Construct Overview

Action plans are detailed plans to perform a distinct subbehaviour of the target behaviour. Action plans direct when, where, and how to carry out this subbehaviour or integrated set of component subbehaviours. If the target behaviour is defined at a very high level of specificity, the action plan may also concern performance of the target behaviour itself.

An action plan is defined as a representation of the plan to perform the plan’s target behavior 1) at a specific time (when; temporal); 2) at a specific location or situation (where; contextual); and 3) in a specific way (how; operational). An incomplete action plan, for example only specifying the contextual or operational aspect, is still considered an action plan, and therefore, the ‘quality’/‘strength’ of action plans is variable.

Representations of aspects of a plan to carry out one or more (sub)behaviors are not action plans: action plans must always represent one or more of the three aspects (contextual, temporal, or operational).

The presence of a greater number of action plans (i.e. an action plan for each of the subbehaviours that together constitute the target behaviour), and greater level of detail of action plans (i.e. specifying exactly when, where, and how to carry out the subbehaviours) increase the likelihood of performance of the target behaviour. The extent to which the various action plans related to the target behaviour are present or absent, is represented by the action plan composite (see dct:action_plancomp_79n2w1bh).

The self-reported presence of action plans can be measured with dichotomous items asking if a specific plan is present (yes/no) for performance of a subbehaviour of the target behaviour. Answers on these items express if the participant has a plan to perform the subbehaviour according to the description provided. Items should specify when, where, and how the particular subbehaviour is performed.

Items measuring the presence of action plans start with a question stem that presents the situation about which the question is asking, followed by a detailed description of an action plan. Participants are asked to indicate if they have (yes=1/no=0) this specific action plan.

An example of a question stem followed by three items measuring the presence of three separate action plans for the target behaviour “protecting skin against UV radiation” is:

When you are going on a day trip this summer and large parts of your skin will be uncovered (e.g. at the pool or beach)…

1. Do you plan to check one day or more beforehand if you have sunscreen at home?
2. Do you plan to check if you have sunscreen with you before leaving on the day trip?
3. Do you plan to use sunscreen every 2 hours?

To assess a participant’s overall presence of action plans for the target behaviour see dct:action_plancomp_79n2w1bh.

Questions or questionnaires that measure the presence of action plans for performance of subbehaviours of the target behaviour. In some instances the target behaviour is defined at a very high level of specificity. In that case, dividing the target behaviour into subbehaviours is not possible or redundant, and action plans for performance of the target behaviour itself are measured.

The presence of action plans, ideally, is measured with dichotomous items asking if a specific plan is present (yes/no) for performance of a subbehaviour of the target behaviour. Answers on these items express if the participant has a plan to perform the subbehaviour according to the description provided. Items should specify when, where, and how the particular subbehaviour is performed.

Items measuring the presence of action plans can start with a question stem that presents the issue and situation about which the question is asking, followed by a detailed description of an action plan. Participants are askes to indicate if they have (yes/no) this specific action plan.

An example of a question stem followed by three items measuring the presence of three separate action plans for the target behaviour “protecting skin against UV radiation” is:

When you are going on a day trip this summer and large parts of your skin will be uncovered (e.g. at the pool or beach)….

1. Do you plan to check one day or more beforehand if you have sunscreen at home?
2. Do you plan to check if you have sunscreen with you before leaving on the day trip?
3. Do you plan to use sunscreen every 2 hours?

Measures of action plans often provide answer options that express the likelihood the participant expects to perform the decribed behaviour, or the level of agreement or diagreement with performance of the behaviour; i.e. with anchors “Very likely” vs “Very unlikely” and “Totally agree” vs “Totally diagree” on a 5- or 7-point Likert scale. For example: “When you are going to a pop concert, music festival, or dance/house event, do you plan to take ear plugs with you?”, with answers ranging from “Very likely” vs “Very unlikely” on a 5-point scale. These items may not measure (the presence of) action plans, but another construct, most likely expectations, and should be coded as such.

See action_plancomp_79n2w1bh for the measurement of a participant’s overall presence of action plans for a particular target behaviour

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, for those participants that have expressed the intention to perform the target behaviour (see: intention_73dnt604), ask for the presence of specific plans how, when, or where to perform the target behaviour.

These questions may be formulated in the following way: “How do you plan to TARGET BEHAVIOUR?”, “When do you plan to TARGET BEHAVIOUR?”, and “Where do you plan to TARGET BEHAVIOUR”, e.g. “How/when/where do you plan to exercise?”.

Expressions that demonstrate or imply the presence or absence of specific plans to perform the target behaviour or subbehaviours of the target behaviour. For example “I have an appointment with a friend to go swimming every Monday and Thursday morning from 09.00 to 10.30” is an expression of an action plan for the target behaviour “engage in physical exercise at least twice a week”.

The overall presence or absence of action plans for the target behaviour. In its totality, the action plan composite indicates the likelihood of performance of the target behaviour.

Action plans (see dct:action_plan_79n2w1bh) are detailed plans to perform a distinct subbehaviour of the target behaviour. Action plans direct when, where, and how to carry out this subbehaviour or integrated set of component subbehaviours. If the target behaviour is defined at a very high level of specificity, the action plan may also concern performance of the target behaviour itself.

The presence of a greater number of action plans (i.e. an action plan for each of the subbehaviours that together constitute the target behaviour), and greater level of detail of action plans (i.e. specifying exactly when, where, and how to carry out the subbehaviours) increase the likelihood of performance of the target behaviour. The extent to which the various action plans related to the target behaviour are present or absent, is represented by the action plan composite.

A composite of action plans is obtained by summing present and absent action plans for the target behaviour. The subsequent approach can thereby be followed:

1. Break downs the target behaviour in subbehaviours.
2. Formulate items measuring the presence of action plans for each of these subbehaviours (see dct:action_plan_79n2w1bh for instructions).
3. Create action plan composite: sum all present (=1) and absent (=0) action plans for the target behaviour

Participants’ overall score expresses the degree action plans for the subbehaviours are present. A high score (e.g. 6 out of 7) makes performance of the target behaviour more likely.

Action plans composites represent the degree action plans are present or absent for the target behaviour. In research, such an overall overview, either graphically or numerically represented, is generally absent.

Therefore, action plans composites have to be deduced from the scores on the different action plan items. One way of doing is, is by creating an overall score by summing the scores of all items measuring action plans (see “measure_dev” for instruction for dichotomous scored items).

Participants’ overall score expresses the degree action plans for the subbehaviours are present. A high score (e.g. 6 out of 7) makes performance of the target behaviour more likely.

Action plan composites can not be directly elicited. See action plans (dct:action_plan_79n2w1bh) for instructions for eliciting their construct content.

Action plan composites can not be directly elicited. See action plans (dct:action_plan_79n2w1bh) for instructions for coding expressions of construct content.

Anticipated regret from engaging in a behaviour.

Use 5-point Likert scales that measure the degree to which participants believe that engagement in a particular behaviour will subsequently lead to regret, e.g. “If you would play video games the whole night, how much regret would you feel?” or “If you would eat a bag of cookies, how much regret would you feel?”, both items with anchors “none” (=0) and “a lot” (=4).

Operationalisations that measure the expectation to feel regret after engagement in the target behavior, e.g. “If you would play video games the whole night, how much regret would you feel?” or “If you would eat a bag of cookies, how much regret would you feel?”, both items with anchors “none” (=0) and “a lot” (=4).

Anticipated regret from not engaging in the target behaviour is inaction regret (see dct:inaction_regret_79n2r0t0).

For the composite of action and inaction regret, either as incentive towards a behaviour (i.e. as sumscore) or the total anticipated regret experienced, see respectively anticip_regret_79n2w1bh and anticip_regret_79n2w1bj.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these questions: ‘What do you expect to feel after (engaging in) TARGET BEHAVIOUR?’ and ‘What emotions do do you expect to experience after (engaging in) TARGET BEHAVIOUR?’

Expressions that demonstrate or imply that one expects to experience feelings of regret if one would engage in the target behaviour, usings verbs such ‘remorse’, ‘sorry’, and ‘lament’.

Manifestations that express feelings of regret if one would not engage in the target behaviour,should be coded dct:inaction_regret_79n2r0t0

Relevant skills and abilities as well as barriers to and facilitators of behavioral performance.

Researchers can utilize three complementary methods to assess the degree of actual control of a participant.

First, they can establish whether control factors, expressed by participants during the elicitation of salient beliefs, constitute actual (real) barriers or facilitators. Fishbein & Ajzen (2010) on p. 330-331 provide an example: ‘Imagine that lack of public transportation is one of the salient factors mentioned as a possible barrier to getting a colonoscopy. We can try to establish if this is a realistic concern by checking with the transportation authorities on the kinds of public transportation available in the area. This would provide a measure of actual control with respect to this particular factor. In addition, we would also need to ascertain whether it is true that lack of public transportation actually interferes with getting to the doctor’s office’.

A second method to assess actual control is to gather background information about the population of interest; specifically concerning factors that may affect their performance of the target behaviour. Identification of these factors, whether personal or environmental, may be enhanced by consulting experts, members of the target population, and those closely living or working together with those members. As a result of increased comprehension of the study population, specific items can be formulated and included in the questionnaire that measure background factors relevant for performance of the target behaviour (e.g. highest level of education achieved), and/or a skills test (see next paragraph) may be developed, and as such these measures may be provide an (indirect) indication of the level of actual control.

Finally, a third method to measure actual control is observation. The current gold standard for objectively measuring enacted behaviour is ‘observational coding’ (i.e. ‘classifying and quantifying verbal and nonverbal behavioural events or psychological states, irrespective of participants’ reports or perceptions’). For example, registering actual control over the intention to be physically active by observing a participant’s choices during the day, e.g taking the stairs instead of the elevator, going by bike, and not by car, etc. As a consequence of its time and resource intensive nature, direct observation is not always a viable option. However, technical innovations have augmented the possiblities to track people’s behaviour, affect, and physiology outside of laboratory settings, e.g. an accelerometer may give an (objective) indication of a person’s ability to remain physically active over a determined time period. Another way of observing actual control is by designing a test in which participants’ skills are put to the proof, e.g. by asking participants to prepare a résumé to test the presence of facilitating factors in applying for a job (example from p. 331 of Fishbein & Ajzen, 2010).

Most research uses perceived behavioural control (see dct:perceivedBehavioralControl_73dnt603) as a proxy of actual control, e.g. questions asessing control beliefs and direct measurements of perceived behavioural control are regarded to reflect actual control.

In some instances, the reality of control beliefs may be objectively verified by the researchers, e.g. a check of the map and neighboorhood may prove or disapprove the belief that no green areas are available for outdoor physical activities. Furthermore, researchers may apply tests to assess the presence of skills essential for the enactment of the target behaviour. Background factors relevant for performance of the target behaviour, such as highest level of education achieved, may also give an indication of the level of actual control.

Conduct a qualitative study where participants are individually interviewed in a free-response format, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these two questions: ‘Please list any factors or circumstances that would make it easy or enable you to target behaviour’ and ‘Please list any factors or circumstances that would make it difficult or prevent you from target behaviour.’

Expressions that demonstrate or imply the presence or absence of confidence about one’s control over performance or nonperformance of the target behaviour (i.e. if (non)performance is “up to them”), and expressions that demonstrate or imply the presence or absence of confidence about one’s ability to perform the target behaviour.

Whether these elicited control beliefs constitute actual (real) barriers or facilitators may be subsequently verified by the researchers, may form the starting point for the development of a skills test or observational coding scheme, or may help gathering the right background information (see: ‘measure_dev’ of this same DCT for more information). These subsequent steps will thus help establish to what extent actual control with regard to the target behaviour is present.

Anticipated regret is the expectation that the decision to perform or not perform a certain behaviour will result in disappointment.

Anticipated regret stems from the the expectation that the decision to perform or not perform the target behaviour will lead to feelings of regret. Therefore, anticipated regret with regard to the target behaviour is assessed via the subconstruct action regret (dct:action_regret_79n2w1bj) and inaction regret (dct:inaction_regret_79n2r0t0).

Whether assessment of action regret or inaction regret, or both, is more appropriate depends on the target behaviour and setting of the study. For example, if the target behaviour is “exercising at least six times over the next two weeks”, then measuring inaction regret appears more logical: “If you would not exercise at least six times in the next two weeks, how much regret would you feel?”, with anchors “none” (=0) and “a lot” (=4). However, one can imagine situations (e.g. exercise as escapism, recovery from injury) in which it makes sense to also measure in the opposite direction (i.e. assess action regret): “If you would exercise at least six times in the next two weeks, how much regret would you feel?”.

Expressed as a numeric score, the level of anticipated regret is equal to the level of action or inaction regret, if only one of the two subconstructs is assessed. When both the level of action and inaction regret are measured (with regard to the same target behaviour), the situation is different. In that case, people that experience anticipated regret tend to have a high score on either action or inaction regret. Therefore, calculating a sum or mean score for the two measures combined does not make sense, as this would make the scores of people with a strong experience of anticipated regret (e.g. 4+0=4 sumscore) equal to people with a moderate experience of anticipated regret (e.g. 2+2=4 sumscore), which would no do justice to their experience. Furthermore, it seems likely that for the former group, the instigation towards action/inaction with respect to the target behaviour is greater, precisely because in this group anticipated regret is experienced stronger. Creating a sum or average score from action and inaction regret scores thus does not make sense, neither to express the level of experienced anticipated regret nor to predict engagement vs non-engagement in the target behaviour.

Substracting one score from the other, or multiplying scores, does not help with the interpretation either. Therefore, no combined score is created from action and inaction regret and scores for both subcontructs are independently assessed and interpreted.

For anticipated regret expressed as a composite score (i.e. the summation of action and inaction regret scores) see dct:anticip_regret_79n2w1bh

Questions or questionnaires that measure the expectation that enagement or non-engagement in the target behaviour will lead to feelings of regret. In most research, it is not specified whether the questions used measure action or inaction regret (the two forms in which anticipated regret may appear), and thus (if of interest) has to be deduced from the formulation of the questions. In most cases either action or inaction regret is used. Which one of the two is used, is mostly guided by the target behaviour.

For example in the study “Deciding to exercise: The role of anticipated regret” (Abraham & Sheeran, 2004), the target behaviour is “exercising at least six times over the next two weeks”, and the two items measuring anticipated regret both assess action regret (although the authors do no mention such a specification): “If I did not exercise at least six times in the next two weeks, I would feel regret” and “‘If I did not exercise at least six times in the next two weeks, I would feel upset’, witch anchors “definitely no” and “definitely yes” for both items.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use one of the following two pairs of questions: ‘What do you expect to feel after engaging in TARGET BEHAVIOUR?’ and ‘What do you expect to feel after not engaging in TARGET BEHAVIOUR?'; or: ‘What emotions do do you expect to experience after engaging in TARGET BEHAVIOUR?’ and ‘What emotions do do you expect to experience after not engaging in TARGET BEHAVIOUR?’

Expressions that demonstrate or imply that one expects to experience feelings of regret if one would engage/not engage in the target behaviour, usings verbs such as ‘remorse’, ‘sorry’, and ‘lament’.

A latent disposition or tendency to respond with some degree of favorableness or unfavorableness to the target behavior.

Attitude consists of two sub-constructs, and is measured by measuring those, i.e.: dct:experientialAttitude_73dnt5z5 and dct:instrumentalAttitude_73dnt5zb

A sample question to measure instrumental and experiental attitude would be (from Ajzen & Fishbein, 2010, on p. 450 of their book “Predicting and changing Behavior: The Reasoned Action Approach”): “My exercising for at least 20 minutes, three times per week, for the next 3 months would be….”. Answer options range from “bad” to “good” (for instrumental attitude), and from “pleasant” to “unpleasant” (for experiental attitude) on a 7-point Likert scale.

Attitude consists of two sub-constructs, and is measured by measuring those: dct:experientialAttitude_73dnt5z5 and dct:instrumentalAttitude_73dnt5zb.

Conduct a qualitative study where participants are individually interviewed in a free-response format, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding. In this qualitative study, use these three questions: ‘What do you see as the advantages of you engaging in TARGET BEHAVIOUR?’, ‘What do you see as the disadvantages of you engaging in TARGET BEHAVIOUR?’ and ‘What else comes to mind when you think about TARGET BEHAVIOUR?’.

Expressions of any global evaluation of the target behaviour as a whole. Note that evaluations of specific aspects or consequences of the behavior relate to aspects of underlying constructs, not of attitude itself.

Also make sure to check the coding instructions for the two sub-constructs that attitude consists of: dct:experientialAttitude_73dnt5z5 and dct:instrumentalAttitude_73dnt5zb.

A latent disposition or tendency to respond with some degree of favorableness or unfavorableness to loud sounds.

Attitude towards noise is a specific subtype of attitude (see dct: attitude_73dnt5zc). It is assessed with instruments that consist of items that measure both the experiential (see dct:experientialAttitude_73dnt5z5) and instrumental (dct:instrumentalAttitude_73dnt5zb) attitude towards noise, as is for example the case for the ‘Youth Attitude to Noise Scale’ (YANS) (Widen et al, 2006), or the instrument may measure attitude towards loud sounds without the items being particularly experiential or instrumental in nature.

When “attitude towards noise” is assessed it is important that this attitude is measured in diverse situations, i.e. the situations must refer to both voluntarily and involuntarily chosen sources of loud noise (e.g. both pop concerts and traffic noise).

Attitude towards noise is a specific subtype of attitude (see dct: attitude_73dnt5zc). It is assessed with instruments that consist of items that measure both the experiential (see dct:experientialAttitude_73dnt5z5) and instrumental (dct:instrumentalAttitude_73dnt5zb) attitude towards noise, as is for example the case for the ‘Youth Attitude to Noise Scale’ (YANS) (Widen et al, 2006), or the instrument may measure attitude towards loud sounds without the items being particularly experiential or instrumental in nature.

When “attitude towards noise” is assessed it is important that this attitude is measured in diverse situations, i.e. the situations must refer to both voluntarily and involuntarily chosen sources of loud noise (e.g. both pop concerts and traffic noise).

Conduct a qualitative study where participants are individually interviewed in a free-response format, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding. In this qualitative study, use these three questions: ‘What do you see as the advantages of LOUD SOUND’, ‘What do you see as the disadvantages of LOUD SOUND?’ and ‘What else comes to mind when you think about LOUD SOUND?’.

To elicit the general attitude towards loud sound, diverse sources of loud sound and situations in which loud sound may be present, should be presented.

Expressions of any global evaluation of loud sound as a whole. Note that evaluations of specific aspects or consequences of the behavior relate to aspects of underlying constructs, not of attitude itself.

Expressions that relate specifically and only to attitudes towards loud sounds in sports, musical, or nightlife settings are coded as dct:attitude_soundculture_7c082lmy

Also make sure to check the coding instructions for the two sub-constructs that attitude consists of: dct:experientialAttitude_73dnt5z5 and dct:instrumentalAttitude_73dnt5zb.

A latent disposition or tendency to respond with some degree of favorableness or unfavorableness to the sound levels at musical and sports activities and events, and in nightlife settings (‘clubbing’).

Attitude towards sound levels is a subgroup of attitude towards noise (see:dct:attitude_noise_7c08258d), which in its turn is a specific subtype of attitude (see dct: attitude_73dnt5zc).

Attitude towards sound levels is assessed with instruments that consist of items that measure both the experiential (see dct:experientialAttitude_73dnt5z5) and instrumental (dct:instrumentalAttitude_73dnt5zb) attitude towards sound levels at musical and sports activities and events, and in nightlife settings (‘clubbing’), or the instrument may measure attitude towards the sound levels at musical and sports activities and events, and in nightlife settings, without the items being particularly experiential or instrumental in nature.

The first factor (phrased “elements of youth culture”) of the ‘Youth Attitude to Noise Scale’ (YANS) (Widen et al, 2006), consisting of 8 items, is an example of an instrument assessing this construct. It consists of items measuring both experiential and instrumental attitudes towards sound levels in musical, sports and nightlife settings.

Attitude towards sound levels is a subgroup of attitude towards noise (see dct:attitude_noise_7c08258d), which in its turn is a specific subtype of attitude (see dct: attitude_73dnt5zc).

Attitude towards sound levels is assessed with instruments that consist of items that measure both the experiential (see dct:experientialAttitude_73dnt5z5) and instrumental (dct:instrumentalAttitude_73dnt5zb) attitude towards sound levels at musical and sports activities and events, and in nightlife settings (‘clubbing’), or the instrument may measure attitude towards the sound levels at musical and sports activities and events, and in nightlife settings, without the items being particularly experiential or instrumental in nature.

The first factor (phrased “elements of youth culture”), consisting of the 8 items, of the ‘Youth Attitude to Noise Scale’ (YANS) (Widen et al, 2006), is an example of an instrument assessing this construct. It consists of items measuring both experiential and instrumental attitudes towards sound levels in musical, sports and nightlife settings.

Conduct a qualitative study where participants are individually interviewed in a free-response format, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding. In this qualitative study, use these three questions: ‘What do you see as the advantages of LOUD SOUND during sport/music/nightlife events’, ‘What do you see as the disadvantages of LOUD SOUND during sport/music/nightlife events?’ and ‘What else comes to mind when you think about LOUD SOUND during sport/music/nightlife events?’.

Expressions of any global evaluation of loud sounds in sport/music/nightlife settings. Note that evaluations of specific aspects or consequences of the behavior relate to aspects of underlying constructs, not of attitude itself.

Expressions that relate to general attitudes towards loud sound (i.e. not specifically refering to sports, musical, or nightlife settings) are coded as dct:attitude_noise_7c08258d

Also make sure to check the coding instructions for the two sub-constructs that attitude consists of: dct:experientialAttitude_73dnt5z5 and dct:instrumentalAttitude_73dnt5zb.

People’s perceptions of the extent to which performance or nonperformance of a behaviour is up to them.

Items to measure autonomy usually ask participants to consider the extent to which extent performance or nonperformance of the behaviour is under their control or is up to them. The focus here is on control over performance as well as nonperformance of the behaviour.

Fishbein & Ajzen provide a sample question to measure autonomy on p. 450 of their Reasoned Action Approach (RAA) book (2010): “My exercising for at least 20 minutes, three times per week, for the next 3 months is up to me”. Answer options range from “Disagree” to “Agree” on a 7-point Likert scale.

Other examples of items that measure autonomy are (p. 166 of RAA book): “How much control do you have over whether you perform behavior x?” (no control–complete control); “I feel in complete control over whether I perform behavior x” (completely false–completely true); “Whether or not I perform behavior x is completely up to me” (disagree–agree); “How much do you feel that performing behavior x is beyond your control?” (not at all–very much); “It is mostly up to me whether or not I perform behavior x” (strongly agree–strongly disagree); “The number of events outside my control which could prevent me from performing behavior x are …” (numerous–very few).

Questions or questionnaires that measure the perceived degree to which performance or nonperformance of the behaviour is under participants’ control or is up to them. Examples of questions measuring autonomy are: “I feel in complete control over whether I perform TARGET BEHAVIOUR” (completely false–completely true); and “Whether or not I perform TARGET BEHAVIOUR is completely up to me”, with responses ranging from “Disagree” to “Agree” on a 7-point Likert scale.

Conduct a qualitative study where participants are individually interviewed in a free-response format, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these two questions: ‘Please list any factors or circumstances that would make it easy or enable you to TARGET BEHAVIOUR’ and ‘Please list any factors or circumstances that would make it difficult or prevent you from TARGET BEHAVIOUR.’

Expressions that demonstrate or imply the presence or absence of confidence about one’s control over performance or nonperformance of the target behaviour (i.e. if (non)performance is “up to them”), for example because of a barrier, obstacle, or facilitating condition or circumstance.

Expressions that relate to one’s ability to perform the target behavior, should be coded as dct:capacity_73dnt602.

Expressions that refer to control in general (not specifically one’s autonomy), should be coded as dct:perceivedBehavioralControl_71w8sfdk.

The sum of the autonomy beliefs. In their totality, autonomy beliefs lead to the perception that performance or nonperformance of the target behaviour is under their control or is up to them.

A composite of autonomy beliefs is obtained by multiplying the strength of each autonomy belief (see dct:autonomy_conditionPresence_73dnt5zr) by its perceived power (see dct:autonomy_conditionPower_73dnt5zs) and then summing the product terms across all autonomy beliefs.

This, preferably, is represented graphically (e.g. in a table or index) and, additionally, in an equation. See Table 5.5 on p.171 of the Reasoned Action Approach (RAA) book (Ajzen & Fishbein, 2010) for an example.

The autonomy beliefs composite reflects the summation of salient autonomy beliefs with regard to the target behaviour. Each autonomy belief (dct:autonomy_belief_73dnt5zt), in its turn, is the product term of the multiplication of the strength of the belief (dct:autonomy_conditionPresence_73dnt5zr) by its perceived power (dct:autonomy_conditionPower_73dnt5zs).

Given the formative nature of the autonomy beliefs composite, for a swift and clear overview of its composition and its effect on autonomy (see: autonomy_73dnt5zx), preferably a graphical representation (e.g. a table or index) and, possibly, an equation is consulted. Table 5.5 on p.171 of Reasoned Action Approach book (Ajzen & Fishbein, 2010) represents such an overview.

Autonomy beliefs and, as a consequence, also autonomy beliefs composites (which are the sum of the former), can not be directly elicited. Autonomy beliefs are the product of the multiplication of a belief’s strength (dct:autonomy_conditionPresence_73dnt5zr) by its perceived power (dct:autonomy_conditionPower_73dnt5zs).

See these two subconstructs for instructions for eliciting their respective construct content.

Autonomy beliefs and, as a consequence, also autonomy beliefs composites (which are the sum of the former), can not be directly elicited. Autonomy beliefs are the product of the multiplication of a belief’s strength (dct:autonomy_conditionPresence_73dnt5zr) by its perceived power (dct:autonomy_conditionPower_73dnt5zs).

See these two subconstructs for instructions for coding expressions of construct content.

A person’s belief about the extent to which performance or nonperformance of a particular behaviour is up to him/her.

The measurement of autonomy beliefs starts with an elicitation study to identify the content of autonomy beliefs about the behaviour which are shared by the target population.

To obtain quantitative data regarding autonomy beliefs, the participant is subsequently asked to respond to two questions with respect to each autonomy belief, one to assess autonomy belief strength (see dct:autonomy_conditionPresence_73dnt5zr) and the other to measure the power of the autonomy belief to facilitate or impede behavioural performance (see dct:autonomy_conditionPower_73dnt5zs).

Finally, to get an overall quantitative score of an autonomy belief, the strength of a belief is multiplied by its perceived power. The total score of an autonomy belief enables the assessment of the importance of the contribution of an autonomy belief, in comparison with other autonomy beliefs, to the autonomy beliefs composite (see dct:autonomy_belCom_73dnt5zw ).

Each autonomy belief is the product term of the multiplication of the strength of the belief (see dct:autonomy_conditionPresence_73dnt5zr) by its perceived power (see dct:autonomy_conditionPower_73dnt5zs). Therefore, whether an autonomy belief exerts influence (and if so, to what extent) on the autonomy beliefs composite (see dct:autonomy_belCom_73dnt5zw ), and whether a particular autonomy belief lowers or increases the sense of perceived autonomy, can only be deduced by concurrent consultation of the two subcomponents of which it is constructed. Table 5.5 on p. 171 of the Reasoned Action Approach (RAA) book (Ajzen & Fishbein, 2010) illustrates how the two subcomponents can exert an influence on individual autonomy beliefs.

Autonomy beliefs can not be directly elicited. They are the product of the multiplication of the strength of the autonomy belief (see dct:autonomy_conditionPresence_73dnt5zr) by its perceived power (see dct:autonomy_conditionPower_73dnt5zs). See these two subconstructs for instructions for eliciting their respective construct content.

Autonomy beliefs can not be directly elicited. They are the product of the multiplication of the strength of the autonomy belief (see dct:autonomy_conditionPresence_73dnt5zr) by its perceived power (see dct:autonomy_conditionPower_73dnt5zs). See these two subconstructs for instructions for coding expressions of construct content.

The perceived power of a condition refers to the perceived impact the presence of a condition has on one’s autonomy. This impact can be perceived to be anywhere from very negative (if the condition is a barrier), through zero if the condition is perceived to have no impact on one’s autonomy regardless of whether it is perceived to be present, to very positive (if the condition is a facilitator). Note that the perception as to whether the condition is present or not is its perceived presence (dct:autonomy_conditionPresence_73dnt5zr).

According to the Reasoned Action Approach, perceptions about the power of conditions combine multiplicatively with perceptions about the presence of conditions (see dct:autonomy_conditionPresence_73dnt5zr) into autonomy beliefs (see dct:autonomy_belief_73dnt5zt).

Examples of the perceived power of a condition are the perception that having to ask for condoms at the counter obstructs buying them; the perception that rain obstructs going for a run; the perception that if food is expensive, that is a barrier to buying it; the perception that paying for STI tests is a barrier to get tested; and the perception that if anticonception is not widely available, it cannot be obtained and used.

The perceived power of a condition is measured by asking the extent to which a condition facilitates or impedes performance of a behavior. As the same condition can be facilitating for some people and impeding for others, this also implies that use of bidimensional scales is required. However, if one of the two dimensions can be excluded a priori, a unidimensional scale suffices.

For bidimensional scales, the items are formulated as ‘If I want to [TARGET BEHAVIOR], (whether) [CONDITION] …’ with anchors ‘Much harder’ and ‘Much easier’. For example, ‘If I want to go for a run, rainy weather makes it [Much harder|Much easier].’

For unidimensional scales, which are only used if one of the dimensions can be excluded a priori, the way the items are formulated depends on whether they concern a barrier or a facilitator. For facilitators, use ‘If I want to [TARGET BEHAVIOR], whether [CONDITION]…’ with anchors ‘Doesn’t matter’ and ‘Makes it much easier’, and for barriers, ‘If I want to [TARGET BEHAVIOR], whether [CONDITION]…’ with anchors ‘Doesn’t matter’ and ‘Makes it much harder’.

For example, ‘If I want to use anticonception, whether it is widely available…’ with anchors ‘Doesn’t matter’ to ‘Makes it much easier’.

The perceived presence of a condition refers to the perceived probability that a condition will be present that may have some power to facilitate or obstruct successfully accomplishing the target behavior. Such conditions are perceived to be external to the individuals direct control (though they may in reality be internal conditions; for example, addicts may perceive their craving to be irresistible, and as such, a condition they have no control over and that negatively impacts their autonomy). Facilitating or obstructing factors that are perceived to be internal (i.e. that an individual can learn over time) are subskills (see dct:capacity_subskillPresence_73dnt5zy).

Each perceived presence of a condition concerns the absence or presence of one condition. The perceived nature of the condition (i.e. whether it facilitates or obstructs the target behavior), i.e. the perceived power of that condition, is covered in dct:autonomy_conditionPower_73dnt5zs.

According to the Reasoned Action Approach, perceptions about the presence of conditions combine multiplicatively with perceptions about the power of conditions (see dct:autonomy_conditionPower_73dnt5zs) into autonomy beliefs (see dct:autonomy_belief_73dnt5zt).

Examples of perceived presence of a condition are the perception that to buy condoms, one will have to ask for them at the counter (which may obstruct buying condoms); the perception that it is raining (which may obstruct going for a run); the perception that healthy food is very expensive (which may obstruct buying healthy food); the perception that testing for STIs is free (which may facilitate getting tested for STIs); and the perception that anticonception is widely available (which may facilitate using anticonception).

The perceived presence of a condition is measured by asking the perceived probability that a condition will be present. A unidimensional scale is required. The items are formulated as ‘How likely do you think it is that [CONDITION]?’ with anchors ‘Very unlikely’ and ‘Very likely’.

For example, ‘How likely do you think it is that it will rain?’ and ‘How likely do you think it is that anticonception will be widely available?’, both with anchors ‘Very unlikely’ and ‘Very likely’.

People’s perceptions of the consistency or inconsistency between their perceived behaviour and a given guideline.

Awareness, in the present definition, is the agreement between self-rated behaviour and a given behavioural guideline. The construct can be assessed in a three-step procedure:

1. Firstly, participants are asked if their behaviour is in line with with the guidelines for that particular behaviour (yes/no). If the guideline is not known, in an introduction to the question this information is provided.

2. Secondly, participants’ actual engagement in the target behaviour is assessed using objective measures (e.g. accelerometer and heart rate monitor to measure actual physical activity, caloric calculations to measure food intake).

3. Thirdly, the accuracy of participants’ perception of their behaviour meeting the guidelines can be assessed by comparing the outcomes of step 1 and 2. As a result of this comparison, four awareness groups can be created: overestimators, underestimators, realistic (not meeting guideline), and realistic (meeting guideline). Thus, the latter two groups are aware of their behaviour, while the first two groups are not.

See “Awareness of physical activity in healthy middle-aged adults: a cross-sectional study of associations with sociodemographic, biological, behavioural, and psychological factors” (Godino et al., 2014) for a sample application of this method.

People’s perception of their behaviour meeting a guideline (yes/no) compared with an objective assessment of this particular behaviour can be employed as a measure of awareness.

People’s perception of their behaviour meeting a guideline can be asssessed with a single question, with yes/no response options.

Objective measures of behaviour are observation and/or the tracking/monitoring of behaviour with mobile apps and technical devices such as accelerometers. If no objective assessment of behaviour can be made, participants can be asked a series of very specific questions concering the target behaviour or asked to keep a diary. These outcomes are subsequently compared with participants’ perception of their behaviour meeting the guideline.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, introduce a specific guideline (for example for physical activity, the number of calories needed a day, or the maximum safe exposure to loud noise in decibels/minutes a day), followed by the following consecutive questions “Do you meet this guideline?” and “Why do you think you meet/do not meet the guideline?”

Expressions that demonstrate or imply the ability to correctly evaluate one’s behaviour in light of the guideline (i.e. “awareness”). For example: “I don’t do sports, I do not do any physical work, I travel by car, so I am pretty sure I don’t get the recommended level of activity” or “I weigh all my food and use an app to calculate the number of calories, so I know my food intake is in line with the energetic recommendations for my age, bodyweight, and activity level”.

The target behaviour is the behaviour to explain using determinants. A behaviour is defined as individual human activity that involves co-ordinated contraction of striated muscles controlled by the brain. Note that this also includes the lack of such co-ordinated contraction.

Depending on the nature of the target behavior, engagement in that behavior can be conceived as binary (i.e. one either does or does not engage in the target behavior, e.g. getting tested for STIs every six months), a matter of frequency (i.e. one engages in the target behavior with a frequency from zero up to a given feasible maximum frequency in a given timespan, e.g. the frequency with which one washes their hands conform the guidelines), a matter of intensity (i.e. one engages in the target behavior to a degree from zero up to a given feasible maximum intensity, e.g. the amount of kilocalories one consumed in a meal), or a combination of these (i.e. one engages in the target behavior with a given frequency and with a given intensity, e.g. how frequently one drinks alcohol, and how many grams of alcohol one consumes when one does).

Ideally, the target behaviour is described unequivocally and consistent in the questions and instructions for participants, and in publications resulting from the research. Moreover, the target behaviour should be described as specific as possible, i.e. specifying action, target, context, and time - and address the level of generality of the behaviour. Diverging and/or ambiguous formulations of the target behaviour confound and limit the applicability of study results.

For example ‘exercising for at least 20 minutes, three times per week, for the next 3 months’ could be a viable target behaviour. However, the definition of ‘exercising’ should specified, as participants be may have different views of what comprises ‘exercising’. When a target behaviour is not sufficiently defined, or when various definitions are used, this has to be taken into account when assessing the quality of the study.

Conduct a qualitative study where participants are individually interviewed in a free-response format, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding. In this qualitative study, ask participants for their performance of the target behaviour. The target behaviour should be described unequivocally and consistent, addressing action, target, context, time, and level of generality. As an introduction to the question a definition of the target behaviour can be provided.

An example can be: ‘Our definition of “exercise” is “all physical activity strenuous enough to make you sweat”. Can you tell me if you currently exercise for at least 20 minutes, three times per week?’. An alternative question could be: ‘In the past 3 months, how often have you exercised for at least 20 minutes, three times per week?’. Note that this latter question measures past behaviour. In practice, asking about current behaviour most often concerns a retrospective asessment of behaviour until the moment of assessment.

Expressions that demonstrate or imply engagement in the target behaviour. The more specific the definition of the target behaviour, the greater the possibility that people partially engage in the target behaviour; i.e. they do not fulfill all behavioural criteria, e.g. when a participant exercises for at least 20 minutes, but not three, but two times per week. How these partial engagements in the target behaviour are to be coded, should be decided a priori.

An individual’s manifest, observable activities in response to external or internal stimuli.

The prefered instruments to measure behaviour are observation by trained observers using a valid and reliable reporting protocol and/or the use of technical measures and devices to record and track behaviour during particular events and over time.

If these objective measures can not be used, self-report of behaviour assessed by questionniares or interviews are an alternative option. However, one should take into account that these latter instruments by definition measure a subjective representation of behaviour. Moreover, whereas there is general agreement about the definition of a specific behaviour (“running”), this is not the case when asking about a broad behavioural category, such as “exercising”. To assess a category of behaviours via self-report, therefore, we must provide a clear definition of the category in question and specify action, target, context, and time.

Depending on the nature of the behaviour, engagement in the behaviour can be conceived as binary (i.e. one either does or does not engage in the target behaviour, e.g. getting tested for STIs every six months), a matter of frequency (i.e. one engages in the target behaviour with a frequency from zero up to a given feasible maximum frequency in a given timespan, e.g. the frequency with which one washes their hands conform the guidelines), a matter of intensity (i.e. one engages in the target behaviour to a degree from zero up to a given feasible maximum intensity, e.g. the amount of kilocalories one consumed in a meal), or a combination of these (i.e. one engages in the target behaviour with a given frequency and with a given intensity, e.g. how frequently one drinks alcohol, and how many grams of alcohol one consumes when one does).

The instructions for developing a self-report instrument to measure behaviour, logically derive from the foregoing consideration regarding the nature of the behaviour of interest. For example, for a dichotomous response option (yes/no) the question can be formulated in the following way: “Did you get tested for STIs during the last six months?”. If the behaviour is a matter of frequency, multiple response options may be given, e.g.: “During the last seven days, how many days did you drink two or more alcoholic drinks?”, with response options ranging from 0 to 7.

Instruments that measure observable behaviour. The prefered instruments for this purpose are observation by trained observers using a valid and reliable reporting protocol and/or the use of technical measures and devices to record and track behaviour during particular events and over time. Self-report of behaviour assessed by questionniares or interviews are alternative options. However, one should take into account that these latter methods by definition measure a subjective representation of behaviour. Moreover, whereas there is general agreement about the definition of a specific behaviour (“running”), this is not the case when asking about a broad behavioural category, such as “exercising”. To assess a category of behaviours via self-report, therefore, a clear definition of the category in question must be provided, specifying action, target, context, and time.

Depending on the nature of the behaviour, engagement in the behaviour can be conceived as binary (i.e. one either does or does not engage in the target behaviour, e.g. getting tested for STIs every six months), a matter of frequency (i.e. one engages in the target behaviour with a frequency from zero up to a given feasible maximum frequency in a given timespan, e.g. the frequency with which one washes their hands conform the guidelines), a matter of intensity (i.e. one engages in the target behaviour to a degree from zero up to a given feasible maximum intensity, e.g. the amount of kilocalories one consumed in a meal), or a combination of these (i.e. one engages in the target behaviour with a given frequency and with a given intensity, e.g. how frequently one drinks alcohol, and how many grams of alcohol one consumes when one does).

The composition of self-report instruments to measure behaviour, logically derive from the foregoing consideration regarding the nature of the behaviour of interest. For example, for a dichotomous response option (yes/no) the question can be formulated in the following way: “Did you get tested for STIs during the last six months?”. If the behaviour is a matter of frequency, multiple response options may be given, e.g.: “During the last seven days, how many days did you drink two or more alcoholic drinks?”, with response options ranging from 0 to 7.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, the formulation of the question(s) depends on the nature of the behaviour, i.e. engagement in the behaviour can be conceived as binary (i.e. one either does or does not engage in the target behaviour, e.g. getting tested for STIs every six months), a matter of frequency (i.e. one engages in the target behaviour with a frequency from zero up to a given feasible maximum frequency in a given timespan, e.g. the frequency with which one washes their hands conform the guidelines), a matter of intensity (i.e. one engages in the target behaviour to a degree from zero up to a given feasible maximum intensity, e.g. the amount of kilocalories one consumed in a meal), or a combination of these (i.e. one engages in the target behaviour with a given frequency and with a given intensity, e.g. how frequently one drinks alcohol, and how many grams of alcohol one consumes when one does).

If engagement in the behaviour is conceived as binary, the interviewer can ask a question in the following fashion: “Do you you perform BEHAVIOUR X?”, “Do you you perform BEHAVIOUR X when at location X?”, or “Did you you perform BEHAVIOUR X during period X?”.

If the interviewer is interested in the frequency the behaviour is performed, he/she can ask a question in the following fashion: “During period X, how many days did you perform BEHAVIOUR X?” or “When you were at location X, how many times did you perform BEHAVIOUR X?”.

If the intensity of performance of the behaviour is the focus of interest, the interviewer can ask a question in the following fashion: “On the days you performed BEHAVIOUR X during period X, how much/many did you take/consume?” or “On the days you performed BEHAVIOUR X during period X, how long did you engage in BEHAVIOUR X?”

If both the frequency and intensity of engagement in the behaviour is of interest, the interviewer can formulate questions in the following way: “During period X, how many days did you perform BEHAVIOUR X and for how long?” or “During period X, how many days did you perform BEHAVIOUR X and how much/many did you take/consume each time?”.

Expressions that demonstrate or imply engagement in the behaviour of interest. These expressions may indicate engagement in the behaviour (yes/no), the frequency the behaviour is performed, the intensity with which the behaviour is performed, or a combination of the foregoing aspects. For example: “I do not drink alcohol”, “I only drink alcohol in the weekends”, “When going clubbing I drink about 5 tot 6 beers”, or “I drink one glass of wine each day with dinner”.

Expressions that refer to either intentions, willingness, or specific plans to engage in the behaviour should not be coded as dct:behaviour_79n2w1bj, but as, respectively, dct:intention_73dnt604, dct:willingness_79n2fh4w, and dct:action_plan_79n2w1bh

Organized sets of information stored schematically linking the action with cues and contexts built up through repeated past experience.

Behavioural scripts describe the sequencing of events involving observable, recurrent activities and patterns of interaction characteristic of a particular setting. As can be deduced from this definition, behavioral script can be observed. This measurement method is, however, very time intensive and unpractical.

The presence of behavioural scripts can also be assessed by self-report, using interviews and questionnaires as research methods. However, no specific questionnaires exist to measure the presence of behavioural scripts. Therefore, one could use a questionnaire to measure habitual behaviour, such as the Self-Report Habit Index (SRHI) (see dct:habit_79n2r0sz), and possibly apply some modifications to this measure, or develop a new questionnaire from scratch.

A questionnaire measuring behavioural scripts for a particular behaviour can present different scenarios (each representing a particular behavioural script) and ask whether the participant performs the target behaviour in this fashion, with “Very likely” and “Very unlikely” at the extreme ends of a 5- or 7-point Likert scale. For example “When I am at a club or festival I usually get offered a couple of alcoholic drinks from friends, even when I don’t aks for them”, “When I am at a club or festival and a stranger bumps into me, I don’t pay attention to it or start an argument, and I continue my way” and “When I am at a club or festival I first walk around at the event to see who is there.”

Questions or questionnaires that measure the degree to which performance of the target behaviour follows certain behavioural scripts. These fixed behavioural patters may be prompted by certain cues or certain situations.

Questions measuring the presence of behavioural scripts for a particular behaviour can present different scenarios (each representing a particualr behavioural script) and ask whether the participant performs the target behaviour in this fashion, with “Very likely” and “Very unlikely” at the extreme ends of a 5- or 7-point Likert scale. For example “When I am at a club or festival, I usually get offered a couple of alcoholic drink from friends, even when I don’t aks for them”, “When I am at a club or festival and a stranger bumps into me, I don’t pay attention to it or start an argument, and I continue my way” and “When I am at a club or festival, I first walk around at the event to see who is there.”

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding. In this qualitative study, use questions similar to the following examples: “Please describe how your engagement in the TARGET BEHAVIOUR looks like when at location X” or “Please describe the things you usually do before, during, and after you engage in TARGET BEHAVIOUR?”. For example, “Please describe the things you usually do before, during, and after you use cannabis?”.

Expressions that demonstrate or imply the presence of behavioural scripts, using expressions suchs as: “When at X I always do X”, “Without thinking about it, I first do X, then I do X, and then I do X”, “Every day I follow the same sequence of actions….”. More specifically, for example: “When I drink my first cup of coffee, I automatically light a cigarette. After I have finished my coffee and cigarette, I go to the toilet”

Behavioural scripts are related to dct:habit_79n2r0sz and the constructs may overlap.

People’s perceptions of their ability to perform a behavior; that is, the degree to which they believe they can, are able to, or are capable of performing the behavior.

Use Likert scales that measure the degree to which participants believe the target behavior to be something they are capable of performing successfully. The items suggested in the book are: ‘I am confident that if I want to, I can TARGET BEHAVIOR.’ with anchors ‘No confidence at all’ vs ‘A lot of confidence’ and ‘If I really wanted to, I could TARGET BEHAVIOR.’ with anchors ‘Unlikely’ vs ‘Likely’.

Note that questions about how easy or difficult participants perceive a target behavior to be fall outside of the constructs of Perceived Behavioral Control, Autonomy, and Capacity as defined in the 2010 RAA book.

Questions or questionnaires that measure the perceived degree to which target behavior or [contrast behavior] is something [target population] are confident they can successfully perform. If perceived control over the behavior is also measured, code this as dct:perceivedBehavioralControl_71w8sfdk. For example, the items suggested in the book are: ‘I am confident that if I want to, I can TARGET BEHAVIOR.’ with anchors ‘No confidence at all’ vs ‘A lot of confidence’ and ‘If I really wanted to, I could TARGET BEHAVIOR.’ with anchors ‘Unlikely’ vs ‘Likely’.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these two questions. ‘Please list any factors or circumstances that would make it easy or enable you to target behavior.’ and ‘Please list any factors or circumstances that would make it difficult or prevent you from target behavior.’

Expressions that demonstrate or imply the presence or absence of confidence about one’s ability to perform the target behavior. Expressions that refer to explicit control over the target behavior (or lack thereof), for example because of a barrier, obstacle, or facilitating condition or circumstance, it should be coded as perceivedBehavioralControl_autonomy_73bg9sq6. Expressions that refer to confidence in general (not in one’s ability), should be coded as perceivedBehavioralControl_71w8sfdk.

The sum of the capacity beliefs. In their totality, capacity beliefs lead to the perception that one has or does not have the ability to carry out the target behaviour.

A composite of capacity beliefs is obtained by multiplying the perceived presence of each relevant capacity (i.e. ‘subskill’) (see dct:capacity_subskillPresence_73dnt5zy) by its perceived importance (see dct:capacity_subskillImportance_73dnt5zz) and then summing the product terms across all capacity beliefs.

This, preferably, is represented graphically (e.g. in a table or index) and, additionally, in an equation. See Table 5.5 on p.171 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) for an example.

The capacity beliefs composite reflects the summation of salient capacity beliefs with regard to the target behaviour. Each capacity belief (dct:capacity_belief_73dnt600), in its turn, is the product term of the multiplication of the perceived presence of the capacity (i.e. ‘subskill’) (dct:capacity_subskillPresence_73dnt5zy) by the perceived importance of this subskill (dct:capacity_subskillImportance_73dnt5zz).

Given the formative nature of the capacity beliefs composite, for a swift and clear overview of its composition and its effect on capacity (dct:capacity_73dnt602), preferably a graphical representation (e.g. a table or index) and, possibly, an equation is consulted. Table 5.5 on p.171 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) represents such an overview.

Capacity beliefs and, as a consequence, also capacity beliefs composites (which are the sum of the former), can not be directly elicited. Capacity beliefs are are the product of the multiplication of perceived subskill presence (dct:capacity_subskillPresence_73dnt5zy) by the perceived importance of this subskill (dct:capacity_subskillImportance_73dnt5zz).

See these two subconstructs for instructions for eliciting their respective construct content.

Capacity beliefs and, as a consequence, also capacity beliefs composites (which are the sum of the former), can not be directly elicited. Capacity beliefs are are the product of the multiplication of perceived subskill presence (dct:capacity_subskillPresence_73dnt5zy) by the perceived importance of this subskill (dct:capacity_subskillImportance_73dnt5zz).

See these two subconstructs for instructions for coding their respective construct content.

A person’s belief about the degree to which he/she can, is able to, or is capable of performing a particular behavior.

The measurement of capacity beliefs starts with an elicitation study to identify the content of capacity beliefs about the behaviour which are shared by the target population.

To obtain quantitative data regarding capacity beliefs, the participant is subsequently asked to respond to two questions with respect to each salient capacity belief, one question to assess the perceived likelihood that the capacity will be present (see dct:capacity_subskillPresence_73dnt5zy), and the other to measure the perceived importance of the capacity to facilitate or impede behavioral performance (see dct:capacity_subskillImportance_73dnt5zz).

Finally, to get a total quantitative score of a capacity belief, the perceived presence of a capacity is multiplied by its perceived importance. The total score of a capacity belief enables the assessment of the importance of the contribution of a capacity belief, in comparison with other capacity beliefs, to the capacity beliefs composite (see dct:capacity_belCom_73dnt601).

Each capacity belief is the product term of the multiplication of the perceived presence of a capacity (i.e. ‘subskill’) (see dct:capacity_subskillPresence_73dnt5zy) by its perceived importance (see dct:capacity_subskillImportance_73dnt5zz).

Therefore, whether an capacity belief exerts influence (and if so, to what extent) on the capacity beliefs composite (see dct:capacity_belCom_73dnt601), and whether a particular capacity belief lowers or increases the sense of perceived capacity, can only be deduced by concurrent consultation of the two subcomponents of which it is constructed. Table 5.5 on p. 171 of the Reasoned Action Approach (RAA) book (Ajzen & Fishbein, 2010) provides an example how the subcomponents exert an influence on individual control beliefs (in Table 5.5 it concerns autonomy beliefs).

Capacity beliefs can not be directly elicited. They are the product of the multiplication of perceived subskill presence (dct:capacity_subskillPresence_73dnt5zy) by the perceived importance of this subskill (dct:capacity_subskillImportance_73dnt5zz). See these two subconstructs for instructions for eliciting their respective construct content.

Capacity beliefs can not be directly elicited. They are the product of the multiplication of perceived subskill presence (dct:capacity_subskillPresence_73dnt5zy) by the perceived importance of this subskill (dct:capacity_subskillImportance_73dnt5zz). See these two subconstructs for instructions for coding their respective construct content.

The perceived importance of a subskill refers to how important possessing the subskill is to successfully accomplishing the target behavior. A subskill can be perceived to anything from completely irrelevant to crucial. Note that the perception of whether one possesses a subskills is its perceived presence (dct:capacity_subskillPresence_73dnt5zy).

According to the Reasoned Action Approach, perceptions about the importance of subskills combine multiplicatively with perceptions about the presence of subskills (see dct:capacity_subskillPresence_73dnt5zy) into capacity beliefs (see dct:capacity_belief_73dnt600).

Examples of perceived subskill importance are the perception that to successfully buy condoms, it is vital one possesses the skill to ask for condoms at the counter; the perception that to successfully moderate one’s alcohol intake, it doesn’t really matter whether one possesses the skill to decline an alcoholic drink when it is offered; the perception that to successfully run 5 kilometers, it is a prerequisite that one is able to run 5 kilometers; the perception that to protect one’s hearing from loud sounds, it helps if one is able to insert earplugs into one’s ear canal properly; and the perception that to prevent skin cancer in one’s child, it doesn’t matter whether one knows how to apply sunscreen on a child.

The perceived importance of a subskill is measured by asking the extent to which possessing the subskill is relevant to performance of the behavior. The items are formulated as ‘For me to successfully [BEHAVIOR], being able to [SUBSKILL] is…’ with ‘Not at all important’ and ‘extremely important’ as anchors.

For example, ‘For me to successfully use condoms, being able to ask for condoms at the counter is [not at all important|extremely important],’ ‘For me to successfully moderate my alcohol intake, being able to decline an alcoholic drink when it is offered is [not at all important|extremely important].’

The perceived presence of a subskill refers to the perceived probability that a person has a certain skill that is required to successfully accomplishing the target behavior (hence ‘subskill’). Such subskills are perceived to be under the individual’s direct control (though they may in reality not be; for example, it is possible that it is not realistic for somebody to learn how to cook five star meals, yet they might believe that they can acquire the necessary subskills). Facilitating or obstructing factors that are perceived to be external to the individual’s control (i.e. that impact an individual’s autonomy) are perceived conditions (see dct:autonomy_conditionPresence_73dnt5zr).

Each perceived presence of a subskill concerns the absence or presence of one subskill. The perceived importance of the subskill to successfully accomplishing the target behavior (i.e. whether the subskill is crucial or merely facilitates the target behavior), i.e. the perceived subskill importance, is covered in dct:capacity_subskillImportance_73dnt5zz.

According to the Reasoned Action Approach, perceptions about the presence of subskills combine multiplicatively with perceptions about the importance subskills (see dct:capacity_subskillImportance_73dnt5zz) into capacity beliefs (see dct:capacity_belief_73dnt600).

Examples of perceived subskill presence are the perception that one possesses the skill to ask for condoms at the counter; the perception that one possesses the skill to decline an alcoholic drink when it is offered; the perception that one is able to run 5 kilometers; the perception that one can use earplugs properly; and the perception that one knows how to one can apply sunscreen on a child.

The perceived presence of a subskill is measured by asking the perceived probability that a person has a certain skill that is required to successfully accomplish the target behavior. It is important that the item stem explicitly describes a scenario where not successfully accomplishing the relevant sub-behavior (that corresponds to the relevant subskill) can only be the consequence of insufficiently possessing that subskill.

In the item stem, start with a word that establishes that the question concerns a regular situation. Then, if the subskill pertains to a specific scenario, insert that condition. Then continue with ‘are you able to [TARGET BEHAVIOR]’. For subskills that do not pertain to specific scenarios, the item stem then becomes ‘Typically, are you able to [TARGET BEHAVIOR]’, and for subskills that do pertain to a specific scenario, the item stem becomes ‘Typically, if [SCENARIO], are you able to [TARGET BEHAVIOR]’.

As anchors, use ‘Absolutely unable’ to ‘Absolutely able’ (this is a unidimensional scale).

For example, ‘Typically, are you able to ask for condoms at the counter?’ and ‘Typically, if an alcoholic drink is offered, are you able to decline?’, both with anchors ‘Absolutely unable’ and ‘Absolutely able’.

Conscious and volitional efforts to regulate emotion, cognition, behaviour, physiology, and the environment in response to stressful events or circumstances.

Skinner (2007) describes the state of assessment in the field of coping as “chaotic and confusing”, and in a review (Skinner et al., 2003) identified over 100 assessments, tapping over 400 ways of coping.

Among the various measures, self-report instruments exist to assess coping styles among members of various groups (adults, children, adolescents, elderly), and scales may be broadly applicable, while others are ment for situation-specific application (e.g. health-related issues). The COPE (Carver, Scheier, & Weintraub, 1989) is the most frequently used instrument for broad application, and also one of the most commonly used measures in situation-specific contexts (Kato, 2015).

The 60-item COPE consists of 15 subscales (e.g. “active coping” and “acceptance”) and its theoretical development is based on the models of behavioral self-regulation (Carver & Scheier, 1981) and stress and coping (Lazarus & Folkman, 1984). The COPE assumes that individual coping strategies are relatively stable across different types of stressors.

Individuals that participate in the COPE are asked to indicate on a 4-points scale what they usually do when they experience a stressful event (1 = I usually don’t do this at all, 2 = I usually do this a little bit, 3 = I usually do this a medium amount, 4 = I usually do this a lot). Examples of items used in the COPE are: “I get upset and let my emotions out”, “I laugh about the situation”, “I daydream about things other than this” and “I think hard about what steps to take”.

The COPE, in its original form, measures coping style across different types of stressors. However, researchers have also adapted the scale to assess coping in specific situations (e.g. by replacing “stressful event” by a more narrow description of a potentially stressful situation), or by using only a selection of the 15 subscales. The COPE does not have an overall score, and the authors do not endorse combining subscales or deducing a dominant coping style from the scores (Carver, 2013).

Researchers that want to create their own coping scale (e.g. for specific groups, situations, or risk behaviours) can use the framework provided by Skinner et al. (2003), which identifies a dozen core families of coping based on action types. Any measure of coping, according to the authors, no matter the domain, developmental level, or time frame, should include ways of coping from each of these 12 families.

Different ways are used to classify coping styles into higher-order families. These classifications are often problematic (see Skinner et al., 2003 for a thorough discussion). Engagement and disengagement coping (see dct:engCoping and dct:disengCoping) is the higher-order structure adhered by Carver (2019), and the one used here. One should, however, take into account that the presence or absence of particular coping styles should be assessed at a higher level of specificity, e.g. coping styles pertaining to the 12 families identified by Skinner et al. (2003) or the 15 subscales of the COPE (Carver, 2013) should be selected. After an assessment is made at this higher level of specificity, subsequently one can potentially classify an individual’s dominant overall coping style as engagement (see dct:engCoping_79n2r0sy) or disengagement coping (see dct:disengCoping_79n2w1bk). Which particular coping styles researchers should direct their focus on, will depend on their appropriateness for the target event and age group (Skinner, 2007).

Questions or questionnaires that measure the way individuals handle potentiially stressful situations. A broad range of (self-report) instruments exist to assess coping styles, e.g there are scales for members of various groups (adults, children, adolescents, elderly) and broad and situation-specific application (e.g. health-related issues). The COPE (Carver, Scheier, & Weintraub, 1989) is the most frequently used instrument for broad application, and also one of the most commonly used measures in situation-specific contexts (Kato, 2015).

The 60-item COPE consists of 15 subscales (e.g. “active coping” and “acceptance”). Individuals that participate in the COPE are asked to indicate on a 4-points scale what they usually do when they experience a stressful event (1 = I usually don’t do this at all, 2 = I usually do this a little bit, 3 = I usually do this a medium amount, 4 = I usually do this a lot). Examples of items used in the COPE are: “I get upset and let my emotions out”, “I laugh about the situation”, “I daydream about things other than this” and “I think hard about what steps to take”.

The COPE, in its original form, measures coping style across different types of stressors. However, researchers have also adapted the scale to assess coping in specific situations (e.g. by replacing “stressful event” by a more narrow description of a potentially stressful situation), or by using only a selection of the 15 subscales. The COPE does not have an overall score, and the authors do not endorse combining subscales or deducing a dominant coping style from the scores (Carver, 2013).

Different ways are used to classify coping styles into higher-order families. Engagement and disengagement coping (see dct:engCoping and dct:disengCoping) is the higher-order structure used here. One should, however, take into account that the presence or absence of particular coping styles should be assessed at a higher level of specificity, e.g. coping styles pertaining to the 12 families identified by Skinner et al. (2003) or the 15 subscales of the COPE (Carver, 2013) should be selected. After an assessment is made at this higher level of specificity, subsequently one can potentially classify an individual’s dominant overall coping style as engagement (see dct:engCoping_79n2r0sy) or disengagement coping (see dct:disengCoping_79n2w1bk).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use one of these two questions: “What do you usually do when you experience a stressful event?” or “What do you usually do when you experience stress in SITUATION X/ from BEHAVIOUR X?”.

The first question is for assessing general coping strategies (i.e. across situations), the second question is for measuring coping styles associated with specific situations or behaviours. An example of the second question is: “What do you usually do when you experience stress at work?” or “What do you usually do when you experience stress from sleeping too little?”.

Expressions that demonstrate or imply a certain response to events or circumstances experienced as stressful (i.e. “coping”), without refering to specific strategies used.

When reference is made to specific coping strategies, these can be classified as either characteristic of engagement (see dct:engCoping_79n2r0sy) or disengagement coping (see dct:disengCoping_79n2w1bk). “Responses that are oriented toward either the source of stress, or toward one’s emotions and thoughts” are typical for engagement coping, while “responses that are oriented away from the stressor or one’s emotions/thoughts” may indicate the presence of a dominantly disengagement coping style (Compas et al., 2001).

Depending on their appropriateness for the target event and age group, the researcher can also adopt or develop a more specific coding scheme, for example based on the coping styles pertaining to the 12 families identified by Skinner et al. (2003) or the 15 subscales of the COPE (Carver, 2013), e.g. identifying denial, acceptance, humor, and other possible coping strategies.

Aspects of the environment or bodily experience perceived by the individual to trigger a strong affective state or bring to mind previously developed intentions or knowledge and, as a consequence, may create an instigation towards a particular (health) behaviour.

Use Likert scales that measure the degree to which participants believe a potential cue affects their engagement in the target behaviour. The following phrasing can be used: “CUE X would make me TARGET BEHAVIOUR”. For example: “If my ears would start ringing, I would start wearing earplugs”, “The smell of cannabis would make me smoke cannabis” and “Seeing a snack commercial on TV would make me eat a snack”. Reponses are given on a 5-point Likert scale ranging from “Very likely” to “Very unlikely”.

Observation and field experiments or other measurement methods that may be employed to measure the presence of cues. A choice for a particular measurement method may be prompted by a subject’s degree of awareness and introspective capacity with regard to the presence and influence (on behaviour) of cues. Diverse methods may potentially be combined to assess this awareness and introspective capacity for a diversity of cues. When self-reflective capacity is low, self-report methods preferably are dismissed.

Questions or questionnaires that measure the perceived degree to which participants believe a potential cue affects their engagement in the target behaviour. The following phrasing can be used: “CUE X would make me TARGET BEHAVIOUR”. For example: “If my ears would start ringing, I would start wearing earplugs”, “The smell of cannabis would make me smoke cannabis” and “Seeing a snack commercial on TV would make me eat a snack”. Reponses are given on a 5-point Likert scale ranging from “Very likely” to “Very unlikely”.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this question: “Please list any elements in the environment/bodily experiences that make it more likely that you engage in the TARGET BEHAVIOUR?” For example: “Please list any elements in the environment that make it more likely that you smoke cannabis?” and “Please list any audiological experiences that make it more likely that you would wear earplugs?”

Expressions that demonstrate or imply the influence of cues on performance of the target behaviour, for example: ““The smell of cannabis makes me smoke cannabis” and “seeing a snack commercial on TV makes me eat a snack”.

Expression that demonstrate or imply the influence of the perception of other people performing the target behaviour, should be coded as a descriptive norms (see dct:descriptiveNorms_73dnt5zp).

The sum of the descriptive norm beliefs.

A composite of descriptive norm beliefs is obtained by multiplying the perceived likelihood that a given normative referent performs the target behaviour (dct:referentBehavior_73dnt5zk) by the identification with that referent (dct:referentIdentification_73dnt5zl), and then summing the product terms across all descriptive norms beliefs (i.e. across all salient normative referents).

This, preferably, is represented graphically (e.g. in a table or index) and, additionally, in an equation. See Table 4.1 on p. 136 of Reasoned Action Approach book (Ajzen & Fishbein, 2010) for an example.

The descriptive norm beliefs composite reflects the summation of salient descriptive norm beliefs with regard to the target behaviour. Each descriptive norm belief (dct:descrNorms_belief_73dnt5zm), in its turn, is the product term of the multiplication of the perceived likelihood that a normative referent performs the target behavior (dct:referentBehavior_73dnt5zk) by the level of identification with the referent (dct:referentIdentification_73dnt5zl).

Given the formative nature of the descriptive norm beliefs composite, for a swift and clear overview of its composition and its effect on descriptive norms (dct:descriptiveNorms_73dnt5zp), preferably a graphical representation (e.g. a table or index) and , possibly, an equation is consulted. Table 4.1 on p. 136 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) represents an example of an overview of a (injunctive) norm beliefs composite.

Descriptive norm beliefs and, as a consequence, also descriptive norm beliefs composites (which are the sum of the former), can not be directly elicited. Descriptive norm beliefs are the product of a participant’s perception of a normative referent’s conduct of the target behaviour multiplied by the degree a participant wants to be like that normative referent with regards to the target behaviour.

See these two subconstructs (i.e. dct:referentBehavior_73dnt5zk and dct:referentIdentification_73dnt5zl) for instructions for eliciting their respective construct content.

Descriptive norm beliefs and, as a consequence, also descriptive norm beliefs composites (which are the sum of the former), can not be directly elicited. Descriptive norm beliefs are the product of a participant’s perception of a normative referent’s conduct of the target behaviour multiplied by the degree a participant wants to be like that normative referent with regards to the target behaviour.

See these two subconstructs (i.e. dct:referentBehavior_73dnt5zk and dct:referentIdentification_73dnt5zl) for instructions for coding their respective construct content.

A person’s belief that important individuals or groups in his/her life perform or don’t perform a particular behavior.

The measurement of descriptive norm beliefs starts with an elicitation study to identify the content of descriptive norm beliefs about the behaviour which are shared by the target population.

To obtain quantitative data regarding descriptive norm beliefs, the participant is subsequently asked to respond to two questions with respect to each salient descriptive norm belief, one to assess the perceived behaviour of a given social referent (dct:referentBehavior_73dnt5zk) and the other to measure the identification with that social referent (dct:referentIdentification_73dnt5zl).

Finally, to get a total quantitative score of a descriptive norm belief, the perceived behaviour of a social referent is multiplied by the identification with that social referent . The total score of a descriptive norm belief enables the assessment of the importance of the contribution of a descriptive norm belief, in comparison with other descriptive norm beliefs, to the descriptive norm beliefs composite (dct:descrNorms_belCom_73dnt5zn).

Each descriptive norm belief is the product term of the multiplication of the perceived behaviour of a given social referent (dct:referentBehavior_73dnt5zk) by the identification with that social referent (dct:referentIdentification_73dnt5zl).

Therefore, whether an descriptive norm belief exerts influence (and if so, to what extent) on the descriptive norm beliefs composite (dct:descrNorms_belCom_73dnt5zn), and whether a particular descriptive norm belief lowers or increases the perception that salient normative referents are performing the behaviour in question, can only be deduced by concurrent consultation of the two subcomponents of which it is constructed. Table 4.1 on p. 136 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) provides an example how the subcomponents exert an influence on individual normative beliefs (in Table 4.1 it concerns injunctive normative beliefs).

Descriptive norm beliefs can not be directly elicited. They are the product of a participant’s perception of a normative referent’s conduct of the target behaviour multiplied by the degree a participant wants to be like that normative referent with regards to the target behaviour

See these two subconstructs (i.e. dct:referentBehavior_73dnt5zk and dct:referentIdentification_73dnt5zl) for instructions for eliciting their respective construct content.

Descriptive norm beliefs can not be directly elicited. They are the product of a participant’s perception of a normative referent’s conduct of the target behaviour multiplied by the degree a participant wants to be like that normative referent with regards to the target behaviour

See these two subconstructs (i.e. dct:referentBehavior_73dnt5zk and dct:referentIdentification_73dnt5zl) for instructions for coding their respective construct content.

Descriptive norms refer to perceptions that others are or are not performing the behavior in question.

Use likert scales that measure perceived behavior of important social referents in general. It is important that no specific individuals are referenced, but that the items do refer to individuals that are similar, important or seen as important to follow. Note that the behavior in these items must always be the target behavior.

The example items in the 2010 RAA book are ‘Most people I respect and admire will TARGET BEHAVIOR.’ with anchors ‘Unlikely’ vs ‘Likely’; and ‘How many people like you TARGET BEHAVIOR?’ with anchors ‘Nobody’ vs ‘Everybody’.

Questions or questionnaires that measure the perceived degree to which important social referents in general perform or do not perform the target behaviour.

Examples of questions measuring descriptive norms (on a 7-point Likert scale) are: “‘Most people I respect and admire will TARGET BEHAVIOUR’ with anchors ‘Unlikely’ vs ‘Likely’; and ‘How many people like you TARGET BEHAVIOUR?’ with anchors ‘Nobody’ vs ‘Everybody’.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use the introduction ‘Sometimes, when we are not sure what to do, we look to see what others are doing’, and then ask the question: ‘Do persons important to you (engage in) TARGET BEHAVIOUR?’.

Reference here is made to a ‘generalized agent’ (i.e. important social referents in general) and no specific individuals are referenced. As reference to a specific individual is often unavoidable, one may opt not to elicit construct content with regard to descriptive norms, but limit the elicitation of construct content to the subconstructs dct:referentBehavior_73dnt5zk and dct:referentIdentification_73dnt5zl.

Expressions of perceived behavior of others, i.e. whether others perform the target behavior or not, where it is unclear who exactly is the other or group that is referred to. If the social referent is clear, this should be coded as the underlying constructs: see dct:referentBehavior_73dnt5zk and dct:referentIdentification_73dnt5zl.

Note that expressions of perceived approval or disapproval should be coded as dct:perceivedNorm_injunctive_73bg2wm7, and expressions of perceived norms where it is unclear whether they concern perceived (dis)approval or perceived approval should be coded as dct:perceivedNorm_71w98kk2.

Desire is a psychological state of motivation for a specific stimulus or experience that is anticipated to be rewarding. This state may or may not be consciously experienced.

In the context of behaviour change, the construct “desire” is covered by experiential attitude (see dct:experientialAttitude_73dnt5z5). For instructions for developing a measurement instrument, see the dct of this latter construct.

Questions or questionnaires that measure whether engagement in the target behaviour is perceived to be rewarding. As this operationalisation is similar to the measurement of experiential attitude (see dct:experientialAttitude_73dnt5z5), for behaviour change purposes, the use of the latter suffices. Therefore, instruments measuring this part of human psychology should be coded as dct:experientialAttitude_73dnt5z5.

In the context of behaviour change, the construct “desire” is covered by experiential attitude (see dct:experientialAttitude_73dnt5z5). See dct:experientialAttitude_73dnt5z5 for instructions for eliciting construct content.

In the context of behaviour change, the construct “desire” is covered by experiential attitude (see dct:experientialAttitude_73dnt5z5). See dct:experientialAttitude_73dnt5z5 for instructions for coding construct content.

Disengagement coping is effort to avoid confronting the threat or the stress-related emotions.

Numerous self-report instruments exist to assess coping styles (see dct:coping_79n2w1bj for a more extensive discourse). Nevertheless, regardless of the specific instrument used, an individual overall response pattern can be classified as either being typical of engagement (dct:engCoping_79n2r0sy) or disengagement coping, with “responses that are oriented toward either the source of stress, or toward one’s emotions and thoughts” representing engagement coping, while “responses that are oriented away from the stressor or one’s emotions/thoughts” indicating the dominant presence of a disengagement coping style (Compas et al., 2001).

Disengagement coping may be conceived at the highest level of a hierarchical factor structure with three levels (Tobin et al., 1989), with one level below, “problem disengagement” and “emotion disengagement”, and at the lowest level, respectively, “problem avoidance” and “wishful thinking” (for problem disengagement), and “self criticism” and “social withdrawal” (for emotion disengagement). Therefore, when an individidual scores high on these lower level coping strategies, and relatively low on the coping strategies falling under engagement coping (dct:engCoping_79n2r0sy), the dominant coping pattern can be classified as “disengagement coping”.

However, a problem is that each coping instrument assesses and classifies coping styles differently. For example, the aforementioned coping styles “problem avoidance” and “wishful thinking” are distinct categories of the Coping Strategies Inventory (Tobin et al., 1989), but not of many other coping style measures. Therefore, whether a particular coping style pertains to the higher-order class of engagement or disengament coping, should be decided based on factor analysis or an a priori categorization of each individual coping style as either falling under the umbrella of engagement or disengament coping.

Questions or questionnaires that measure a so-called “disengagement” coping style (i.e. efforts to avoid confronting the threat or the stress-related emotions) when faced with stressful situations. A broad range of (self-report) instruments exist for this purpose, e.g the COPE (Carver, Scheier, & Weintraub, 1989). Each of these instruments assesses and classifies the different coping styles differently.

Therefore, whether a particular coping style pertains to the higher-order class of engagement or disengament coping, should be decided based on factor analysis or an a priori categorization of each individual coping style as either falling under the umbrella of engagement (dct:engCoping_79n2r0sy) or disengament coping, with “responses that are oriented toward either the source of stress, or toward one’s emotions and thoughts” representing engagement coping, while “responses that are oriented away from the stressor or one’s emotions/thoughts” indicating the dominant presence of a disengagement coping style (Compas et al., 2001).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use one of these two questions: “What do you usually do when you experience a stressful event?” or “What do you usually do when you experience stress in SITUATION X/ from BEHAVIOUR X?”.

The first question is for assessing general coping strategies (i.e. across situations), the second question is for measuring coping styles associated with specific situations or behaviours. An example of the second question is: “What do you usually do when you experience stress at work?” or “What do you usually do when you experience stress from sleeping too little?”.

“Responses that are oriented away from the stressor or one’s emotions/thoughts” (Compas et al., 2001) may indicate the dominant presence of a disengagement coping style. Coping responses that indicate the presence of a disengagement coping style are, for example, denial, restraint, and substance use. Therefore, expressions that demonstrate or imply the presence of such coping styles should be coded as disengagement coping (dct:disengCoping_79n2w1bk).

“Responses that are oriented toward either the source of stress, or toward one’s emotions and thoughts” (Compas et al., 2001) are typical for engagement coping. Coping responses that indicate the presence of an engagement coping style are, for example, acceptance, humor, and planning. Therefore, expressions that demonstrate or imply the presence of such coping styles should be coded as engagement coping (dct:engCoping_79n2r0sy).

Expressions that demonstrate or imply a certain response to events or circumstances experienced as stressful (i.e. “coping”), without refering to specific strategies used, should be coded as dct:coping_79n2w1bj.

Depending on their appropriateness for the target event and age group, the researcher can also adopt or develop a more specific coding scheme, for example based on the coping styles pertaining to the 12 families identified by Skinner et al. (2003) or the 15 subscales of the COPE (Carver, 2013), e.g. identifying coping styles such as denial, restraint, substance use, and other coping strategies typical of disengagement coping.

People’s perceptions of situations and circumstances in which their ability is reduced to control over dominant responses with respect to the target behaviour.

Present participants with items describing situations and circumstances that may potentially reduce the capacity for self-control, and ask participants to which extent they believe their ability is reduced to control over dominant responses with respect to the target behaviour. Responses are given on a 5-point Likert scale ranging from “Severly reduced” to “Not reduced at all”.

Examples of such items are: “Is your capacity to control the expression of vicious comments to friends or family reduced when preparing for an important test or exam?”, “Is your capacity to control your alcohol intake reduced when tired from an exhausting day at work?”, “Is your ability to control your exposure to loud music (e.g. by wearing ear plugs, restricting volume, or keeping a distance from loudspeakers) reduced when under the influence of alcohol of drugs? “.

Note that various situations and circumstances are depicted that may temporarily reduce self-control. The definition of “ego depletion” in this DCT is thus broader than in the original definition of ego depletion by Baumeister et al. (1994). “Ego depletion” in the present DCT can be the result of a prolonged effort to exert self-control (in line with the original defintion by Baumeister et al.), but it can also be prompted by other internal or external factors.

For each behaviour and group of persons, different situations and circumstances may reduce the capacity for self-control. Therefore, specific items are composed for each target behaviour and target group.

“Ego depletion” is its original meaning (i.e. ego depletion is reduced performance on self-control tasks resulting from engaging in preceding tasks requiring self-control) as formulated by Baumeister et al. 1994) is usually assessed by using a sequential-task experimental paradigm, in which participants engage in two consecutive tasks.

The range of situations and circumstances that may elicit “ego depletion” in the definition of the present DCT is broader. Moreover, it reflects an individual’s perception of situations and circumstances in which ego depletion occurs. This perception is not necessarily a reflection of actual ego depletion taking place. Therefore, the sequential-task experimental paradigm is not an appropriate method to measure the present construct.

To measure the broader meaning of ego depletion, questions are presented in which participants are asked to which extent they believe their ability is reduced to control dominant responses with respect to the target behaviour in the situation or circumstances described. For example: “Is your capacity to control the expression of vicious comments to friends or family reduced when preparing for an important test or exam?” or “Is your ability to control your exposure to loud music (e.g. by wearing ear plugs, restrict the volume, or keeping a distance from loudspeakers) reduced when under the influence of alcohol of drugs? “. Responses on these items can be given on a 5-point Likert scale ranging from “Severly reduced” to “Not reduced at all”.

Note that to accurately measure ego depletion, items have to match with the specific target behaviour and target group.

Ego depletion, as defined in the present construct (dct:ego_depletion_79n2r0sy) can be regarded as a subconstruct of dct:capacity_73dnt602, i.e. it represents a temporary reduced capacity for self-control (dct:self_control_79n2fh4t), induced by the situation or by circumstances.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this question: “In what situations or circumstances is your ability to control TARGET BEHAVIOUR reduced?”. For example: “In what situations or circumstances is your ability to control your food consumption reduced?”.

Expressions that describe situations or circumstances in which one’s ability to exert self-control with respect to the target behaviour is temporarily reduced. For example, expressions such as “When I have drunk alcohol, it is very difficult to control my eating behaviour” or “When I have a lot to do at work, I usually no longer control my eating behaviour and frequently eat junk food”.

Expressions that refer to self-control with regard to the target behaviour in general, not refering to situations or circumstances in which this ability is temporarily reduced, should be coded as dct:self_control_79n2fh4t

Emotions are states that follow from the appraisal of a specific situation as facilitating or endangering the achievement of an important goal, involving behavioural, physiological, affective, and cognitive elements.

Emotions affect an individual on different levels (behaviour, physiology, affect, cognition). As a consequence, multiple measurement methods may be employed, and possibly integrated, to measure emotion. Self-report, other-report, observation, and neurophysiology are the measurement methods most dominantly used in emotion research.

A three-step system for choosing a measure of emotion can be employed (Ekkekakis, 2008): 1) Decide whether you wish to study core affect, emotion, or mood (these are related but different constructs); 2) Choose the most appropriate theoretical framework for the chosen construct; 3) Select the psychometrically strongest measure based on the chosen theoretical framework.

In the context of behaviour change, emotions may exert an influence on the target behaviour in various ways. The mere thought of performing the behaviour may trigger a strong emotion and in- or decrease the likelihood of performing the behaviour. Emotions may also affect the actual performance of the behaviour. The kinds of emotion that are experienced before, during, and after the behaviour may also be diverse (e.g. anger, shame, or joy).

The foregoing must be taken into account when a measurement instrument is chosen or developed, e.g. if a (fictitious) research hypothesis is that the experience of “positive” emotions (e.g. joy, pride) and “negative” emotions (e.g. anger, shame), respectively, de- and increase the number of cigarettes smoked during a day, then retrospective assessment of experienced emotions with a self-report questionnaire likely is not the most appropriate instrument, but an measurement method that is able to record emotions at various moments of the day such as ecological momentary assessments (EMAs) is a better candidate.

Self-report, other-report, observation, and neurophysiology measures that assess emotions associated with the target behaviour, and/or emotions experienced before, during, and/or after engagement in the target behaviour.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use one or more of these questions: “Please list any emotions you experience when you think about engaging in the TARGET BEHAVIOUR”, “Please list any emotions you experience before engaging in the TARGET BEHAVIOUR”, “Please list any emotions you experience while engaging in the TARGET BEHAVIOUR”, and “Please list any emotions you experience after engaging in the TARGET BEHAVIOUR”.

Expressions that demonstrate or imply the presence or absence of emotions, either by directly refering to the experience of an emotion (e.g. “I was very happy at that moment” or “That made me extremely angry”) or by expressing behaviour typical of the experience of emotions (“I collapsed to the ground and cried when I heard the bad news” or “My head turned bright red and I didn’t dare to look at my friends when i saw the video from the night before”).

Engagement coping is aimed at dealing with the stressor or reducing stress-related emotions.

Numerous self-report instruments exist to assess coping styles (see dct:coping_79n2w1bj for a more extensive discourse). Nevertheless, regardless of the specific instrument used, an individual overall response pattern can be classified as either being typical of engagement or disengagement coping (see dct:disengCoping_79n2w1bk), with “responses that are oriented toward either the source of stress, or toward one’s emotions and thoughts” representing engagement coping, while “responses that are oriented away from the stressor or one’s emotions/thoughts” indicating the dominant presence of a disengagement coping style (Compas et al., 2001).

Engagement coping may be conceived at the highest level of a hierarchical factor structure with three levels (Tobin et al., 1989), with one level below, “problem engagement” and “emotion engagement”, and at the lowest level, respectively, “problem solving” and “cognitive restructuring” (for problem engagement), and “express emotions” and “social support” (for emotion engagement). Therefore, when an individidual scores high on these lower level coping strategies, and relatively low on the coping strategies falling under disengagement coping (dct:disengCoping_79n2w1bk), the dominant coping pattern can be classified as “engagement coping”.

However, a problem is that each coping instrument, measures and classifies coping styles differently. For example, the aforementioned coping styles “problem solving” and “cognitive restructuring” are distinct categories of the Coping Strategies Inventory (Tobin et al., 1989), but not of many other coping style measures. Therefore, whether a particular coping style pertains to the higher-order class of engagement or disengament coping, should be decided based on factor analysis or an a priori categorization of each individual coping style as either falling under the umbrella of engagement or disengagement coping.

Questions or questionnaires that measure a so-called “engagement” coping style (i.e. dealing with the stressor or reducing stress-related emotions) when faced with stressful situations. A broad range of (self-report) instruments exist for this purpose, e.g the COPE (Carver, Scheier, & Weintraub, 1989). Each of these instruments classifies the different coping styles differently.

Therefore, whether a particular coping style pertains to the higher-order class of engagement or disengament coping (see dct:disengCoping_79n2w1bk), should be decided based on factor analysis or an a priori categorization of each individual coping style as either falling under the umbrella of engagement or disengament coping, with “responses that are oriented toward either the source of stress, or toward one’s emotions and thoughts” representing engagement coping, while “responses that are oriented away from the stressor or one’s emotions/thoughts” indicating the dominant presence of a disengagement coping style (Compas et al., 2001).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use one of these two questions: “What do you usually do when you experience a stressful event?” or “What do you usually do when you experience stress in SITUATION X/ from BEHAVIOUR X?”.

The first question is for assessing general coping strategies (i.e. across situations), the second question is for measuring coping styles associated with specific situations or behaviours. An example of the second question is: “What do you usually do when you experience stress at work?” or “What do you usually do when you experience stress from sleeping too little?”.

“Responses that are oriented toward either the source of stress, or toward one’s emotions and thoughts” (Compas et al., 2001) are typical for engagement coping. Coping responses that indicate the presence of an engagement coping style are, for example, acceptance, humor, and planning. Therefore, expressions that demonstrate or imply the presence of such coping styles should be coded as engagement coping (dct:engCoping_79n2r0sy).

“Responses that are oriented away from the stressor or one’s emotions/thoughts” (Compas et al., 2001) may indicate the dominant presence of a disengagement coping style. Coping responses that indicate the presence of a disengagement coping style are, for example, denial, restraint, and substance use. Therefore, expressions that demonstrate or imply the presence of such coping styles should be coded as disengagement coping (dct:disengCoping_79n2w1bk).

Expressions that demonstrate or imply a certain response to events or circumstances experienced as stressful (i.e. “coping”), without refering to specific strategies used, should be coded as dct:coping.

Depending on their appropriateness for the target event and age group, the researcher can also adopt or develop a more specific coding scheme, for example based on the coping styles pertaining to the 12 families identified by Skinner et al. (2003) or the 15 subscales of the COPE (Carver, 2013), e.g. identifying denial, restraint, substance use, and other possible coping strategies typical of disengagement coping.

The sum of the experiential attitude beliefs.

A composite of experiential attitude beliefs (i.e. the expectancy-value index) for the target behaviour is obtained by multiplying each experiential attitude belief’s expectation strength (dct:expAttitude_expectation_73dnt5z1) by its evaluation (dct:expAttitude_evaluation_73dnt5z2) and then summing the product terms across all experiential attitude beliefs. This, preferably, is represented graphically (e.g. in a table or index) and, additionally, in an equation.

Ajzen & Fishbein state on p.104 of the Reasoned Action Approach book (2010) that the composite measure of beliefs is assumed to determine the (experiential) attitude, and should not be interpreted as an indirect measure of (experiential) attitude. A low correlation between the beliefs composite and a direct measure of experiential attitude may be an indication that the expectancy-value index is based on an inappropriate set of (nonsalient) beliefs, or that the measure of belief strength may be scored inappropriately.

Questions or questionnaires that measure the expectations (dct:expAttitude_expectation_73dnt5z1) and evaluations (dct:expAttitude_evaluation_73dnt5z2) of all salient experiential attitude beliefs with regard to the target behaviour.

Experiential attitude belief expectations combine multiplicatively with experiential attitude belief evaluations into experiental attitude beliefs. The product terms for each belief and the sum of the products terms of all salients beliefs are represented in the expectancy-value index (i.e. the compostite of attitude beliefs) for the target behaviour. See Table 3.2 on p. 98 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) for an example of such an index.

Experiential attitude beliefs and, as a consequence, also experiential attitude beliefs composites (which are the sum of the former), can not be directly elicited. Experiential attitude beliefs are the product of the product of expected experiential consequences of a target behaviour (expAttitude_expectation_73dnt5z1) multiplied by the evaluation of those consequences (expAttitude_evaluation_73dnt5z2).

See these two subconstructs for instructions for eliciting their respective construct content.

Experiential attitude beliefs and, as a consequence, also experiential attitude beliefs composites (which are the sum of the former), can not be directly elicited. Experiential attitude beliefs are the product of the product of expected experiential consequences of a target behaviour (expAttitude_expectation_73dnt5z1) multiplied by the evaluation of those consequences (expAttitude_evaluation_73dnt5z2).

See these two subconstructs for instructions for coding their respective construct content.

Beliefs are defined as the subjective probability that an object has a certain attribute, where ‘object’ and ‘attribute’ are used in the generic sense, and refer to any discriminable aspect of an individual’s world. Conform the expectancy-value model of attitude, beliefs are the product of the expectation that engaging in a behavior will have a given consequence and the evaluation of that consequence as (very) negative or (very) positive.

Experiential attitude beliefs concern beliefs about consequences of engaging in the behavior that influence one’s experiences, for example by causing pleasure or pain.

Experiential attitude beliefs are measured by measuring both of the component constructs, i.e.: experiential attitude belief expectations (see dct:expAttitude_expectation_73dnt5z1) and experiential attitude belief evaluation (see dct:expAttitude_evaluation_73dnt5z2).

Each experiential attitude belief is the product term of the multiplication of the evaluation of the perceived experiential consequences of the target behaviour (dct:expAttitude_evaluation_73dnt5z2) by the perceived likelihood that engaging in the behaviour will actually have these consequences (dct:expAttitude_expectation_73dnt5z1).

Therefore, whether an experiential attitude belief exerts influence (and if so, to what extent) on the experiential attitude beliefs composite (dct:expAttitude_belCom_73dnt5z4), and whether a particular experiential attitude belief leads to a more favourable or more unfavourable experiential attitude towards performing the target behaviour, can only be deduced by concurrent consultation of the two subcomponents of which it is constructed. Table 3.2 on p. 98 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) illustrates how the two subcomponents can exert an influence on individual attitude beliefs.

Experiential attitude beliefs can not be directly elicited. They are the product of expected experiential consequences of a target behaviour (dct:expAttitude_expectation_73dnt5z1) multiplied by the evaluation of those consequences (dct:expAttitude_evaluation_73dnt5z2).

See these two subconstructs for instructions for eliciting their respective construct content.

Experiential attitude beliefs can not be directly elicited. They are the product of expected experiential consequences of a target behaviour (dct:expAttitude_expectation_73dnt5z1) multiplied by the evaluation of those consequences (dct:expAttitude_evaluation_73dnt5z2).

See these two subconstructs for instructions for coding their respective construct content.

An experiential attitude belief evaluation is the evaluation as positive or negative (i.e. desirable or undesirable) of one specific experiential potential consequence of a behavior. The experiential nature of this evaluation means that this concerns mostly expected experiences and sensations, such as pleasure or pain. Consequences that render the target behavior less or more instrumental given more long-term goals are captured in instrumental attitude and the underlying beliefs (see dct:expAttitude_expectation_73dnt5z1 and dct:instrAttitude_expectation_73dnt5z6 for more details about this distinction).

According to the Reasoned Action Approach, experiential attitude belief evaluations combine multiplicatively with experiential attitude belief expectations (see dct:expAttitude_expectation_73dnt5z1) into experiential attitude beliefs (see dct:expAttitude_belief_73dnt5z3).

Examples of experiential attitude belief evaluations are (note that the examples provided for the experiential attitude belief expectations should probably be read first, see dct:expAttitude_expectation_73dnt5z1): the degree to which one finds it pleasant to sit down during a bus journey to the city centre; the evaluation of watching the scenery as desirable; one’s evaluation of a pizza’s taste as nice; one appraising a relaxed feeling as pleasant; one enjoying the sensation of sex without a condom.

Experiential attitude belief evaluations are measured using questions in a questionnaire that are referred to as ‘items’. Experiential attitude belief evaluation always relate to specific experiential attitude belief expectations, so the first step is to identify the expectation for which the evaluation should be measured (e.g. ‘Drinking alcohol makes me feel … [much more relaxed|much more excited]'). Evaluations of expectations are always measured on a bidimensional scale.

As item stem, use ‘I prefer …’, and as anchors, include the scale extremes used when measuring the expectation. For example, an item could be ‘I prefer … [being much more relaxed|being much more excited]’. Make sure to formulate the item stem such that it makes clear that you are asking people about their evaluation (not what they think holds more generally). Use a seven-point response scale and try to always be consistent in the scale valence; in languages that are read from left to right, always place the most passive/low/less/weak/unlikely scale extreme (anchor) on the left, and the most active/high/more/strong/likely scale extreme (anchor) on the right. Do not reverse this order for one or more items.

The item or set of items should be accompanied by an instruction that makes clear that you are asking people about their evaluation (not what they think holds more generally).

Experiential attitude belief evaluations are measured using questions in a questionnaire (these are referred to as ‘items’). Items can be coded as measuring an experiential attitude belief evaluation if they measure participants’ evaluation of one specific experiential potential consequence of a behavior as positive or negative (i.e. desirable or undesirable).

The experiential nature of this evaluation means that this concerns mostly expected experiences and sensations, such as pleasure or pain.

The question should include, either in its stem or in its anchors, the specific dimension of which the evaluation is being measured (e.g. whether people prefer feeling relaxed or feeling excited, or whether people evaluate being excited as positive or negative).

Note that evaluations of consequences that render the target behavior less or more instrumental given more long-term goals are captured in instrumental attitude and the underlying beliefs. (see dct:expAttitude_expectation_73dnt5z1 and dct:instrAttitude_expectation_73dnt5z6 for more details about this distinction, and see dct:instrAttitude_evaluation_73dnt5z7 for the coding instruction for instrumental attitude belief evaluations).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources (i.e. transcripts or notes) are then coded using the instruction for aspect coding. In this qualitative study, use these questions: ‘What do you see as the advantages of you engaging in [target behavior]?’, ‘What do you see as the disadvantages of you engaging in [target behavior]?’, and ‘What else comes to mind when you think about [target behavior]?’.

Expressions of one’s evaluations as positive or negative (i.e. desirable or undesirable) of experiential potential consequences of a behavior. The experiential nature of this evaluation means that this these evaluations concern mostly expected experiences and sensations, such as pleasure or pain.

Note that expressions of evaluations of consequences that render the target behavior less or more instrumental given more long-term goals are captured in instrumental attitude and the underlying beliefs (see dct:instrAttitude_evaluation_73dnt5z7).

One’s actual expectation (i.e. whether a consequence is likely or unlikely) should be coded as dct:expAttitude_expectation_73dnt5z1.

An experiential attitude belief expectation is the expectation of how probable (i.e. unlikely versus likely) it is that engaging in the target behavior will cause one specific experiential potential consequence to come about. The experiential nature of this expectation means that these expected consequences must concern experiences and sensations, such as pleasure or pain.

Each experiential attitude belief expectation covers one specific acute hedonic expectation, disconnected from potential long-term consequences the target behavior may have. Experiential attitude belief expectations always refer to immediately experienced consequences of a behavior. Expectations about consequences that render the target behavior less or more desirable without the immediate experiential effects of the behavior playing a role are captured in instrumental attitude belief expectations (see dct:instrAttitude_expectation_73dnt5z6).

According to the Reasoned Action Approach, experiential attitude belief expectations combine multiplicatively with experiential attitude belief evaluations (see dct:expAttitude_evaluation_73dnt5z2) into experiental attitude beliefs (see dct:expAttitude_belief_73dnt5z3).

Examples of experiential attitude belief expectations are: the expectation that taking the bus to the city centre will allow one to sit down during the journey, which one finds a pleasant feeling; the expectation that walking to the city centre will allow one to watch the scenery, which one enjoys; the expectation that eating a pizza will taste nice; the expectation that going for a run will feel relaxing; and the expectation that having sex without a condom will feel more pleasant than having sex with a condom.

Experiential attitude belief expectations are measured using questions in a questionnaire that are referred to as ‘items’. To measure an experiential attitude belief expectation, first, identify exactly which potential experiential consequence of the target behavior (e.g. a specific experience or sensation) you want to measure. Then, establish whether accurately describing the spectrum of possibilities regarding this experiential consequence requires a unidimensional scale or a bidimensional scale. Most potential consequences of a target behavior can be perceived either to increase or decrease upon performance of the target behavior, requiring a bidimensional scale. For example, some people might expect they will feel less relaxed if they drink alcohol, while others might expect they will feel more relaxed. However, in rare cases, one of the two dimensions can be excluded a priori, in which case a unidimensional scale suffices. For example, the degree to which people will expect they will feel less hungry may vary; but it is excessively unlikely that somebody might expect that if they eat a whole pizza, they will then feel more hungry. Therefore, in that case, one might want to choose a unidimensional scale. In general, a rule of thumb is that if the ‘default state’ of this experiential consequence resembles absence of the experience, and therefore, engaging in the target behavior can only have an effect in one direction, a unidimensional scale can be used. However, when engaging in the target behavior can conceivably increase or decrease this experiential consequence, a bidimensional scale is required.

Once it is clear whether a bidimensional or unidimensional scale should be used, the construction of the item stem and the two anchors can start. Make sure to formulate the item stem such that it makes clear that you are asking people about their expectation (not what they think holds more generally).

For unidimensional scales, create an item stem that explicitly lists the single dimension that expresses the experiential consequence (e.g. feeling full after eating a pizza). As anchors, always use ‘Very unlikely’ and ‘Very likely’. An example item would be: ‘If I eat a whole pizza at once, it is … that it will make me feel full. [Very unlikely|Very likely]’. However, as explained above, unidimensional scales can rarely be used in most circumstances, so bidimensional scales are usually required.

For bidimensional scales, the item stem cannot explicitly list only one of the two dimensions (e.g. ‘feeling much less relaxed’ or ‘feeling much more relaxed’). This is because this would create a unidimensional or ambiguous response scale. People who would score low on the unidimensional scale might mean either that they don’t think that consequence will occur, or that they think that the opposite consequence will occur. For example, when creating an item ‘If I drink a glass of alcohol, I will feel much more relaxed. [Very unlikely|Very likely]’, people who respond ‘Very unlikely’ can mean either that they expect to feel much less relaxed, or that they expect that drinking a glass of alcohol will have no effect on how relaxed they will feel.

Therefore, capturing the full potential breadth of the beliefs of your target population requires asking what they expect exactly. To do this, create an item stem that contains the target behavior, and anchors that express the extremes of the bidimensional scale. For example, ‘Drinking alcohol makes me feel … [much less relaxed|much more relaxed]’. Sometimes, the two extremes of the dimension you want to measure can be expressed in two antonyms, such as ‘Drinking alcohol makes me feel … [much more relaxed|much more excited]’.

Once the item stem and the two anchors have been determined, decide which response scale to use. For bidimensional scales, seven-point scales are preferred, as these leave three degrees of expression in each dimension (the mid-point representing the expectation that the behavior does not have a consequence regarding this specific experiential attitude belief). For unidimensional scales, five-point scales suffice. Try to always be consistent in the scale valence; in languages that are read from left to right, always place the most passive/low/less/weak/unlikely scale extreme (anchor) on the left, and the most active/high/more/strong/likely scale extreme (anchor) on the right. Do not reverse this order for one or more items.

When combining multiple items in one measurement instrument, if both evaluations of unidimensional consequences and evaluations of bidimensional consequences are measured, either use two matrices or combine them in one that uses seven-point scales for all items.

The item or set of items should be accompanied by an instruction that makes clear that you are asking people about their expectation (not what they think holds more generally).

Experiential attitude belief expectations are measured using questions in a questionnaire (these are referred to as ‘items’). Items can be coded as measuring an experiential attitude belief expectation if they measure participants’ expectation of how probable (i.e. unlikely versus likely) it is that engaging in the target behavior will cause one specific experiential potential consequence to come about. The experiential nature of this expectation means that these expected consequences must concern experiences and sensations, such as pleasure or pain. These experiential attitude belief expectations cover acute hedonic expectations, disconnected from potential long-term consequences the target behavior may have. Experiential attitude belief expectations always refer to immediately experienced consequences of a behavior. The question should include, either in its stem or in its anchors, the specific expectation being measured. Note that items concerning consequences that render the target behavior less or more desirable without the immediate experiential effects of the behavior playing a role are captured in instrumental attitude belief expectations (see dct:instrAttitude_expectation_73dnt5z6).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources (i.e. transcripts or notes) are then coded using the instruction for aspect coding. In this qualitative study, use these questions: ‘What do you see as the advantages of you engaging in [target behavior]?’, ‘What do you see as the disadvantages of you engaging in [target behavior]?’, and ‘What else comes to mind when you think about [target behavior]?’.

Expressions of expectations of how probable (i.e. unlikely versus likely) it is that engaging in the target behavior will cause an experiential potential consequence to come about. The experiential nature of this expectation means that these expected consequences must concern experiences and sensations, such as pleasure or pain. These experiential attitude belief expectations cover acute hedonic expectations, disconnected from potential long-term consequences the target behavior may have. Experiential attitude belief expectations always refer to immediately experienced consequences of a behavior. Note that expressions of expectations of consequences that render the target behavior less or more desirable without the immediate experiential effects of the behavior playing a role are captured in instrumental attitude belief expectations (see dct:instrAttitude_expectation_73dnt5z6). One’s evaluation in terms of valence (i.e. positive versus negative) of the consequence should be coded as dct:expAttitude_evaluation_73dnt5z2.

A latent disposition or tendency to respond with some degree of favorableness or unfavorableness to the target behavior based on what one expects to experience if engaging in the target behavior.

Use semantic differentials with root ‘For me, TARGET BEHAVIOR is …’ and a bidimensional scale where the right-most anchor expresses a pleasant affective state and the left-most anchor expresses the opposite unpleasant affective state (e.g. ‘unpleasant’ versus ‘pleasant’).

The example items in the 2010 RAA book are ‘For me, TARGET BEHAVIOR is …’ with anchors ‘Bad’ vs ‘Good’; ‘For me, TARGET BEHAVIOR is …’ with anchors ‘Unpleasant’ vs ‘Pleasant’; ‘For me, TARGET BEHAVIOR is …’ with anchors ‘Harmful’ vs ‘Beneficial’; and ‘For me, TARGET BEHAVIOR is …’ with anchors ‘Boring’ vs ‘Interesting’.

Operationalisations that measure affective aspects of the latent disposition or tendency to respond favourably versus unfavourably to target behavior, for example using the semantic differentials ‘pleasant’ vs ‘unpleasant’ or ‘fun’ vs ‘boring’.

Experiential attitude is defined as a construct that is the consequence of a person’s evaluation of the experiential attitude belief composite.

Expressions that demonstrate or imply affective aspects of the latent disposition or tendency to respond favourably versus unfavourably to the target behaviour at a very general level (e.g. “Face masks just feel bad."). If the expressions are more specific, relating to specific experiences, sensations, or feelings (e.g. “I don’t like the way face masks feel against my skin”), code them as [[……….]].

Manifestations that express favourableness or unfavourableness to the target behaviour based on the perceived usefulness of engaging in the target behaviour to achieving one’s goals (e.g. “Face masks don’t help”) should be coded as dct:instrumentalAttitude_73dnt5zb. If the expressions are so general and vague that they don’t clearly relate to experiential aspects of attitude, code them as dct:attitude_73dnt5zc.

Behaviour is externally regulated when an activity is adopted to acquire an outcome that is separable from the activity itself.

The degree to which engagement in the target activity is externally regulated can be assessed by asking participants “Why are you currently engaged in this activity?” and representing a series of items containing potential answers to this question. Responses to each item are given on a 7-point Likert scale ranging from “Does not correspond at all” to “Corresponds exactly”.

Examples of items expressing externally regulated behaviour are: “Because I am supposed to do it”, “Because it is something that I have to do”, and “Because I feel that I have to do it”.

The procedure and items described here are from the Situational Motivation Scale (SIMS) (Guay et al., 2000). Beside items containing statements deemed typical for externally regulated behaviour, the SIMS contains items to assess identified regulation, intrinsic regulation, and amotivation. The items belonging to each particular subscale are summed to assess which motivational construct dominates. The SIMS is used to assess motivation for various activities such as physical exercise and academic performance.

Instruments that measure the degree to which engagement in the target activity is externally regulated.

The Situational Motivation Scale (SIMS) (Guay et al., 2000) is a frequently used self-report instrument to assess whether behaviour is externally regulated. Beside external regulation of behaviour, the SIMDS assesses identified regulation, intrinsic regulation, and amotivation, and is used to assess motivation for various activities (e.g. physical exercise, academic performance).

The SIMS asks participants “Why are you currently engaged in this activity?”. Scale items represent potential answers to this question. Responses are given on a 7-point Likert scale ranging from “Does not correspond at all” to “Corresponds exactly”.

Examples of statements expressing external regulation of behaviour are: “Because I am supposed to do it”, “Because it is something that I have to do”, and “Because I feel that I have to do it”.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this question: “Why are you currently engaged in TARGET BEHAVIOUR?”

For example: “Why do you currently wear ear plugs at concerts?” or “Why do you currently follow a calorie-restricted diet?”

Expressions that demonstrate or imply that engagement in the target behaviour is externally regulated, using phrases such as “Because I am supposed to do it” and “Because it is something that I have to do”.

Expressions that demonstrate or imply the presence of introjected regulation of behaviour should be coded as dct:introjBehaviour_79n2r0t0.

Expressions that demonstrate or imply the presence of identified regulation of behaviour should be coded as dct:identifiedBehaviour_79n2r0t0.

Expressions that demonstrate or imply the presence of integrated regulation of behaviour should be coded as dct:integratedBehaviour_79n2r0t0.

In practice, it can be difficult to distinguish expressions belonging to the four types of extrinsic motivation (external, introjected, identified, and integrated self-regulation). Therefore, a less precise categorization can be made between controlled and autonomous motivation: external and introjected regulation belonging to the first category, and identified and integrated self-regulation belonging to the first category.

Expressions that demonstrate or imply the presence of intrinsically regulated behaviour should be coded as dct:intrinsicBehaviour_79n2r0t0.

Expressions that demonstrate or imply the absence of motivation to engage in the target behaviour (i.e. “amotivation”) should be coded as dct:motivation_79n2fh4q, and in a supplementary comment scored “0” or “absent”.

Expressions that refer to the presence of motivation with regard to the target behaviour without specifying the nature of its regulation or expressions that refer to the presence of motivation in general should be coded dct:motivation_79n2fh4q.

The representation of a desired state.

Make 5- or 7-point Likert scales that measure the extent participants think engagement in the target behavior will help them reach their goals. For example: “TARGET BEHAVIOUR will help me reach my goals”, with anchors “Totally agree” and “Totally disagree”.

Note that no specified goals are formulated. If one wants to identify the goals that are associated with a target behaviour, an elicitation study among the study population can be performed. This will identify the most salient goal beliefs.

After the elicitation study, participants can be asked to respond to two questions with respect to each salient goal belief, one question to assess the perceived likelihood that engagement in the target behaviour will lead to attainment of the goal, and the other to measure the perceived importance of this goal.

To get a total quantitative score of a goal belief, the perceived likelihood that engagement in the target behaviour will lead to attainment of the goal is multiplied by the perceived importance of the goal. The total score of a goal belief enables the assessment of the impact of a specific goal belief, in comparison with other goal beliefs, and thus helps to identify the most important goals people may have for engaging in the target behaviour.

Questions or questionnaires that measure that the extent participants think engagement in the target behaviour will help them reach their goals. For example: “TARGET BEHAVIOUR will help me reach my goals”, with anchors “Totally agree” and “Totally disagree”.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, ask the question: “What role does TARGET BEHAVIOUR play in reaching your goals?”.

In the aforementioned question “goal” is not specified. If one wants to identify salient goal beliefs, the following question may be asked: “What do you consider your most important goals for (engaging in) TARGET BEHAVIOUR?”

Expressions that demonstrate or imply the degree of importance, performance of the target behaviour has for the participant in reaching his or her goals, e.g. using words and phrases such as “important/unimportant” and “major/minor role”.

A habit is an action or behavioural tendency that it enacted spontaneously, with little conscious awareness or reflection, in response to a set of associated conditions or contextual cues.

The four most widely used instruments to measure habit are so-called “association measures”, the Behavior Frequency x Context Stability measure (BFCS), the Self-Report Habit Index (SRHI), and the Self-Report Behavioral Automaticity Index (SRBAI) (Van Bree, 2018). The choice of a habit measure depends on the goal of the measurement and the setting where the assessment is ought to take place; each of the four aforementioned instruments tap on different aspects of habit (e.g. automaticity, repetition, and self-identity in case of the SRHI), and while the latter three measures can be incorprated into a survey, association measures, in contrast, can only be administered in a controlled laboratory setting.

The 12-item SRHI (Verplanken & Orbell, 2003) is the instrument most frequently used to measure habit (Gardner, 2015). The SRHI is supposed to measure behaviour performed across contexts. However, specific context cues can be added by modifying the stem of SRHI-items.

All 12 SRHI-items start with the sentence “TARGET BEHAVIOUR is something…”, after which a particular statement follows for each individual item, such as “I do frequently”, “I do automatically”, and “That would require effort not to do”. The items are accompanied by response scales anchored by agree/disagree and contain five or more response categories.

An example of an item measuring “smoking cigarettes” (the target behaviour) as habit, thus, would be: “Smoking cigarettes is something I do automatically”.

A critique on self-report measures of habit, such as the SRHI, is that they require respondents to reflect on behaviours that are automatic and, as such, outside of conscious awareness (Van Bree, 2018).

Questions or questionnaires that measure the perceived degree to which target behaviour is habitual. Depending on the theoretical framework in which an author operates, and the specific instrument used, the focus may be on different aspects of habitual behaviour (e.g. automaticity, repetition, and self-identity).

Sample items measuring habit on a 5-point Likert scale with anchors “Totally agree” and “Totally disageree” are: “TARGET BEHAVIOUR is something I do frequently” and “TARGET BEHAVIOUR is something I do automatically”.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these questions: “What role does TARGET BEHAVIOUR play in your daily life?” or “How would you describe your engagement in the TARGET BEHAVIOUR in your daily living?”.

Expressions that demonstrate or imply the presence or absence of habitual behaviour, using words and phrases such as “automatically/deliberate”, “frequent/infrequent”, “spontaneous/planned”, and “hard not do/easy to cease” (with the first component of each pair of opposites indicating the possible presence of habitual behaviour).

In identified regulation a person endorses or identifies with the value or importance of a behaviour or health practice.

Use Likert scales that measure the degree to which engagement in the target activity is controlled by identified regulation.

The Situational Motivation Scale (SIMS) (Guay et al., 2000) is a frequently used self-report instrument to assess identified regulation of behaviour. Beside identified regulation of behaviour, the SIMS assesses intrinsic regulation, external regulation, and amotivation, and is used to assess motivation for various activities (e.g. physical exercise, academic performance).

The SIMS asks participants “Why are you currently engaged in this activity?”. Scale items represent potential answers to this question. Responses are given on a 7-point Likert scale ranging from “Does not correspond at all” to “Corresponds exactly”.

Examples of statements expressing identified regulation of behaviour are: “Because I am doing it for my own good”, “Because I think this activity is good for me”, and “Because I believe this activity is important for me”.

Instruments that measure the degree to which engagement in the target activity is controlled by identified regulation.

The Situational Motivation Scale (SIMS) (Guay et al., 2000) is a frequently used self-report instrument to assess identified regulation of behaviour. Beside identified regulation of behaviour, the SIMDS assesses intrinsic regulation, external regulation, and amotivation, and is used to assess motivation for various activities (e.g. physical exercise, academic performance).

The SIMS asks participants “Why are you currently engaged in this activity?”. Scale items represent potential answers to this question. Responses are given on a 7-point Likert scale ranging from “Does not correspond at all” to “Corresponds exactly”.

Examples of statements expressing identified regulation of behaviour are: “Because I am doing it for my own good”, “Because I think this activity is good for me”, and “Because I believe this activity is important for me”.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this question: “Why are you currently engaged in TARGET BEHAVIOUR?”

For example: “Why do you currently wear ear plugs at concerts?” or “Why do you currently follow a calorie-restricted diet?”

Expressions that demonstrate or imply that engagement in the target behaviour is controlled by identified regulation, using phrases such as “Because I am doing it for my own good” and “Because I believe this activity is important for me”.

Expressions that demonstrate or imply the presence of externally regulated behaviour should be coded as dct:extBehaviour_79n2r0sz.

Expressions that demonstrate or imply the presence of introjected regulation of behaviour should be coded as dct:introjBehaviour_79n2r0t0.

Expressions that demonstrate or imply the presence of integrated regulation of behaviour should be coded as dct:integratedBehaviour_79n2r0t0.

In practice, it can be difficult to distinguish expressions belonging to the four types of extrinsic motivation (external, introjected, identified, and integrated self-regulation). Therefore, a less precise categorization can be made between controlled and autonomous motivation. External and introjected regulation belonging to the first category, and identified and integrated self-regulation belonging to the first category. When this dichotomous categorization of extrinsic motivation is maintained, expressions typical of identified and integrated self-regulation can be grouped together.

Expressions that demonstrate or imply the presence of intrinsic regulation of behaviour should be coded as dct:intrinsicBehaviour_79n2r0t0.

Expressions that demonstrate or imply the absence of motivation to engage in the target behaviour (i.e. “amotivation”) should be coded as dct:motivation_79n2fh4q, and in a supplementary comment scored “0” or “absent”.

Expressions that refer to the presence of motivation with regard to the target behaviour without specifying the nature of its regulation or expressions that refer to the presence of motivation in general should be coded dct:motivation_79n2fh4q.

An impulsive action is defined as a non-deliberate action that serves the purpose of rendering one’s relation to the object, event, or state of the world more pleasant or less unpleasant.

To measure impulsive action in humans, several tests exist in which participants have to carry out an assignment or a series of assignments; among these tests the go-no go task, stop signal task (SST), Stroop task, and the continuous performance task (CPT) are most often utilized.

In addition, self-report instruments are used to measure impulsive choice and/or impulsive action, of which the Barratt Impulsiveness Scale (BIS-11) is most commonly used. The BIS-11 consists of 30 items, and answers on each item are measured on a 4-point scale (rarely/never, occasionally, often, almost always), with higher scores indicative of greater impulsivity. Examples of items used are: “I act on the spur of the moment” and " I say things without thinking”.

The aforementioned tests measure impulsive action as a general trait, not related to specific behaviours or objects. If one wants to assess impulsivity with regard to a particular behaviour or object, specific questions can be composed that measure the frequency impulses are experienced, and/or the strength of these impulses, in situations related to the target behaviour/object.

To measure impulsivity related to a specific behaviour, the stem “When it comes to TARGET BEHAVIOUR….” can be composed, followed by a series of phrases that express behaviour considered indicative for the presence or absence of impulsivity. For example: “When it comes to smoking cigarettes… 1) I do it without thinking about it; 2) I am self controlled; 3) I don’t manage to plan my behaviour 4) I act on the spur of the moment 5) I often suddenly abandon my behavioural intentions. Responses on these items can be given on a 5-point Likert scale ranging from “totally agree” to “totally disagree”.

Alternatively, a series of unique items can be composed to measure behaviour- or object-related impulsivity, such as: “I don’t wear ear plugs because when I decide to go to an event or concert it’s on the spur of the moment” (for target behaviour “wearing ear plugs when exposed to loud music”) and “When I hear a good track, I just have to turn off the volume restriction” (for target behaviour “use volume restriction on music players”) with responses “totally agree” and “totally disagree” as anchors on a 5-point Likert scale.

Impulsivity can be regarded as the converse of self-control. Therefore, the same instrument to assess self control (dct:self_control_79n2fh4t) can potentially be used to measure impulsivity. However, the scoring for the questionnaire items should be reversed (see dct:self_control_79n2fh4t for an example).

Questions or questionnaires that measure impulsivity related to a specific behaviour or object. For self-report asessment, a statement such as “When it comes to TARGET BEHAVIOUR….” can be composed, followed by a series of phrases that express behaviour considered indicative for the presence or absence of impulsivity. For example: “When it comes to smoking cigarettes… 1) I do it without thinking about it; 2) I am self controlled; 3) I don’t manage to plan my behaviour; 4) I act on the spur of the moment; 5) I often suddenly abandon my behavioural intentions; 6) I afterwards frequently regret that I smoked. Responses on these items can be given on a 5-point Likert scale ranging from “totally agree” to “totally disagree”.

Alternatively, unique items can be created to measure behaviour- or object-related impulsivity, such as: “When I hear a good track, I just have to turn off the volume restriction” (for target behaviour “use volume restriction on music players”) with responses “totally agree” and “totally disagree” as anchors on a 5-point Likert scale.

Beside behaviour- or object-related impulsivity, impulsivity as a trait can be measured. The Barratt Impulsiveness Scale (BIS-11) is the most commonly used self-report questionnaire for this pupose. Furthermore, numerous tests exist in which participants have to carry out an assignment or a series of assignments; among these tests the go-no go task, stop signal task (SST), Stroop task, and the continuous performance task (CPT) are most often utilized.

Impulsivity can be regarded as the converse of self-control. Therefore, the same instrument to assess self control (see dct:self_control_XXX) can potentially be used to measure impulsivity. However, the scoring for the questionnaire items should be reversed (see dct:self_control_XXX for an example).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these two questions (with preferably some questions in between): “When it comes to TARGET BEHAVIOUR how deliberate and controlled do you see your actions?” and “When it comes to TARGET BEHAVIOUR how spontaneous do you see your actions?”.

These same questions are used to elicite expressions indicative of self-control (see dct:self_control_XXX).

In addition, for the assessment of clinical levels of impulsivity in adults (e.g. in Attention Deficit Disorder), (semi) structured interview instruments exist such as the ACE+ (Young, 2016) and the MIDI (Grant, 2008).

Expressions that demonstrate or imply the presence or absence of impulsivity. In the context of behaviour change interventions (i.e. health behaviour changes), this impulsivity is regarded with respect to a specific behaviour or object. Trait impulsivity, i.e. impulsivity across situations, not specifically linked to a particular behaviour or object, is usually not the focus of interest. Therefore, expressions indicative of impulsivity should be coded as such only when they concern the behaviour or object of interest, unless beforehand another decision has been made. Expressions indicative of the presence of impulsivity may contain words such as “unplanned”, “undeliberate”, and “on the spur of the moment”.

Expressions indicative of self-control (e.g. ““I sticked to my plan, in spite of all these temptations”) should be coded as dct:self_control_79n2fh4t.

Expressions that describe situations or circumstances that evoke a desire to engage in the target behaviour (e.g. “When I see somebody drink a beer it is impossible to resist”) should be coded as dct:tempation_79n2fh4w.

Anticipated regret from not engaging in a behaviour.

Use 5-point Likert scales that measures the degree to which participants believe that not performing a particular behaviour will subsequently lead to regret, e.g. “If you would not use ear plugs when going clubbing, how much regret would you feel the next day?” or “If you would not use a condom when having sex with a one night stand, how much regret would you feel the next day?”, both items with anchors “None” (=0) and “A lot” (=4).

Operationalisations that measure the expectation to feel regret after not performing the target behaviour, e.g. “If you would not use ear plugs when going clubbing, how much regret would you feel the next day?” or “If you would not use a condom when having sex with a one night stand, how much regret would you feel the next day?”, both items with anchors “None” (=0) and “A lot” (=4).

Anticipated regret from engaging in the target behaviour is action regret (dct:action_regret_79n2w1bj).

For the composite of action and inaction regret, either as incentive towards a behaviour (i.e. as sumscore) or the total anticipated regret experienced, see respectively anticip_regret_79n2w1bh and anticip_regret_79n2w1bj.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these questions: ‘What do you expect to feel if you would not (engage in) TARGET BEHAVIOUR?’, ‘What emotions do do you expect to experience if you would not (engage in) TARGET BEHAVIOUR?’

Expressions that demonstrate or imply that one expects to experience feelings of regret if one would not engage in the target behaviour, i.e. usings verbs such ‘remorse’, ‘sorry’, and ‘lament’.

Manifestations that express feelings of regret if one would engage in the target behaviour,should be coded dct:action_regret_79n2w1bj.

Biological influences consist of an individual’s genotypic and phenotypic characteristics (i.e., respectively, an organism’s hereditary information and its actual observed properties) that determine, or may potentially determine, its behaviour.

Depending on the relevance for the research in question, a list of potentially relevant biological influences can be composed. Examples of frequently assessed biological influences are sex, height, and presence of (hereditary) diseases. The biological influences deemed of importance are assessed in different ways: if the number of potential responses on a question is limited, fixed answer options should be offered. When answers are aspected to be diverse, or participants may need to elaborate, open answer options is the format of preference.

Questions or questionnaires that measure biological factors (e.g. sex or height) with a potential influence on the behaviour of interest.

In behaviour change research, biological influences are most often regarded to exert a “distal” effect on behaviour, i.e. exerting influence on behaviour through other variables with a more direct, “proximal”, effect. As these latter variables have a greater potential for change, the focus in questionnaires is mostly on these variables, and biological influences are considered background factors, measured with a few questions, most often at the beginning of the questionnaire.

Note that certain variables, such as age and sex, may be coded as both a biological and demographic influence (i.e. dct:inflDemographic_79n2w1bk). Whether classification as the former or latter is more appropriate depends on the focus of the research, e.g. if the impact of legislation on alcohol consumption in adolescents is studied, classifying “age” as a demographic influence rather than a biological influence is appropriate.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this question: ‘Please list any biological factors that may have an influence on your engagement in the TARGET BEHAVIOUR’. For example: “Please list any biological influences that may have an influence on your consumption of alcohol”.

The aformentioned question may to difficult for some target groups and/or participants may not be able to consider, or are not aware of, all possible biological influences. In those cases a more ‘narrow’ question zooming in on a particular biological influence (e.g. hereditary addiction sensitivity) may be prefered, e.g.: “Do you have or did you have any family members with dependency or addiction problems?”.

See also dct:inflDemographic_79n2w1bk for more examples of questions that may be used to elicit contruct content for biological influences.

Expressions that demonstrate or imply the presence or absence of biological influences on engagement in the target behaviour. For example, expressions such as “Although we don’t eat much, obesity is a problem in my family” and “My father and uncle had a drinking problem as well” may indicate that biological factors exert an influence on the (problem) behaviour of interest.

Note that certain expressions, such as those refering to age and sex, may be coded as both a biological and demographic (i.e. dct:inflDemographic_79n2w1bk) influence. Whether classification as the former or latter is more appropriate depends on the context the expression is made in, e.g. classifying “age” as a demographic influence rather than a biological influence in the expression “I don’t drink alcohol because I cannot buy it, as I am underage” is appropriate.

The influence of an individual’s broad characteristics (e.g. age, marital status, nationality) as determined in demographic research and/or known to public institutions on the behaviour of interest.

Depending on the relevance for the research in question, a list of potentially relevant demographic influences can be composed. Examples of frequently assessed demographic influences are age, marital status, income, religion, and education. The demographic influences deemed of importance are assessed in different ways: if the number of potential responses on a question is limited, fixed answer options should be offered. When answers are aspected to be very diverse, or participants may need to elaborate, open answer options is the format of preference.

Questions or questionnaires that measure demographic factors (e.g. income, education, marital/partner status) with a potential influence on the behaviour of interest.

In behaviour change research, demographic influences are most often regarded to exert a “distal” effect on behaviour, i.e. exerting influence on behaviour through other variables with a more direct, “proximal”, effect. As these latter variables have a greater potential for change, the focus in questionnaires is mostly on these variables, and demographic influences are considered background factors, measured with a few questions, most often at the beginning of the questionnaire.

Note that certain variables, such as age and gender, may be coded as both a biological (i.e. dct:inflBiological_79n2w1bj) and demographic influence. Whether classification as the former or latter is more appropriate depends on the focus of the research, e.g. if the impact of legislation on alcohol consumption in adolescents is studied, classifying “age” as a demographic influence rather than a biological influence is appropriate.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, one of two different approaches can be followed:

1. The participant is encouraged to name demographic factors that he/she perceives to have an influence on his/her engagement in the target behaviour.

2. a) The participant is asked if he/she perceives a particular demographic factor to have an influence on his/her engagement in the target behaviour. This question can be repeated for other demographic factors the researcher thinks to be relevant.

   b) The participant is asked to name the factors out of a series of demographic factors that he/she perceives to have an influence on his/her engagement in the target behaviour.

When the first approach is chosen, use the following question ‘Please list any demographic factors that may have an influence on your engagement in the TARGET BEHAVIOUR’.

One should consider that the concept of “demographic influences” for most study populations may be too difficult to understand, or the range of demographic influences considered may be too narrow. In that case, the researcher can adopt the second approach and either:

a) ask if the participant perceives a particular demographic factor to have an influence on his/her engagement in the target behaviour, e.g.“Do you think that being single has an influence on your alcohol consumption?”. Possibly followed by the same question for other demographic factors, e.g. “Do you think that being unemployed has an influence on your alcohol consumption?”. or b) name a number of demographic influences, using the following introduction and question: “I will name a couple of so-called demographic factors: [name demograhic variables of interest]”. “Please list those factors that you think have an influence on your engagement in the TARGET BEHAVIOUR”. For example: “I will name a couple of so-called demographic factors: partner status, employment status, nationality, and religion. Please list those factors that you think have an influence on your alcohol consumption.”

Any of the aforementioned questions is followed by “Please explain” if the participant names a demographic factor without giving an explanation for this particular response.

Expressions that demonstrate or imply the presence or absence of demographic influences on the engagement in the target behaviour. For example, expressions such as “I don’t drink alcohol because I am a muslim” or “Eating healthy for me is difficult as I am unemployed”.

Note that certain expressions, such as those refering to age and gender, may be coded as both a biological (i.e. dct:inflBiological_79n2w1bj) and demographic influence. Whether classification as the former or latter is more appropriate depends on the context the expression is made in, e.g. classifying “age” as a demographic influence rather than a biological influence in the expression “I don’t drink alcohol because I cannot buy it, as I am underage” is appropriate.

Aspects of a person’s physical and social surroundings that discourage or encourage the behaviour of interest.

Depending on the relevance for the research in question, a list of potentially relevant environmental influences can be composed. Examples of frequently assessed environmental influences are the living environment (e.g. the neighboorhood), weather/climate, health care, public transport, sport and recreation facilities, legislation, and politics.

The environmental influences deemed of importance are assessed in different ways: if the number of potential responses on a question is limited, fixed answer options should be offered. When answers are aspected to be very diverse, or participants may need to elaborate, open answer options is the format of preference.

Questions or questionnaires that measure environmental factors (e.g. health care, public transport) with a potential influence on the behaviour of interest.

In behaviour change research, environmental influences are most often regarded to exert a “distal” effect on behaviour, i.e. exerting influence on behaviour through other variables with a more direct, “proximal”, effect. As these latter variables have a greater potential for change, the focus in questionnaires is mostly on these variables, and environmental influences are considered background factors, measured with a few questions, most often at the beginning of the questionnaire.

Note that certain variables may be coded as both a environmental and demographic influence (dct:inflDemographic_79n2w1bk). Whether classification as the former or latter is more appropriate depends on the focus of the research, e.g. if the impact of green neigboorhood spaces on physical activity is studied, it is appropraite to classify living in a deprived neighboorhood with limited spaces for leisure time activities as an environmental influence rather than a demographic influence.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, one of two different approaches can be followed:

1. The participant is encouraged to name environmental factors that he/she perceives to have an influence on his/her engagement in the target behaviour.

2. a) The participant is asked if he/she perceives a particular environmental factors factor to have an influence on his/her engagement in the target behaviour. This question can be repeated for other environmental factors the researcher thinks to be relevant.

   b) The participant is asked to name the factors out of a series of environmental factors that he/she perceives to have an influence on his/her engagement in the target behaviour.

When the first approach is chosen, use the following question ‘Please list any environmental factors that may have an influence on your engagement in the TARGET BEHAVIOUR’.

One should consider that the concept of “environmental influences” for most study populations may be too difficult to understand, or the range of environmental influences considered may be too narrow. In that case, the researcher can adopt the second approach and either:

a) ask if the participant perceives a particular environmental factor to have an influence on his/her engagement in the target behaviour, e.g.“Do you think that the number of parks in your neighboorhood has an influence on your obesity?”. Possibly followed by the same question for other environmental factors, e.g. “Do you think that the number of fitness centres in your neighboorhood has an influence on your obesity?”. or b) name a number of environmental influences, using the following introduction and question: “I will name a couple of so-called environmental factors: [name environmental variables of interest]”. “Please list those factors that you think have an influence on your engagement in the TARGET BEHAVIOUR”. For example: “I will name a couple of so-called environmental factors: the presence of parks, the presence of fitness centres, the presence of bicycle lanes, and the presence of food markets. Please list those factors that you think have an influence on your obesity.”

Any of the aforementioned questions is followed by “Please explain” if the participant names an environmental factor without giving an explanation for this particular response.

Expressions that demonstrate or imply the presence or absence of environmental influences on the engagement in the target behaviour. For example, expressions such as “I don not do any sports because there are no sport clubs in my neighboorhood” and “I would like to travel by bike, but it is too dangerous to ride a bike in my neighboorhood”.

Note that certain expressions may be coded as both an environmental and demographic influence (dct:inflDemographic_79n2w1bk). Whether classification as the former or latter is more appropriate depends on the context the expression is made in, e.g. classifying “neighboorhood” as an environmental influence rather than a demographic influence in the expression “I don’t engage in outdoor physical activity, as in my neighboorhood there are no spaces for leisure time activities” is appropriate.

The sum of the injunctive norm beliefs.

A composite of injunctive norms beliefs is obtained by multiplying the perceived approval or disapproval of a normative referent (dct:referentApproval_73dnt5zd) by the motivation to comply with that referent (dct:motivationToComply_73dnt5zf) and then summing the product terms across all salient injunctive norms beliefs. This, preferably, is represented graphically (e.g. in a table or index) and, additionally, in an equation. See Table 4.1 on p. 136 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) for an example.

The injunctive norm beliefs composite reflects the summation of injunctive norm beliefs with regard to the target behaviour. Each injunctive norm belief (dct:injNorms_belief_73dnt5zg), in its turn, is the product term of the multiplication of the perceived referent approval (dct:referentApproval_73dnt5zd) by the motivation to comply with that referent (dct:motivationToComply_73dnt5zf).

Given the formative nature of the injunctive norm beliefs composite, for a swift and clear overview of its composition and its effect on injunctive norms (dct:injunctiveNorms_73dnt5zj), preferably a graphical representation (e.g. a table or index) and, possibly, an equation is consulted. Table 4.1 on p. 136 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) represents an example of an overview of a injunctive norm beliefs composite.

Injunctive norm beliefs and, as a consequence, also injunctive norm beliefs composites (which are the sum of the former), can not be directly elicited. Injunctive norm beliefs are the product of the perceived referent approval of the target behaviour (dct:referentApproval_73dnt5zd) multiplied by the motivation to comply with that referent (dct:motivationToComply_73dnt5zf).

See these two subconstructs for instructions for eliciting their respective construct content.

Injunctive norm beliefs and, as a consequence, also injunctive norm beliefs composites (which are the sum of the former), can not be directly elicited. Injunctive norm beliefs are the product of the perceived referent approval of the target behaviour (dct:referentApproval_73dnt5zd) multiplied by the motivation to comply with that referent (dct:motivationToComply_73dnt5zf).

See these two subconstructs for instructions for coding their respective construct content.

A person’s belief what should or ought to be done with respect to performing a particular behaviour.

The measurement of injunctive norm beliefs starts with an elicitation study to identify the content of injunctive norm beliefs about the behaviour which are shared by the target population.

To obtain quantitative data regarding injunctive norm beliefs, the participant is subsequently asked to respond to two questions with respect to each injunctive norm belief, one to assess the perceived approval or disapproval of a social referent (see dct:referentApproval_73dnt5zd) and the other to measure the motivation to comply with that referent (see dct:motivationToComply_73dnt5zf).

Finally, to get a total quantitative score of a descriptive norm belief, the perceived behaviour of a social referent is multiplied by the identification with that social referent. The total score of a injunctive norm belief enables the assessment of the importance of the contribution of an injunctive norm belief, in comparison with other injunctive norm beliefs, to the injunctive norm beliefs composite (see dct:injNorms_belCom_73dnt5zh).

Each injunctive norm belief is the product term of the multiplication of the perceived approval or disapproval of a given normative referent (see dct:referentApproval_73dnt5zd) by the motivation to comply with that referent (see dct:motivationToComply_73dnt5zf).

Therefore, whether an injunctive norm belief exerts influence (and if so, to what extent) on the injunctivenorm beliefs composite (see dct:injNorms_belCom_73dnt5zh), and whether a particular injunctive norm belief lowers or increases the perception that salient injunctive referents think one should perform the behaviour in question, can only be deduced by concurrent consultation of the two subcomponents of which it is constructed. Table 4.1 on p. 136 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) provides an example how the two subcomponents exert an influence on the formation of individual injunctive normative beliefs.

Injunctive norm beliefs can not be directly elicited. They are the product of the perceived referent approval of the target behaviour (dct:referentApproval_73dnt5zd) multiplied by the motivation to comply with that referent (dct:motivationToComply_73dnt5zf).

See these two subconstructs for instructions for eliciting their respective construct content.

Injunctive norm beliefs can not be directly elicited. They are the product of the perceived referent approval of the target behaviour (dct:referentApproval_73dnt5zd) multiplied by the motivation to comply with that referent (dct:motivationToComply_73dnt5zf).

See these two subconstructs for instructions for coding their respective construct content.

Injunctive norms refer to perceptions concerning what should or ought to be done with respect to performing a given behaviour. People may be motivated to behave in accordance with what they believe others think they should do (injunctive norms) because the social agent has the power to reward or punish them, because the agent has the right to request it, because the agent is an expert, or because they want to be like the agent. The operation of reward, coercive, and legitimate power all imply that the social agent expects a particular kind of behavior; that is, it implies an injunctive norm.

Use likert scales that measure perceived social approval of important social referents in general. It is important that no specific individuals are referenced, but that the items do refer to individuals that are similar, important or seen as important to follow. Note that the behaviour in these items must always be the target behavior. The example items in the 2010 RAA book are ‘Most people who are important to me think I should TARGET BEHAVIOUR.’ with anchors ‘False’ vs ‘True’; and ‘Most people whose opinions I value would … of my TARGET BEHAVIOUR’ with anchors ‘Disapprove’ vs ‘Approve’.

Questions or questionnaires that measure perceptions concerning what should or ought to be done with respect to performing the target behaviour. No specific individuals are referenced, but items refer to individuals that are similar, important or seen as important to follow. Examples of questions measuring injunctive norms (on a 7-point Likert scale) are: ‘Most people who are important to me think I should TARGET BEHAVIOuR.’ with anchors ‘Agree’ vs ‘Disagree’; and ‘Most people whose opinions I value would … of my TARGET BEHAVIOUR’ with anchors ‘Disapprove’ vs ‘Approve’.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, ask these questions: ‘Would persons important to you approve or think you should TARGET BEHAVIOUR’; and ‘Would persons important to you disapprove or think you should not TARGET BEHAVIOUR?’

Reference here is made to a ‘generalized agent’ (i.e. important social referents in general) and no specific individuals are referenced. As reference to a specific individual is often unavoidable, one may opt not to elicit construct content with regard to injunctive norms, but limit the elicitation of construct content to the subconstructs dct:motivationToComply_73dnt5zf and dct:referentApproval_73dnt5zd

Expressions of perceived approval or disapproval of the target behaviour by others.

Note that expressions of perceived performance of the target behaviour by others should be coded as dct:descriptiveNorms_73dnt5zp, and expressions of perceived norms where it is unclear whether they concern perceived (dis)approval or perceived approval should be coded as dct:perceivedNorm_71w98kk2.

The sum of the instrumental attitude beliefs.

A composite of instrumental attitude beliefs is obtained by multiplying the expectations of each instrumental attitude belief (see dct:instrAttitude_expectation_73dnt5z6) by its evaluation (see dct:instrAttitude_evaluation_73dnt5z7) and then summing the product terms across all instrumental attitude beliefs. This, preferably, is represented graphically (e.g. in a table or index) and, additionally, in an equation.

The authors state on p.104 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) that the composite measure of beliefs is assumed to determine the (instrumental) attitude, and should not be interpreted as an indirect measure of (instrumental) attitude. A low correlation between the beliefs composite and a direct measure of instrumental attitude may be an indication that the expectancy-value index (i.e. the attitude beliefs composite) is based on an inappropriate set of (nonsalient) beliefs, or that the measure of belief strength may be scored inappropriately.

Questions or questionnaires that measure the expectations (dct:instrAttitude_expectation_73dnt5z6) and evaluations (dct:instrAttitude_evaluation_73dnt5z7) of all salient instrumental attitude beliefs with regard to the target behaviour.

Instrumental attitude belief expectations combine multiplicatively with instrumental attitude belief evaluations into instrumental attitude beliefs. The product terms for each belief and the sum of the products terms of all salients beliefs are represented in the expectancy-value index (i.e. the compostite of attitude beliefs) for the target behaviour. See Table 3.2 on p. 98 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) for an example of such an index.

Instrumental attitude beliefs and, as a consequence, also instrumental attitude beliefs composites (which are the sum of the former), can not be directly elicited. Instrumental attitude beliefs are the product of the product of the expected instrumental consequences of a target behaviour (instrAttitude_expectation_73dnt5z6) multiplied by the evaluation of those consequences (instrAttitude_evaluation_73dnt5z7).

See these two subconstructs for instructions for eliciting their respective construct content.

Instrumental attitude beliefs and, as a consequence, also instrumental attitude beliefs composites (which are the sum of the former), can not be directly elicited. Instrumental attitude beliefs are the product of the product of the expected instrumental consequences of a target behaviour (instrAttitude_expectation_73dnt5z6) multiplied by the evaluation of those consequences (instrAttitude_evaluation_73dnt5z7).

See these two subconstructs for instructions for coding their respective construct content.

Beliefs are defined as the subjective probability that an object has a certain attribute, where ‘object’ and ‘attribute’ are used in the generic sense, and refer to any discriminable aspect of an individual’s world. Conform the expectancy-value model of attitude, beliefs are the product of the expectation that engaging in a behavior will have a given consequence and the evaluation of that consequence as (very) negative or (very) positive. Instrumental attitude beliefs concern beliefs about consequences of engaging in the behavior for achieving one’s goals.

Instrumental attitude beliefs are measured by measuring both of the component constructs, i.e.: dct:instrAttitude_evaluation_73dnt5z7 and dct:instrAttitude_expectation_73dnt5z6.

Each instrumental attitude belief is the product term of the multiplication of the evaluation of the perceived instrumentall consequences of the target behaviour (dct:instrAttitude_evaluation_73dnt5z7) by the perceived likelihood that engaging in the behaviour will actually have these consequences (dct:instrAttitude_expectation_73dnt5z6).

Therefore, whether an instrumental attitude belief exerts influence (and if so, to what extent) on the instrumental attitude beliefs composite (dct:instrAttitude_belCom_73dnt5z9), and whether a particular instrumental attitude belief leads to a more favourable or more unfavourable instrumental attitude towards performing the target behaviour, can only be deduced by concurrent consultation of the two subcomponents of which it is constructed. Table 3.2 on p. 98 of the Reasoned Action Approach book (Ajzen & Fishbein, 2010) illustrates how the two subcomponents can exert an influence on individual attitude beliefs.

Instrumental attitude beliefs can not be directly elicited. They are the product of the expected instrumental consequences of a target behaviour (instrAttitude_expectation_73dnt5z6) multiplied by the evaluation of those consequences (instrAttitude_evaluation_73dnt5z7).

See these two subconstructs for instructions for eliciting their respective construct content.

Instrumental attitude beliefs can not be directly elicited. They are the product of the expected instrumental consequences of a target behaviour (instrAttitude_expectation_73dnt5z6) multiplied by the evaluation of those consequences (instrAttitude_evaluation_73dnt5z7).

See these two subconstructs for instructions for coding their respective construct content.

An instrumental attitude belief evaluation is the evaluation as positive or negative (i.e. desirable or undesirable) of one specific instrumental potential consequence of a behavior. The instrumental nature of this evaluation means that this these evaluations contain information about people’s longer term goals. Consequences that concern experiences and sensations, such as pleasure or pain, are captured in experiential attitude and the underlying beliefs (see dct:instrAttitude_expectation_73dnt5z6 and dct:expAttitude_expectation_73dnt5z1 for more details about this distinction).

According to the Reasoned Action Approach, instrumental attitude belief evaluations combine multiplicatively with instrumental attitude belief expectations (see dct:instrAttitude_expectation_73dnt5z6) into instrumental attitude beliefs (see dct:instrAttitude_belief_73dnt5z8).

Examples of instrumental attitude belief evaluations are (note that the examples provided for the instrumental attitude belief expectations should probably be read first, see dct:instrAttitude_expectation_73dnt5z6): the degree to which one evaluates the expectation that one saves time by taking the bus to the city center as positive; the extent to which one considers it desirable that exercising will contribute to one’s goal of being healthy (assuming one holds that expectation, which is captured by dct:instrAttitude_expectation_73dnt5z6); whether somebody likes to prevent starving, a goal that eating a pizza contributes to; one’s evaluation of burning calories (an expectation one may have of the behavior going for a run); and whether somebody wants to produce offspring.

Instrumental attitude belief evaluations are measured using questions in a questionnaire that are referred to as ‘items’. Instrumental attitude belief evaluation always relate to specific instrumental attitude belief expectations, so the first step is to identify the expectation for which the evaluation should be measured (e.g. ‘Drinking four cups of coffee every day leads to me being … [much less healthy|much more healthy]’ or ‘Drinking four cups of coffee makes me … [very idle|very productive]'). Evaluations of expectations are always measured on a bidimensional scale.

As item stem, use ‘I prefer …’, and as anchors, include the scale extremes used when measuring the expectation. For example, an item could be ‘I prefer … [being much less healthy|being much more healthy]’, and another items can be ‘I prefer … [being very idle|being very productive]’. Make sure to formulate the item stem such that it makes clear that you are asking people about their evaluation (not what they think holds more generally). Use a seven-point response scale and try to always be consistent in the scale valence; in languages that are read from left to right, always place the most passive/low/less/weak/unlikely scale extreme (anchor) on the left, and the most active/high/more/strong/likely scale extreme (anchor) on the right. Do not reverse this order for one or more items.

The item or set of items should be accompanied by an instruction that makes clear that you are asking people about their evaluation (not what they think holds more generally).

Instrumental attitude belief evaluations are measured using questions in a questionnaire (these are referred to as ‘items’). Items can be coded as measuring an instrumental attitude belief evaluation if they measure participants’ evaluation of one specific instrumental potential consequence of a behavior as positive or negative (i.e. desirable or undesirable).

The instrumental nature of this evaluation means that this these evaluations contain information about people’s longer term goals.

The question should include, either in its stem or in its anchors, the specific dimension of which the evaluation is being measured (e.g. whether people prefer being unhealthy versus healthy, or whether people evaluate being healthy as positive or negative).

Note that evaluations of consequences that concern experiences and sensations, such as pleasure or pain, are captured in experiential attitude and the underlying beliefs (see dct:instrAttitude_expectation_73dnt5z6 and dct:expAttitude_expectation_73dnt5z1 for more details about this distinction, and see [[dct::expAttitude_evaluation_73dnt5z2]] for the coding instruction for experiential attitude belief evaluations).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources (i.e. transcripts or notes) are then coded using the instruction for aspect coding. In this qualitative study, use these questions: ‘What do you see as the advantages of you engaging in [target behavior]?’, ‘What do you see as the disadvantages of you engaging in [target behavior]?’, and ‘What else comes to mind when you think about [target behavior]?’.

Expressions of one’s evaluations as positive or negative (i.e. desirable or undesirable) of instrumental potential consequences of a behavior. The instrumental nature of this evaluation means that this these evaluations contain information about people’s longer term goals.

Note that expressions of evaluations of consequences that concern experiences and sensations, such as pleasure or pain, are captured in experiential attitude and the underlying beliefs (see dct:expAttitude_evaluation_73dnt5z2).

One’s actual expectation (i.e. whether a consequence is likely or unlikely) should be coded as dct:instrAttitude_expectation_73dnt5z6.

An instrumental attitude belief expectation is the expectation of how probable (i.e. unlikely versus likely) it is that engaging in the target behavior will cause one specific instrumental potential consequence to come about. The instrumental nature of this consequence denotes that instrumental attitude belief expectations must concern consequences that facilitate or hinder achieving one or more longer term goals.

Each instrumental attitude belief expectation covers one specific long-term consequence, disconnected from acute hedonic consequences the target behavior may have. Instrumental attitude belief expectations always refer to consequences that cause a behavior to contribute more or less to goals an individual has. Expectations about experiencing the consequences of a behavior, for example expectations relating to experiences and sensations, are captured in experiential attitude belief expectations (see dct:expAttitude_expectation_73dnt5z1).

According to the Reasoned Action Approach, instrumental attitude belief expectations combine multiplicatively with instrumental attitude belief evaluations (see dct:instrAttitude_evaluation_73dnt5z7) into instrumental attitude beliefs (see dct:instrAttitude_belief_73dnt5z8).

Examples of instrumental attitude belief expectations are: the expectation that taking the bus to the city centre will save time, which contributes to the goal of being punctual; the expectation that walking to the city centre will allow one to exercise, which contributes to the goal of being healthy; the expectation that eating a pizza will prevent one from starving, which contributes to the goal of surviving; the expectation that going for a run will burn calories, which contributes to one’s goal of losing weight; and the expectation that having sex without a condom is more likely to lead to conception, which is required to achieve one’s goal of producing offspring.

Instrumental attitude belief expectations are measured using questions in a questionnaire that are referred to as ‘items’. To measure an instrumental attitude belief expectation, first, identify exactly which potential instrumental consequence of the target behavior (e.g. a specific benefit or advantage) you want to measure. Then, establish whether accurately describing the spectrum of possibilities regarding this instrumental consequence requires a unidimensional scale or a bidimensional scale. Most potential consequences of a target behavior can be perceived either to increase or decrease upon performance of the target behavior, requiring a bidimensional scale. For example, some people might expect that drinking four cups of coffee every day contributes to their goal of being healthy, whether others might expect that that is unhealthy instead, expecting that not consuming any coffee at all is more healthy. However, in rare cases, one of the two dimensions can be excluded a priori, in which case a unidimensional scale suffices. For example, the degree to which people will expect that exercising regularly for a month will increase their health may vary; but is it excessively unlikely that somebody might expect that if they exercise regularly for a month, their health will decrease. Therefore, in that case, one might want to choose a unidimensional scale. In general, a rule of thumb is that if the ‘default state’ of this instrumental consequence resembles absence of the potential consequence, and therefore, engaging in the target behavior can only have an effect in one direction, a unidimensional scale can be used. However, when engaging in the target behavior can conceivably increase or decrease this instrumental consequence, a bidimensional scale is required.

Once it is clear whether a bidimensional or unidimensional scale should be used, the construction of the item stem and the two anchors can start. Make sure to formulate the item stem such that it makes clear that you are asking people about their expectation (not what they think holds more generally).

For unidimensional scales, create an item stem that explicitly lists the single dimension that expresses the instrumental consequence (e.g. exercising regularly for a month leading to increased health). As anchors, always use ‘Very unlikely’ and ‘Very likely’. An example item would be: ‘If I exercise regularly for a month, it is … that my health will increase. [Very unlikely|Very likely]’. However, as explained above, unidimensional scales can rarely be used in most circumstances, so bidimensional scales are usually required.

For bidimensional scales, the item stem cannot explicitly list only one of the two dimensions (e.g. ‘become much less healthy’ or ‘become much more healthy’). This is because this would create a unidimensional or ambiguous response scale. People who would score low on the unidimensional scale might mean either that they don’t think that consequence will occur, or that they think that the opposite consequence will occur. For example, when creating an item ‘If I drink four cups of coffee every day, I will become much more healthy. [Very unlikely|Very likely]’, people who respond ‘Very unlikely’ can mean either that they expect to feel much less healthy, or that they expect that drinking four cups of coffee every day will have no effect on their health.

Therefore, capturing the full potential breadth of the beliefs of your target population requires asking what they expect exactly. To do this, create an item stem that contains the target behavior, and anchors that express the extremes of the bidimensional scale. For example, ‘Drinking four cups of coffee every day leads to me being … [much less healthy|much more healthy]’. Sometimes, the two extremes of the dimension you want to measure can be expressed in two antonyms, such as ‘Drinking four cups of coffee makes me … [very idle|very productive]’.

Once the item stem and the two anchors have been determined, decide which response scale to use. For bidimensional scales, seven-point scales are preferred, as these leave three degrees of expression in each dimension (the mid-point representing the expectation that the behavior does not have a consequence regarding this specific instrumental attitude belief). For unidimensional scales, five-point scales suffice. Try to always be consistent in the scale valence; in languages that are read from left to right, always place the most passive/low/less/weak/unlikely scale extreme (anchor) on the left, and the most active/high/more/strong/likely scale extreme (anchor) on the right. Do not reverse this order for one or more items.

When combining multiple items in one measurement instrument, if both evaluations of unidimensional consequences and evaluations of bidimensional consequences are measured, either use two matrices or combine them in one that uses seven-point scales for all items.

The item or set of items should be accompanied by an instruction that makes clear that you are asking people about their expectation (not what they think holds more generally).

Instrumental attitude belief expectations are measured using questions in a questionnaire (these are referred to as ‘items’). Items can be coded as measuring an instrumental attitude belief expectation if they measure participants’ expectation of how probable (i.e. unlikely versus likely) it is that engaging in the target behavior will cause one specific instrumental potential consequence to come about. The instrumental nature of this consequence denotes that instrumental attitude belief expectations must concern consequences that facilitate or hinder achieving one or more longer term goals.

These instrumental attitude belief expectations cover long-term consequences, disconnected from acute hedonic consequences the target behavior may have. Instrumental attitude belief expectations always refer to consequences that cause a behavior to contribute more or less to goals an individual has.

The question should include, either in its stem or in its anchors, the specific expectation being measured.

Note that items concerning experiencing the consequences of a behavior, for example expectations relating to experiences and sensations, are captured in experiential attitude belief expectations (see dct:expAttitude_expectation_73dnt5z1).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources (i.e. transcripts or notes) are then coded using the instruction for aspect coding. In this qualitative study, use these questions: ‘What do you see as the advantages of you engaging in [target behavior]?’, ‘What do you see as the disadvantages of you engaging in [target behavior]?’, and ‘What else comes to mind when you think about [target behavior]?’.

Expressions of expectations of how probable (i.e. unlikely versus likely) it is that engaging in the target behavior will cause an instrumental potential consequence to come about. The instrumental nature of this expectation means that these expected consequences must concern facilitation or obstruction of achieving one or more longer term goals. These instrumental attitude belief expectations cover expectations of potential long-term consequences, disconnected from acute hedonic consequences the target behavior may have. Instrumental attitude belief expectations always refer to assumed benefits or costs of a behavior.

Note that expressions of expectations of consequences that render the target behavior less or more desirable where the immediate experiential effects of the behavior play a role are captured in experiential attitude belief expectations (see dct:expAttitude_expectation_73dnt5z1).

One’s evaluation in terms of valence (i.e. positive versus negative) of the consequence should be coded as dct:instrAttitude_evaluation_73dnt5z7.

A latent disposition or tendency to respond with some degree of favorableness or unfavorableness to the target behavior based on the perceived usefulness of engaging in that target behavior to achieving one’s goals.

Use semantic differentials with root ‘For me, TARGET BEHAVIOR is …’ and a bidimensional scale where the right-most anchor expresses a generally desirable instrumental state/goal and the left-most anchor expresses the opposite undesirable state/goal (e.g. ‘unwise’ versus ‘wise’ and ‘bad’ versus ‘good’). The example items in the 2010 RAA book are ‘For me, TARGET BEHAVIOR is …’ with anchors ‘Bad’ vs ‘Good’; ‘For me, TARGET BEHAVIOR is …’ with anchors ‘Unpleasant’ vs ‘Pleasant’; ‘For me, TARGET BEHAVIOR is …’ with anchors ‘Harmful’ vs ‘Beneficial’; and ‘For me, TARGET BEHAVIOR is …’ with anchors ‘Boring’ vs ‘Interesting’.

Operationalisations that measure aspects of the latent disposition or tendency to respond favourably versus unfavourably to the target behavior based on the perceived usefulness of engaging in that target behavior to achieving one’s goals, for example using the semantic differentials ‘wise’ vs ‘unwise’ or ‘good’ vs ‘bad’.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these three questions: ‘What do you see as the advantages of you engaging in TARGET BEHAVIOUR?’, ‘What do you see as the disadvantages of you engaging in TARGET BEHAVIOUR?’, and ‘What else comes to mind when you think about engaging in TARGET BEHAVIOUR?’.

Expressions that demonstrate or imply aspects of the latent disposition or tendency to respond favourably versus unfavourably to the target behaviour based on the perceived usefulness of engaging in the target behaviour to achieving one’s goals, for example using the semantic differentials ‘wise’/‘unwise’ or ‘good’/‘bad’.

Manifestations that express affective aspects of favourableness or unfavourableness to the target behaviour should be coded dct:experientialAttitude_73dnt5z5.

In integrated regulation a person not only values a behaviour, but has also aligned it with other central values and lifestyle patterns.

Use Likert scales that measure the degree to which participants’ engagement in the target activity is determined by integrated regulation. Examples of statements expressing integrated regulation of behaviour are: “It is essential to my identity and sense of self”, “It is genuinely part of me”, “It is consistent with my values, goals and aims in life”, and “Doing TARGET BEHAVIOUR and being myself are inseparable”. Responses on these items are given on a 4-point scale ranging from “Not true at all” to “Very true”. These sample items are form the Integrated regulation (INTEG) scale developed by McLachlan, Spray, and Hagger (2011) which assesses integrated regulation in exercise behaviour.

The frequently used Situational Motivation Scale (SIMS) (Guay et al., 2000) does not measure integrated (nor introjected) regulation. The SIMS assesses self-reported intrinsic regulation, identified regulation, external regulation, and amotivation, and is used to assess motivation for various activities (e.g. physical exercise, academic performance).

In practice, it is difficult to distinguish the four types of extrinsic motivation (external, introjected, identified, and integrated self-regulation) using self-report instruments (i.e. insufficient discriminant validity). Therefore, a less precise categorization can be made between controlled and autonomous motivation. External and introjected regulation belonging to the first category, and identified and integrated self-regulation belonging to the first category. When this dichotomous categorization of extrinsic motivation is maintained, items to assess identified and integrated self-regulation can be grouped together.

Instruments that measure the degree to which participants’ engagement in the target activity is determined by integrated regulation. Examples of statements expressing integrated regulation of behaviour are: “It is essential to my identity and sense of self”, “It is genuinely part of me”, “It is consistent with my values, goals and aims in life”, and “Doing TARGET BEHAVIOUR and being myself are inseparable”. Responses on these items are given on a 4-point scale ranging from “Not true at all” to “Very true”. These sample items are form the Integrated regulation (INTEG) scale developed by McLachlan, Spray, and Hagger (2011) which assesses integrated regulation in exercise behaviour.

The frequently used Situational Motivation Scale (SIMS) (Guay et al., 2000) does not measure integrated (nor introjected) regulation. The SIMS assesses self-reported intrinsic regulation, identified regulation, external regulation, and amotivation, and is used to assess motivation for various activities (e.g. physical exercise, academic performance).

In practice, it is difficult to distinguish the four types of extrinsic motivation (external, introjected, identified, and integrated self-regulation) using self-report instruments (i.e. insufficient discriminant validity). Therefore, a less precise categorization can be made between controlled and autonomous motivation. External and introjected regulation belonging to the first category, and identified and integrated self-regulation belonging to the first category. When this dichotomous categorization of extrinsic motivation is maintained, items to assess identified and integrated self-regulation can be grouped together.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this question: “Why are you currently engaged in TARGET BEHAVIOUR?”

For example: “Why do you currently wear ear plugs at concerts?” or “Why do you currently follow a calorie-restricted diet?”

Expressions that demonstrate or imply that engagement in the target behaviour is determined by integrated regulation, using phrases such as “It is genuinely part of me”, “It is consistent with my values, goals and aims in life”, and “Doing TARGET BEHAVIOUR and being myself are inseparable”.

Expressions that demonstrate or imply the presence of externally regulated behaviour should be coded as dct:extBehaviour_79n2r0sz.

Expressions that demonstrate or imply the presence of identified regulation of behaviour should be coded as dct:identifiedBehaviour_79n2r0t0.

Expressions that demonstrate or imply the presence of introjected regulation of behaviour should be coded as dct:introjBehaviour_79n2r0t0.

In practice, it can be difficult to distinguish expressions belonging to the four types of extrinsic motivation (external, introjected, identified, and integrated self-regulation). Therefore, a less precise categorization can be made between controlled and autonomous motivation. External and introjected regulation belonging to the first category, and identified and integrated self-regulation belonging to the first category. When this dichotomous categorization of extrinsic motivation is maintained, expressions typical of identified and integrated self-regulation can be grouped together.

Expressions that demonstrate or imply the presence of intrinsically regulated behaviour should be coded as dct:intrinsicBehaviour_79n2r0t0.

Expressions that demonstrate or imply the absence of motivation to engage in the target behaviour (i.e. “amotivation”) should be coded as dct:motivation_79n2fh4q, and in a supplementary comment scored “0” or “absent”.

Expressions that refer to the presence of motivation with regard to the target behaviour without specifying the nature of its regulation or expressions that refer to the presence of motivation in general should be coded dct:motivation_79n2fh4q.

The readiness to engage in the behavior, incorporating concepts such as willingness, behavioral expectation, and trying.

Use a likert scale to ask participants to what degree they intend to, are willing to, or plan to perform the target behavior. The items suggested in the 2010 RAA book are: ‘I intend to TARGET BEHAVIOR.’ with anchors ‘Definitely do not’ vs ‘Definitely do’; ‘I will TARGET BEHAVIOR.’ with anchors ‘Unlikely’ vs ‘Likely’; ‘I am willing to TARGET BEHAVIOR.’ with anchors ‘False’ vs ‘True’; and ‘I plan to TARGET BEHAVIOR.’ with anchors ‘Absolutely not’ vs ‘Absolutely’.

Operationalisations that measure the degree to which a target population member has a deliberate (reasoned) plan/intention to engage in TARGET BEHAVIOR. For example, the items suggested in the book are: ‘I intend to TARGET BEHAVIOR.’ with anchors ‘Definitely do not’ vs ‘Definitely do’; ‘I will TARGET BEHAVIOR.’ with anchors ‘Unlikely’ vs ‘Likely’; ‘I am willing to TARGET BEHAVIOR.’ with anchors ‘False’ vs ‘True’; and ‘I plan to TARGET BEHAVIOR.’ with anchors ‘Absolutely not’ vs ‘Absolutely’.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, partcipants are asked whether they intend to perform the target behaviour, for example: ‘Do you intend to exercise for at least 20 minutes, three times per week, for the next 3 months?’

Any expressions that somebody has the intention, goal, or plan to perform a target behavior.

Behaviour is intrinsically regulated when an activity is driven by inherent satisfactions provided by the activity itself and not by an external positive or negative reward that is contingent upon the completion of the activity.

The degree to which engagement in the target activity is intrinsically regulated can be assessed by asking participants “Why are you currently engaged in this activity?” and representing a series of items containing potential answers to this question. Responses to each item are given on a 7-point Likert scale ranging from “Does not correspond at all” to “Corresponds exactly”.

Examples of items expressing intrinsically regulated behaviour are: “Because I think that this activity is interesting”, “Because I think that this activity is pleasant”, “Because this activity is fun”, and “Because I feel good when doing this activity”.

The procedure and items described here are from the Situational Motivation Scale (SIMS) (Guay et al., 2000). Beside items containing statements deemed typical for intrinsically regulated behaviour, the SIMS contains items to assess identified regulation, external regulation, and amotivation. The items belonging to each particular subscale are summed to assess which motivational construct dominates. The SIMS is used to assess motivation for various activities such as physical exercise and academic performance.

Instruments that measure the degree to which engagement in the target activity is intrinsically regulated.

The Situational Motivation Scale (SIMS) (Guay et al., 2000) is a frequently used self-report instrument to assess whether behaviour is intrinsically regulated. Beside intrinsic regulation of behaviour, the SIMS assesses identified regulation, external regulation, and amotivation, and is used to assess motivation for various activities (e.g. physical exercise, academic performance).

The SIMS asks participants “Why are you currently engaged in this activity?”. Scale items represent potential answers to this question. Responses are given on a 7-point Likert scale ranging from “Does not correspond at all” to “Corresponds exactly”.

Examples of statements expressing intrinsic regulation of behaviour are: “Because I think that this activity is interesting”, “Because I think that this activity is pleasant”, “Because this activity is fun”, and “Because I feel good when doing this activity”.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this question: “Why are you currently engaged in TARGET BEHAVIOUR?”

For example: “Why do you currently wear ear plugs at concerts?” or “Why do you currently follow a calorie-restricted diet?”

Expressions that demonstrate or imply that engagement in the target behaviour is intrinsically regulated, using phrases such as “Because I think that TARGET BEHAVIOUR is interesting” and “Because I think that TARGET BEHAVIOUR is pleasant”.

Expressions that demonstrate or imply the presence of externally regulated behaviour should be coded as dct:extBehaviour_79n2r0sz.

Expressions that demonstrate or imply the presence of introjected regulation of behaviour should be coded as dct:introjBehaviour_79n2r0t0.

Expressions that demonstrate or imply the presence of identified regulation of behaviour should be coded as dct:identifiedBehaviour_79n2r0t0.

Expressions that demonstrate or imply the presence of integrated regulation of behaviour should be coded as dct:integratedBehaviour_79n2r0t0.

In practice, it can be difficult to distinguish expressions belonging to the four types of extrinsic motivation (external, introjected, identified, and integrated self-regulation). Therefore, a less precise categorization can be made between controlled and autonomous motivation: external and introjected regulation belonging to the first category, and identified and integrated self-regulation belonging to the first category.

Expressions that demonstrate or imply the absence of motivation to engage in the target behaviour (i.e. “amotivation”) should be coded as dct:motivation_79n2fh4q, and in a supplementary comment scored “0” or “absent”.

Expressions that refer to the presence of motivation with regard to the target behaviour without specifying the nature of its regulation or expressions that refer to the presence of motivation in general should be coded dct:motivation_79n2fh4q.

In introjected behaviour the regulation of action has been partially internalized and is energized by factors such as an approval motive, avoidance of shame, contingent self-esteem, and ego-involvements.

Use Likert scales that measure the degree to which participants’ engagement in the target activity is determined by introjected regulation. Examples of statements expressing introjected regulation of behaviour are: “To show myself that I am an intelligent person”, “Because I want to show myself that I can succeed in my studies”, and “Because of the fact that when I succeed in school I feel important”. Responses on these items are given on a 7-point scale ranging from “Does not correspond at all” to “Corresponds exactly”. These sample items are form the Academic Motivation Scale (AMS) developed by Vallerand, Blais, Brière, & Pelletier (1989) which assesses motivation in post-secondary students (AMS).

The often used Situational Motivation Scale (SIMS) (Guay et al., 2000) does not measure introjected (nor integrated) regulation. The SIMS assesses self-reported intrinsic regulation, identified regulation, external regulation, and amotivation, and is used to assess motivation for various activities (e.g. physical exercise, academic performance).

In practice, it is difficult to distinguish the four types of extrinsic motivation (external, introjected, identified, and integrated self-regulation) using self-report instruments (i.e. insufficient discriminant validity). Therefore, a less precise categorization can be made between controlled and autonomous motivation. External and introjected regulation belonging to the first category, and identified and integrated self-regulation belonging to the first category. When this dichotomous categorization of extrinsic motivation is maintained, items to assess external and introjected self-regulation can be grouped together.

Instruments that measure the degree to which engagement in the target activity is determined by introjected regulation. Examples of statements expressing introjected regulation of behaviour are: “To show myself that I am an intelligent person”, “Because I want to show myself that I can succeed in my studies”, and “Because of the fact that when I succeed in school I feel important”. Responses on these items are given on a 7-point scale ranging from “Does not correspond at all” to “Corresponds exactly”. These items are form the Academic Motivation Scale (AMS) developed by Vallerand, Blais, Brière, & Pelletier (1989) which assesses motivation in post-secondary students (AMS).

The often used Situational Motivation Scale (SIMS) (Guay et al., 2000) does not measure introjected (nor integrated) regulation. The SIMS assesses self-reported intrinsic regulation, identified regulation, external regulation, and amotivation, and is used to assess motivation for various activities (e.g. physical exercise, academic performance).

In practice, it is difficult to distinguish the four types of extrinsic motivation (external, introjected, identified, and integrated self-regulation) using self-report instruments. Therefore, a less precise categorization can be made between controlled and autonomous motivation. External and introjected regulation belonging to the first category, and identified and integrated self-regulation belonging to the first category. When this dichotomous categorization of extrinsic motivation is maintained, items to assessexternal and introjected self-regulation can be grouped together.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this question: “Why are you currently engaged in TARGET BEHAVIOUR?”

For example: “Why do you currently wear ear plugs at concerts?” or “Why do you currently follow a calorie-restricted diet?”

Expressions that demonstrate or imply that engagement in the target behaviour is determined by introjected regulation , using phrases such as “Because I want to show myself that I can succeed in my studies” and “Because of the fact that when I succeed in school I feel important”.

Expressions that demonstrate or imply the presence of externally regulated behaviour should be coded as dct:extBehaviour_79n2r0sz.

Expressions that demonstrate or imply the presence of identified regulation of behaviour should be coded as dct:identifiedBehaviour_79n2r0t0.

Expressions that demonstrate or imply the presence of integrated regulation of behaviour should be coded as dct:integratedBehaviour_79n2r0t0.

In practice, it can be difficult to distinguish expressions belonging to the four types of extrinsic motivation (external, introjected, identified, and integrated self-regulation). Therefore, a less precise categorization can be made between controlled and autonomous motivation: external and introjected regulation belonging to the first category, and identified and integrated self-regulation belonging to the first category.

Expressions that demonstrate or imply the presence of intrinsically regulated behaviour should be coded as dct:intrinsicBehaviour_79n2r0t0.

Expressions that demonstrate or imply the absence of motivation to engage in the target behaviour (i.e. “amotivation”) should be coded as dct:motivation_79n2fh4q, and in a supplementary comment scored “0” or “absent”.

Expressions that refer to the presence of motivation with regard to the target behaviour without specifying the nature of its regulation or expressions that refer to the presence of motivation in general should be coded dct:motivation_79n2fh4q.

A representation of the world that one believes to accurately describe a part of the world.

Knowledge can be measured by presenting statements about parts of reality (i.e. about objects or relationships that exist or do not exist in the world) related to the target behaviour, and determining a participant’s belief concerning this statement by using a 5-point Likert scale that ranges from “certainly not true” through “probably not true”, “no idea”, “certainly true”, “certaintly true”.

An example of such an item for the target behaviour “protecing hearing against exposure to loud noise” is: “Exposure to very loud music (>120 decibel) can lead to acute and permanent hearing loss” or “When volume restriction is switched on for a MP-3 player sold in E.U., the maximum attainable level of decibels is 85.”

Questions that measure participants’ beliefs concerning statements about reality (i.e. about objects or relationships that exist or do not exist in the world) related to the target behaviour. These beliefs can be determined by using a 5-point Likert scale that ranges from “certainly not true” through “probably not true”, “no idea”, “certainly true”, “certaintly true”.

Two examples for the target behaviour “protecing hearing against exposure to loud noise” are: “Exposure to very loud music (>120 decibel) can lead to acute and permanent hearing loss” and “When volume restriction is switched on for a MP-3 player sold in E.U. the maximum attainable level of decibels is 85.”

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, participants are asked to express their knowledge regarding the target behaviour or topics related to the target behaviour.

These question can be formulated in the following way: “What do you know about TARGET BEHAVIOUR?” and “What do you know about POTENTIAL CONSEQUENCE OF (NOT) ENGAGING IN TARGET BEHAVIOUR?” For example: “What do you know about noise-induced hearing loss?” and “What do you know about the effect of loud noise on the hearing?”

Expressions that demonstrate or imply the presence or absence of knowledge concerning the target behaviour or topics related to the target behaviour. These expressions pertain to structural or causal characteristics of the natural world. Importantly, these are separate from, for example, one’s evaluations of those characteristics. E.g. “smoking causes lung cancer” is knowledge; “if I smoke, I get lung cancer” is expectation; “I don’t like lung cancer” is evaluation.

Motivation to comply with a specific social referent refers to an individual’s motivation to behave in a way that is most approved of by that social referent. Note that motivation to comply is separate from the approval or disapproval itself, which is captured in dct:referentApproval_73dnt5zd.

A social referent can be a specific individual, a group of distinct individuals (e.g. one’s siblings), or a more generically defined group (e.g. the residents of my city).

The motivation to comply with a given social referent combines multiplicatively with the perceived approval of that referent (dct:referentApproval_73dnt5zd) into an injunctive norm belief (dct:injNorms_belief_73dnt5zg).

To measure the motivation to comply with a given social referent, measure how much people want to do what the relevant social referents think they should do with regard to the target behavior.

In most behaviors and populations, unidimensional scales can be used. As item stem, use ‘When it comes to [TARGET BEHAVIOR], I want to do what [SOCIAL REFERENT] think(s) I should do.’, and as anchors, ‘Not at all’ and ‘Very much’. For example, ‘When it comes to whether I avoid meat during dinner, I want to do what my partner thinks I should do. [‘Not at all'|'Very much’]’ or ‘When it comes to my coffee consumption, I want to do what my colleagues think I should do. [‘Not at all'|'Very much’]’.

However, for some behaviors and some populations, it’s not only possible that individuals want to do what a social referent thinks they should do to a certain degree, but it’s also possible that the individual actively wants to do what that social referent would disapprove of. For example, people may strongly want to clearly demonstrate distancing themselves from members of a perceived outgroup (e.g. adolescents may want to disobey middle-aged people).

In such situations, you can use a bidimensional scale to account for this variation. Use item stem ‘When it comes to [TARGET BEHAVIOR], I want to do what [SOCIAL REFERENT]…’ with anchors ‘Do(es)n’t want me to do’ and ‘Want(s) me to do’. For example, ‘When it comes to condom use with a partner, I want to do what other adolescents… [Don’t want me to do|Want me to do]’, or ‘When it comes to how late I go to bed, I want to do what my older sibling … [Doesn’t want me to do|Wants me to do]’

Motivation is the experience of desire or aversion that instigates, directs, and maintains goal-oriented behaviour, and is energized intrinsically, by internal factors related to the individual, and extrinsically, by external factors acting on the individual.

The presence or absence of motivation for engagement in the target behaviour can be measured with the question “Are you motivated to engage in TARGET BEHAVIOUR?” followed by either a dichotomous response option (“yes”/“no”) or a 5-point Likert scale ranging from “Very Much” to “Not at all”.

According to self-determination theory (Deci & Ryan, 1985), different types of motivation underlie human behaviour. These different types of motivation may be positioned on a continuum from high to low levels of self-determination, with intrinsic motivation and amotivation at the extreme ends. Instructions for measuring these different categories of motivation (each one with its specific place on the motivation continuum) can be found at their respective dcts, i.e.: dct:intrinsicBehaviour_79n2r0t0, dct:integratedBehaviour_79n2r0t0, dct:identifiedBehaviour_79n2r0t0, dct:introjBehaviour_79n2r0t0, and dct:extBehaviour_79n2r0sz.

Questions or questionnaires that measure the presence or absence of motivation for engagement in the target behaviour. In its simplest form, this can be measured with the question “Are you motivated to engage in TARGET BEHAVIOUR?” followed by either a dichotomous response option (“yes”/“no”) or a 5-point Likert scale ranging from “Very Much” to “Not at all”.

Definitions and operationalisations of motivation vary greatly across studies. One influential theory on human motivation is self-determination theory (Deci & Ryan, 1985). According to this theory, different types of motivation underlie human behaviour. These different types of motivation are positioned on a continuum from high to low levels of self-determination, with intrinsic motivation and amotivation at the extreme ends. The Situational Motivation Scale (SIMS) (Guay et al., 2000) is a frequently used self-report instrument to assess which type of motivation dominates for a particular behaviour or activity (see dct:intrinsicBehaviour_79n2r0t0, dct:integratedBehaviour_79n2r0t0, dct:identifiedBehaviour_79n2r0t0, dct:introjBehaviour_79n2r0t0, and dct:extBehaviour_79n2r0sz).

Questions that measure specific types of motivation (such as the questions used in the SIMS), and not motivation in general (for engagement in the target behaviour) should not be coded as dct:motivation_79n2fh4q, but should be given the code pertaining to the specific subconstruct measured (e.g. dct:extBehaviour_79n2r0sz).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this question “Are you motivated to engage in TARGET BEHAVIOUR?” For example: “Are you motivated to follow a calorie-restricted diet?” or “Are you motivated to stop smoking?”

Expressions that demonstrate or imply the presence or absence of motivation to perform the target behavior. For example, using phrases as “I am motivated to follow a calorie-restricted diet” or “I do not have enough motivation to stop smoking”

Expressions that demonstrate or imply that engagement in the target behaviour is externally regulated, should be coded as dct:extBehaviour_79n2r0sz.

Expressions that demonstrate or imply the presence of introjected regulation of behaviour should be coded as dct:introjBehaviour_79n2r0t0.

Expressions that demonstrate or imply the presence of identified regulation of behaviour should be coded as dct:identifiedBehaviour_79n2r0t0.

Expressions that demonstrate or imply the presence of integrated regulation of behaviour should be coded as dct:integratedBehaviour_79n2r0t0.

Expressions that demonstrate or imply the presence of intrinsically regulated behaviour should be coded as dct:intrinsicBehaviour_79n2r0t0.

An individual’s perception whether an activity or situation satisfies an universal human need ( i.e. what is considered essential or necessary for well-being and healthy functioning).

Numerous self-report instruments exist to measure need satisfaction. These instruments measure the degree needs are satisfied in a particular domain (e.g. work, relationship), the degree needs are satisfied by a particular activity (e.g. a certain type of sports), or they assess need satisfaction in general.

Instruments that measure need satisfaction most often assume that humans have universal needs. In particular, Self-Determination Theory (SDT) (Deci & Ryan, 1985) and Maslow’s (1968) Hierarchy of Needs have instigated the development of instruments to measure the satisfaction of these universal needs. The Basic Psychological Needs Scale (BPNS) (Gagné et al., 2003) and the Balanced Measure of Psychological Needs (BMPN) (Sheldon et al., 2012) measure needs from a SDT-perspective, while the Basic Need Satisfaction Inventory (BNSI) (Leidy et al., 1994)) adopts Maslow’s standpoint on human needs.

For behaviour change (e.g. instigating healthy lifestyle choices), measuring general, as well as measuring domain- or activity-specific need satisfaction may have its merits. For example, Timmerman et al. (2001) found that the lower the level of basic need satisfaction, the more likely one is engaged in emotional eating. Need satisfaction at this general level, however, is harder to change than needs being met by specific activities or in a narrow domain. Therefore, instruments pertaining to the latter category are preferably used.

Need satisfaction can be asssessed by following a series of four steps: Firstly, depending on the theoretical background one adheres to (e.g. SDT) a list of of universal needs is composed. Secondly, to measure whether these needs are met by the situation or activity of interest, for each of these needs a number of items are used to assess the level of needs satisfaction. Thirdly, the item sumscores for each need, and the overall sumcore of all needs together are calculated. Fourth, the degree individual needs and all need together are satisfied by an activity or situation can be deduced from the two sumscores.

Need satisfaction in a particular domain can, for example, be assessed with the following sample items. These sample items pertain to the assessment of need satisfaction at work but (with slight adjustments) could also be used to measure need satisfaction in other domains or concerning specific activities. Participants are asked to indicate how true each statement is for them, with answer option ranging from “Not at all true” to “Very true” on a 7-point Likert scale: “I feel like I can make a lot of inputs to deciding how my job gets done”, “People at work care about me”, and “Most days I feel a sense of accomplishment from working”. These three items reflect what SDT regards as the three basic human needs: autonomy, relatedness, and competence.

Numerous self-report instruments exist to measure need satisfaction. These instruments measure the degree needs are satisfied in a particular domain (e.g. work, relationship), the degree needs are satisfied by a particular activity (e.g. a certain type of sports), or they assess need satisfaction in general.

Instruments that measure need satisfaction most often assume that humans have universal needs. In particular, Self-Determination Theory (SDT) (Deci & Ryan, 1985) and Maslow’s (1968) Hierarchy of Needs have instigated the development of instruments to measure the satisfaction of these universal needs. The Basic Psychological Needs Scale (BPNS) (Gagné et al., 2003) and the Balanced Measure of Psychological Needs (BMPN) (Sheldon et al., 2012) measure needs from a SDT-perspective, while the Basic Need Satisfaction Inventory (BNSI) (Leidy et al., 1994)) adopts Maslow’s standpoint on human needs.

For behaviour change (e.g. instigating healthy lifestyle choices), measuring general, as well as measuring domain- or activity-specific need satisfaction may have its merits. For example, Timmerman et al. (2001) found that the lower the level of basic need satisfaction, the more likely one is engaged in emotional eating. Need satisfaction at this general level, however, is harder to change than needs being met by specific activities or in a narrow domain. Therefore, instruments pertaining to the latter category are preferably used.

Need satisfaction can be asssessed by following a series of four steps: Firstly, depending on the theoretical background one adheres to (e.g. SDT) a list of of universal needs is composed. Secondly, to measure whether these needs are met by the situation or activity of interest, for each of these needs a number of items are used to assess the level of needs satisfaction. Thirdly, the item sumscores for each need, and the overall sumcore of all needs together are calculated. Fourth, the degree individual needs and all need together are satisfied by an activity or situation can be deduced from the two sumscores.

Need satisfaction in a particular domain can, for example, be assessed with the following sample items. These sample items pertain to the assessment of need satisfaction at work but (with slight adjustments) could also be used to measure need satisfaction in other domains or concerning specific activities. Participants are asked to indicate how true each statement is for them, with answer option ranging from “Not at all true” to “Very true” on a 7-point Likert scale: “I feel like I can make a lot of inputs to deciding how my job gets done”, “People at work care about me”, and “Most days I feel a sense of accomplishment from working”. These three items reflect what SDT regards as the three basic human needs: autonomy, relatedness, and competence.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use the general question: “Do you feel your needs are being met by TARGET ACTIVITY/BEHAVIOUR?"; or use a more narrow question to ask whether a specific universal need is met: “Do you feel you have a connection with the people at work?”. The latter question is an example to assess whether “relatedness”, one of the universal needs according to SDT, is met.

Expressions that demonstrate or imply that needs are satisfied or not satisfied by the activity or behaviour of interest, for example: “I feel like I can make a lot of inputs to deciding how my job gets done”, “People at work do not care about me”, and “Most days I feel a sense of accomplishment from working”.

Self-Determination Theory (SDT) (Deci & Ryan, 1985) regards relatedness, autonomy, and competence as the three basic human needs. Therefore, expressions that indicate that autonomy and competence are regarded as important individual needs should be coded as dct:need_79n2fh4r and as, respectively, dct:autonomy_73dnt5zx and dct:capacity_73dnt602.

In a similar fashion, other expressions of needs may be given the coding dct:need_79n2fh4r and of another dct, e.g. dct:injunctiveNorms_73dnt5zj, if the participant expresses that approval of people around him/her of his/her behaviour is regarded as an important individual need.

People’s perceptions of the degree to which they are capable of, or have control over, performing a given behavior.

Use Likert scales that measure the degree to which participants believe the target behavior to be under their control and something they are capable of performing successfully, for example by measuring their perceived capacity and perceived autonomy to perform the target behavior.

The items suggested in the 2010 RAA book are: ‘I am confident that if I want to, I can TARGET BEHAVIOR.’ with anchors ‘No confidence at all’ vs ‘A lot of confidence’; ‘Whether I TARGET BEHAVIOR is …’ with anchors ‘Not up to me’ vs ‘Completely up to me’; ‘If I really wanted to, I could TARGET BEHAVIOR.’ with anchors ‘Unlikely’ vs ‘Likely’; and ‘For me to TARGET BEHAVIOR is under my control.’ with anchors ‘Not at all’ vs ‘Completely’. If only the first and third of these items are measured, the ‘Capacity’ construct is measured instead. If only the second and fourth items are measured, the ‘Autonomy’ construct is measured instead.

Note that questions about how easy or difficult participants perceive a target behavior to be fall outside of the constructs of Perceived Behavioral Control, Autonomy, and Capacity as defined in the 2010 RAA book (see also the discussion on page 164).

Questions or questionnaires that measure the perceived degree to which the target behavior or the contrast behavior is both under the control of the target population individual and something they are confident they can successfully perform. If only one aspect is measured (i.e. only control or only confidence), code this as the ‘Autonomy’ or ‘Capacity’ constructs.

For example, the items suggested in the book are: ‘I am confident that if I want to, I can TARGET BEHAVIOR.’ with anchors ‘No confidence at all’ vs ‘A lot of confidence’; ‘Whether I TARGET BEHAVIOR is …’ with anchors ‘Not up to me’ vs ‘Completely up to me’; ‘If I really wanted to, I could TARGET BEHAVIOR.’ with anchors ‘Unlikely’ vs ‘Likely’; and ‘For me to TARGET BEHAVIOR is under my control.’ with anchors ‘Not at all’ vs ‘Completely’. If only the first and third of these items are measured, the ‘Capacity’ construct should be coded instead. If only the second and fourth items are measured, the ‘Autonomy’ construct should be coded instead.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these questions: ‘Please list any factors or circumstances that would make it easy or enable you to target behavior.’ and ‘Please list any factors or circumstances that would make it difficult or prevent you from target behavior.’

Expressions that demonstrate or imply that one perceives behavioural control with regard to the target behaviour to be present or absent. These are general expressions of perceived behavioural control, i.e. not refering specifically to the ability or autonomy to perform the behaviour, e.g.: “Only time will tell if I will be able to follow the entire exercise program”.

Expressions that demonstrate or imply either the presence/absence of confidence about one’s ability to perform the target behavior, should be coded as dct:capacity_73dnt602. An example of a statement expressing the capacity to perform the target behaviour is: “I am confident that I have the required muscle strength to perform all the exercises of the prescribed program”.

Expressions that demonstrate or imply the presence/absence of confidence about one’s control over performance or nonperformance of the target behavior (i.e. if (non)performance is “up to them”), for example because of a barrier, obstacle, or facilitating condition or circumstance, should be coded as dct:autonomy_73dnt5zx. An example of a statement expressing the perceived autonomy to perform the target behaviour is: “My exercising for at least 20 minutes, three times per week, is up to me”.

Perceived social pressure to perform (or not to perform) the target behavior.

Use likert scales that measure both perceived social approval and perceived behavior of important social referents in general. It is important that no specific individuals are referenced, but that the items do refer to individuals that are similar, important or seen as important to follow. Note that the behavior in these items must always be the target behavior.

The example items in the 2010 RAA book are ‘Most people who are important to me think I should TARGET BEHAVIOR.’ with anchors ‘False’ vs ‘True’; ‘Most people whose opinions I value would … of my TARGET BEHAVIOR’ with anchors ‘Disapprove’ vs ‘Approve’; ‘Most people I respect and admire will TARGET BEHAVIOR.’ with anchors ‘Unlikely’ vs ‘Likely’; and ‘How many people like you TARGET BEHAVIOR?’ with anchors ‘Nobody’ vs ‘Everybody’.

Questionnaires that measure both perceived social approval and perceived behavior of important social referents in general. It is important that no specific individuals are referenced, but that the items do refer to individuals that are similar, important or seen as important to follow. Note that the behavior in these items must always be the target behavior.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use the introduction ‘When it comes to TARGET BEHAVIOR, there might be individuals or groups who would think you should or should not perform this behavior.’, and then ask these questions: ‘Please list any individuals or groups who would approve or think you should TARGET BEHAVIOR’; ‘Please list any individuals or groups who would disapprove or think you should not TARGET BEHAVIOR’; ‘Sometimes, when we are not sure what to do, we look to see what others are doing. Please list the individuals or groups who are most likely to TARGET BEHAVIOR’; and ‘Please list the individuals or groups who are least likely to TARGET BEHAVIOR’.

Expressions that refer to social pressure in general. However, expressions about perceived behavior of others are coded as dct:perceivedNorm_descriptive_73bg61tx and expressions about perceived (dis)approval of others are coded as dct:perceivedNorm_injunctive_73bg2wm7.

Feelings of moral obligation to perform or refrain from specific action.

Whether a person holds a certain personal norm concerning the topic of interest can be measured by proposing statements concerning the perceived moral obligation to (not) engage in a particular behaviour and asking participants to repond on a 5- or 7-points Likert scale with anchors “Strongly disagree” and “Strongly agree”.

Two sample items to measure personal norms are: “Because of my own values/principles I feel an obligation to use public transport instead of the car for everyday trips here in X” and “Regardless of what other people do, because of my own values/principles I feel an obligation to use public transport instead of the car for everyday trips here in X” (from “Social context, personal norms and the use of public transportation” by Bamberg et al., 2007).

Two sample questions used by Doran et al., 2015 in “The relative importance of social and personal norms in explaining intentions to choose eco‐friendly travel options” are: “I do feel a moral obligation to pay more for a trip if this helps to protect the environment” and “I do feel a moral obligation to make an effort to stay at environmentally friendly accommodation when travelling”.

Questions or questionnaires that measure the level of agreement/disagreement with a certain personal norm concerning the topic of interest (e.g. the target behaviour).

A 5- or 7-points Likert scale with anchors “Strongly disagree” and “Strongly agree” are most often used to measure personal norms, as is the case in the following sample items: “Because of my own values/principles I feel an obligation to use public transport instead of the car for everyday trips here in X” and “Regardless of what other people do, because of my own values/principles I feel an obligation to use public transport instead of the car for everyday trips here in X” (from “Social context, personal norms and the use of public transportation” by Bamberg et al., 2007).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these questions: “Why do you engage/not engage in TARGET BEHAVIOUR?”, e.g. “Why do you travel by public transport?” or “Why do you not donate money for charity?”.

Expressions that demonstrate or imply the presence or absence of personal norms concerning the target behaviour. Expressions containing words and phrases such as “values”, “principles”, and “moral obligation” may indicate the presence of personal norms linked to the behaviour in question.

Expressions that refer to the perceived approval or disapproval of the behaviour by others, should be coded as dct: injunctiveNorms_73dnt5zj.

Expressions that refer to the perceived behaviour of others (i.e. whether others perform the target behaviour or not), should be coded as dct:descriptiveNorms_73dnt5zp.

Expressions that demonstrate or imply either the presence/absence of confidence about one’s ability to perform the target behaviour, should be coded as dct:capacity_73dnt602

Expressions that demonstrate or imply the presence/absence of confidence about one’s control over performance or nonperformance of the target behaviour (i.e. if (non)performance is “up to them”), for example because of a barrier, obstacle, or facilitating condition or circumstance, should be coded as dct:autonomy_73dnt5zx.

Personal norms most often are a crucial component of one’s self-identity (see dct:self_identity_79n2fh4t). Therefore, if a participants explicitly expresses that a personsal norm is central to his/her self-identity, the expression should also be coded as such (i.e. dct:self_identity_79n2fh4t).

Personality is the set of psychological traits and mechanisms within the individual that are organized and relatively enduring and that influence his or her interactions with, and adaptations to, the intrapsychic, physical, and social environments.

Numerous instruments exist that aim to measure human personality. Most instruments are self-report questionnaires. However, other measures of personality exist such as performance tests, instruments for direct observation, semi-structured interviews, and projective tests.

The choice for a particular instrument is largely driven by the specific personality theory one adheres to; for example the influential Big Five model of personality traits, or five-factor model (FFM), has inspired the creation of a number of self-report instruments (such as the the Revised NEO Personality Inventory [NEO PI-R]) that assess the relative presence of each of the Big Five personality traits (openness to experience, conscientiousness, extraversion, agreeableness, and neuroticism) in a person.

The NEO PI-R consists of 240 items such as “I like to have a lot of people around me” and “I often find myself in disagreements with my family and colleagues”. Answers are given on a 5-points Likert scale ranging from “Totally disagree” to “Totally agree”. Each of the five personality dimensions in the NEO PI-R consists of six facets. At the broadest level, personality can be described in terms of five basic dimensions by summing the six facets in each domain. More precise estimates of standing on the five factors, can be provided by factor scores (i.e. a weighted combination of scores on all 30 facets). For interpretation, an individual’s scores on the NEO PI-R are compared to a relevant reference group (e.g. adult men).

Self-report questions that are assumed to measure stable facets of personality (i.e. “traits”) across time and situations such as “I like to have a lot of people around me” and “I often find myself in disagreements with my family and colleagues”. These examples are from the NEO Personality Inventory (NEO PI-R) and answers here are given on a 5-points Likert scale with the anchors “Totally disagree” to “Totally agree”.

Beside self-report questionnaires, other measures of personality exist such as performance tests, instruments for direct observation, semi-structured interviews, and projective tests.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, people are asked for their typical behaviour in a series of different situations (deemed appropriate to incite representative behaviour). For example: “You are invited to a party where you hardly know any of the guests. How would you describe your behaviour at the party?”.

The Structured Interview for the Five-Factor Model of Personality (SIFFM; T J. Rull & T A. Widiger, 1997) is a 120-item semistructured interview to asses personality according to the five-factor model (FFM). An example of an item measuring neuroticism is: “Do you worry a lot about the future or things that might go wrong?” [If yes:] “What kinds of things do you worry about?” [After response] “What proportion of the time?”. An SIFFM-item for the assessment of extraversion is: “Would you describe yourself as a leader?” [If yes:] “Please give me some examples of yourself as a leader?” [After response] “Do others ever consider you to be bossy or domineering?'’.

Expressions that demonstrate or imply the presence of behaviour assumed to be typical for a particular personality type, e.g. answers to the question “You are invited to a party where you hardly know any of the guests; how would you describe your behaviour at the party?” that include elements such “withdrawn”, “shy”, and “uncomfortable” point to a low degree of extraversion.

Responses to items on the Structured Interview for the Five-Factor Model of Personality (SIFFM; T J. Rull & T A. Widiger, 1997) are scored as either 0 (absent), 1 (present and does not result in significant dysfunction), or 2 (present and may result in significant dysfunction). Raw SIFFM-scores are not compared with a reference group but, beside reflecting the degree to which a personality trait is endorsed, directly indicate the level of dysfunction associated with the trait (i.e. criterion referenced, as opposed to norm referenced).

The degree of likability of a image representing perceptions of typical persons engaging in or refraining from a certain behaviour.

Prototype favourability can be assessed by rating adjectives relating to the prototype, for instance, how “cool” or “likeable” the prototype is, and measuring responses on a 5- or 7-point Likert scale.

For example: “How likeable or dislikeable do you think the type of person who drinks four units of alcohol (over the legal limit) and drives thereafter is?”, with the scale anchors “Very likeable” and “Very dislikeable” (example from “Understanding young and older male drivers’ willingness to drive while intoxicated” by Rivis et al., 2011).

Higher prototype favourability is assumed to be related to the increased likelihood that individuals will actually engage in the behaviour or will exhibit the willingness or intention to engage in the health-related behaviour associated with the prototype.

Questions or questionnaires that measure prototype favourability by rating adjectives relating to the prototype, for instance, how “cool” or “likeable” the prototype is, and measuring responses on a 5- or 7-point Likert scale.

For example: “How likeable or dislikeable do you think the type of person who drinks four units of alcohol (over the legal limit) and drives thereafter is?”, with the scale ancors “Very likeable” and “Very dislikeable” (example from “Understanding young and older male drivers’ willingness to drive while intoxicated” by Rivis et al., 2011).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use the questions: “How would you describe the typical person that engages in TARGET BEHAVIOUR?” or “What is your image of the typical person that engages in TARGET BEHAVIOUR?”

Expressions that demonstrate or imply the degree of favourability, containing words such as “smart/dumb”, “attractive/unattractive”, and “desirable/undiserable”.

The words chosen to express the degree of (un)favourability associated with the prototype in itself is also informative; i.e. it tells us which association is most salient when refering to a prototype. For example, indicating that a person wearing ear plugs when in a club is “cool/uncool” and not saying that this particular person is “smart/dumb”, may indicate that “coolness” is an important factor (and likely more important than “smartness”) to assess the level of favourability with regard to the prototype.

Expressions that demonstrate or imply the degree of similarity, containing verbs such as “familiar/unfamiliar”, “related/unrelated”, “resemble/differ” should be coded as dct:prototypeSimilarity_79n2fh4t.

Expressions that exhibit the presence of a prototype image associated with the behaviour without a reference to favourability or similarity should be coded as dct:prototype_79n2fh4t.

The perceived resemblance to the self with an image representing perceptions of typical persons engaging in or refraining from a certain behaviour.

Prototype similarity can be assessed by directly asking “how similar are you to the PROTOTYPE ?” and measuring responses on a 5- or 7-point Likert scale.

For example, prototype similarity can be assessed with the following question: “In general, how similar are you to the type of person who drinks four units of alcohol (above the legal limit) and drives thereafter?”, with the scale anchors “Very similar” and “Not at all similar (example from “Understanding young and older male drivers’ willingness to drive while intoxicated” by Rivis et al., 2011).

Higher prototype similarity is assumed to be related to the increased likelihood that individuals will actually engage in the behaviour or will exhibit the willingness or intention to engage in the health-related behaviour associated with the prototype.

Questions or questionnaires that measure prototype similarity by asking “how similar are you to the PROTOTYPE ?” (or similar) and measuring responses on a 5- or 7-point Likert scale.

For example, prototype similarity can be assessed with the following question: “In general, how similar are you to the type of person who drinks four units of alcohol (above the legal limit) and drives thereafter?”, with the scale anchors “Very similar” and “Not at all similar (example from “Understanding young and older male drivers’ willingness to drive while intoxicated” by Rivis et al., 2011).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use the questions: “How would you describe the typical person that engages in TARGET BEHAVIOUR?” or “What is your image of the typical person that engages in TARGET BEHAVIOUR?”

Expressions that demonstrate or imply the degree of similarity, containing verbs such as “familiar/unfamiliar”, “related/unrelated”, and “resemble/differ”.

Expressions that demonstrate or imply the degree of favourability, containing verbs such as “smart/dumb”, “attractive/unattractive”, and “desirable/undiserable” should be coded as dct:prototypeFavourability_79n2fh4t.

Expressions that exhibit the presence of a prototype image associated with the behaviour without a reference to favourability or similarity should be coded as dct:prototype_79n2fh4t.

A person’s image of the typical person who performs (or does not perform) a given behaviour.

An individual’s image of the typical person who performs (or does not perform) a given behaviour (i.e. the “prototype”) can be assessed by measuring the subconstructs prototype favourability (see dct:prototypeFavourability_79n2fh4t) and prototype similarity (see dct:prototypeSimilarity_79n2fh4t).

When the prototype performs the target behaviour, higher prototype favourability and similarity are, in general, related to an increased likelihood that an individual will engage in the behaviour, or will exhibit the willingness or intention to engage in the behaviour associated with the prototype (vice versa, when the protoptype abstains from performing the behaviour, higher prototype favourability and similarity are related to a decreased likelihood that an individual will engage in the behaviour).

In addition, analyzing the interaction between prototype favourability and similarity can increase predictive validity (Van Lettow et al., 2016). For example, in the study by Rivis et al. (2011) more favourable perceptions of the drink-driver prototype were associated with greater willingness to drink and drive at high (but not moderate or low) levels of prototype similarity.

Prototype images with regard to the target behaviour can be assessed by measuring the subconstructs prototype favourability (see dct:prototypeFavourability_79n2fh4t) and prototype similarity (see dct:prototypeSimilarity_79n2fh4t).

When the prototype performs the target behaviour, higher prototype favourability and similarity are, in general, related to an increased likelihood that an individual will engage in the behaviour, or will exhibit the willingness or intention to engage in the behaviour associated with the prototype. When the protoptype abstains from performing the behaviour, higher prototype favourability and similarity are related to a decreased likelihood that an individual will engage in the behaviour.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use the questions: “How would you describe the typical person that engages (or: does not engage) in TARGET BEHAVIOUR?” or “What is your image of the typical person that engages (or: does not engage) in TARGET BEHAVIOUR?”

Expressions that exhibit the presence of a prototype image associated with the performance of a behaviour (or abstaining from it) without a reference to favourability or similarity. For example in the answer “These people are very health concious” in response to the question “What is your image of the typical person that does not drink alcohol?” no clear aspect of favourability or similarity is apparent. However, a reference to a prototypical image is nevertheless made.

Expressions that demonstrate or imply the degree of favourability, containing adjunctives such as “smart/dumb”, “attractive/unattractive”, “desirable/undiserable” should be coded as dct:prototypeFavourability_79n2fh4t.

Expressions that demonstrate or imply the degree of similarity, containing adjunctives such as “familiar/unfamiliar”, “related/unrelated”, “resemble/differ” should be coded as dct:prototypeSimilarity_79n2fh4t.

The perceived approval or disapproval of a social referent refers to whether a given individual believes that social referent thinks that that given individual should engage in the target behavior. Note that this concerns whether the individual should engage in the target behavior, not whether the social referent themselves should engage in the target behavior. Also note that this is distinct from whether the social referent themselves does actually engage in the target behavior, which is captured in dct:referentBehavior_73dnt5zk.

A social referent can be a specific individual, a group of distinct individuals (e.g. one’s siblings), or a more generically defined group (e.g. the residents of my city).

The perceived approval for a given social referent combines multiplicatively with the motivation to comply with that referent (dct:motivationToComply_73dnt5zf) into an injunctive norm belief (dct:injNorms_belief_73dnt5zg).

Use item stems that list the target behavior and the social referent, and use disapproval and approval as anchors, in that order, with an intensifying adjective. Specifically, as item stem, use ‘If I were to (engage in) [TARGET BEHAVIOR], [SOCIAL REFERENT] would…’ with anchors ‘strongly disapprove’ versus ‘strongly approve’. Note that the ‘engage in’ is optional, depending on the kind of behavior. For example, ‘If I were to engage in a demonstration, my neighbour would… [Strongly disapprove|Strongly approve]’

Because this is a bidimensional scale, the scale midpoint reflects neutrality (i.e. one does not think the relevant social referent disapproves or approves of this target behavior. This means that a seven-point response scale is slightly preferred over a five-point response scale to allow participants to express three degrees of perceived (dis)approval.

Page 135 of the RAA book

The perceived behavior of a social referent is a given individual’s perception of whether that person engages in the target behavior. Note that this perception only relates to the behavior of the social referent, not to what the social referent might believe. Perceptions of the social referent’s (dis)approval of whether the given individual engages in the target behavior is captured in dct:referentApproval_73dnt5zd.

A social referent can be a specific individual, a group of distinct individuals (e.g. one’s siblings), or a more generically defined group (e.g. the residents of my city).

Depending on the nature of the target behavior, engagement in that behavior can be conceived as binary (i.e. one either does or does not engage in the target behavior, e.g. getting tested for STIs every six months), a matter of frequency (i.e. one engages in the target behavior with a frequency from zero up to a given feasible maximum frequency in a given timespan, e.g. the frequency with which one washes their hands conform the guidelines), a matter of intensity (i.e. one engages in the target behavior to a degree from zero up to a given feasible maximum intensity, e.g. the amount of kilocalories one consumed in a meal), or a combination of these (i.e. one engages in the target behavior with a given frequency and with a given intensity, e.g. how frequently one drinks alcohol, and how many grams of alcohol one consumes when one does).

The perceived behavior of a given social referent combines multiplicatively with the identification with that social referent (dct:referentIdentification_73dnt5zl) into a descriptive norm belief (dct:descrNorms_belief_73dnt5zm).

To measure the perceptions individuals have of social referents’ behavior, measure the probability that they engage in that target behavior. Use an item stem listing both the target behavior and the social referent, and use anchors ‘improbable’ versus ‘probable’, with intensifying adjectives.

‘How likely do you think it is that [SOCIAL REFERENT] engage(s) in [TARGET BEHAVIOR]?’ with anchors ‘Very improbable’ and ‘Very probable’.

For example, ‘How likely do you think it is that your close family members engage in recycling? [Very improbable|Very probable]’

Identification with a social referent is defined as the degree to which an individual wants to be like that social referent with regards to the target behavior.

A social referent can be a specific individual, a group of distinct individuals (e.g. one’s siblings), or a more generically defined group (e.g. the residents of my city).

The identification with a given social referent combines multiplicatively with the perceived behavior of that social referent (dct:referentBehavior_73dnt5zk) into a descriptive norm belief (dct:descrNorms_belief_73dnt5zm).

To measure the identification with a given social referent, measure how much people want to be like the relevant social referents with regard to the target behavior.

Note that for any given social referent, it’s possible that an individual wants to be like that referent, but it’s also possible that the individual actively wants to be unlike that social referent. For example, people may strongly want to distance themselves from (be unlike) members of a perceived outgroup (e.g. adolescents may want to be unlike middle-aged people).

Therefore, bidimensional scales are required to account for this variation. As item stem, use ‘Concerning [BEHAVIOR], I want to be like [SOCIAL REFERENT]…’, with anchors ‘as little as possible’ versus ‘as much as possible’. For example, ‘Concerning being able to speak in public, I want to be like a movie star… [As little as possible|As much as possible]’

Because this is a bidimensional scale, the scale midpoint reflects neutrality (i.e. the participant neither wants to identify with, nor distance themselves from, the relevant social referent regarding this target behavior). This means that a seven-point response scale is slightly preferred over a five-point response scale to allow participants to express three degrees of desire to be similar or dissimilar.

Risk perception is the perceived susceptibility to and the perceived severity of a threat.

Risk perception is determined by the perceived susceptibility (see dct:threat_susceptibility_79n2fh4s) to and the perceived severity (see dct:threat_severity_79n2fh4r) of a (health) threat, i.e. it is the product term of the multiplication of these component constructs.

See dct:threat_susceptibility_79n2fh4s and dct:threat_severity_79n2fh4r for instructions for developing measurement instruments for these subconstructs. The resulting measurements express the degree of susceptibility and severity as a numeric score which, subsequently, can be multiplicated to estimate an individual’s perception of the level of risk associated with a threat.

Instead of multiplying the two subcomponents, threat susceptibility and threat severity are also summed by some risk perception measures, e.g. the 12-item Risk Behaviour Diagnosis (RBD) Scale (Witte et al., 1995) includes 3 items measuring threat susceptibility and 3 items measuring threat severity on a 5-point Likert scale. Together these 6 (summed) items scores determine a participant’s risk perception (there named “perceived threat”).

Risk perception is measured by measuring both of the component constructs: perceived threat susceptibility (see dct:threat_susceptibility_79n2fh4s) and perceived threat severity (see dct:threat_severity_79n2fh4r) and, subsequently, multiplying or summing the scores of these two components.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use the questions: “How would you rate the risk associated with HEALTH THREAT (e.g. TARGET BEHAVIOUR)?” or “Can you describe your personal ideas of the risk associated with HEALTH THREAT (e.g. TARGET BEHAVIOUR)?”

Expressions that illustrate the participant’s risk perception with regard to the health threat (e.g. the target behaviour) without a reference to susceptibility or severity.

For example, if the question is “Can you describe your personal ideas of the risk associated with smoking cigarettes?” and the response is “I know there is a risk associated with smoking, so I try not to smoke to much” no clear reference is made to to the severity of the threat or to personal susceptibility. However, a reference to a risk perception is nevertheless made.

Expressions that demonstrate or imply the degree of perceived threat susceptibility, containing phrases and words such as “high/low risk”, “high/low prevalence”, “likely/unlikely”, “nobody I know”/“a lot of people I know” (the latter two responses refering to adverse consequences of engaging in the target behaviour) should be coded as dct:threat_susceptibility_79n2fh4s.

Expressions that demonstrate or imply the degree of perceived threat severity, containing verbs such as “very/not very severe”, “malicious/benign”, “great disturbance”/“minor annoyance” should be coded as dct:threat_severity_79n2fh4r.

People’s perceptions of their ability to effortful control over dominant responses with respect to the target behaviour.

In a meta-analysis of self-control measures, Duckwort et al. (2011) distinguised four distinct approaches to the measurement of self-control: executive function tasks, delay of gratification tasks, selfreport questionnaires, and informant-report questionnaires.

Among the self- and informant-report questionnaires, over 100 unique instruments were discerned (Duckwort et al., 2011). Items on these questionnaires suggest considerable heterogeneity in the underlying constructs assessed. However, Duckwort et al. (2011) found that urgency (the inability to override strong impulses, e.g., “I have trouble controlling my impulses”), (lack of) premeditation (acting before thinking , e.g., “My thinking is usually careful and purposeful” ), and (lack of) perseverance (the inability to focus on boring or difficult tasks, e.g., “I tend to give up easily”) are three central and related features of self-control. Therefore, when self-constructing an instrument to assess self-control, these three facets of self-control should be included. Moreover, for behaviour change, self-control related to a specific behaviour or object is of greater interest than self-control as a trait (i.e. stable across time and situations). Therefore, in a behaviour change setting, questionnaire items should refer to the specific behaviour or object of interest.

Impulsivity can be regarded as the converse of self-control. Questionnaire items that attempt to measure self-control often present mutually exclusive responses. One response is aligned with long-term goals and standards, while the alternative response is more gratifying or automatic in the short-term. Self-controlled individuals tend to choose the first response, whereas impulsive individuals tend to choose the latter response.

If impulsivity is regarded as the converse of self-control, the same instrument used to assess impulsivity (see dct:impulse_79n2r0t0) can potentially also be used to measure self-control. However, the scoring for each questionnaire item should be reversed, i.e. if a high score on an item indicates a high level of impulsivity, it indicates a low level of self-control. For example: “When it comes to TARGET BEHAVIOUR I act on the spur of the moment” with responses ranging from “Totally agree” (=5) to “Totally disagree” (=1) on a 5-point Likert scale. A high score (=5) here is indicative of a high level of impulsivity and a low level of self-control.

Questions or questionnaires that measure self-control related to a specific behaviour or object. For self-report asessment, a statement such as “When it comes to TARGET BEHAVIOUR….” can be composed, followed by a series of phrases that express behaviour considered indicative for the presence or absence of self-control. For example: “When it comes to smoking cigarettes… 1) I do it without thinking about it; 2) I am self controlled; 3) I don’t manage to plan my behaviour; 4) I act on the spur of the moment; 5) I often suddenly abandon my behavioural intentions; 6) I afterwards frequently regret that I smoked. Responses on these items can be given on a 5-point Likert scale ranging from “totally agree” to “totally disagree”.

Beside behaviour- or object-related self-control, self-control as a trait can be measured. In fact, most self-control instruments measure self-control as a trait (i.e. stable across time and situations). Furthermore, beside selfreport questionnaires, executive function tasks, delay of gratification tasks, and informant-report questionnaires are used to measure self-control. Duckwort et al. (2011) discerned over 100 unique self- and informant-report questionnaires. Items on these questionnaires suggest considerable heterogeneity in the underlying constructs assessed.

If impulsivity is regarded as the converse of self-control, the same instrument to assess impulsivity (see dct:impulse_79n2r0t0) can potentially be used to measure self-control, applying reverse scoring.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use these two questions (with preferably some questions in between): “When it comes to TARGET BEHAVIOUR how deliberate and controlled do you see your actions?” and “When it comes to TARGET BEHAVIOUR how spontaneous do you see your actions?”.

These same questions are used to elicite expressions indicative of impulsivity (see dct:impulse_79n2r0t0).

Expressions that demonstrate or imply the presence of self-control. In the context of behaviour change interventions (i.e. health behaviour changes), this self-control is regarded with respect to a specific behaviour or object.

Trait self-control, i.e. self-control across situations, not specifically linked to a particular behaviour or object, is usually not the focus of interest. Therefore, expressions indicative of self-control should be coded as such only when they concern the behaviour or object of interest, unless beforehand another decision has been made.

Expressions indicative of the presence of self-control may contain phrases such as “I did not succumb to temptation X because I wanted to achieve goal X” and “I sticked to my plan, but sometimes it was difficult”.

Expressions indicative of impulsivity should be coded as dct:impulse_79n2r0t0. Expressions that refer to an individual’s perceptions of the degree to which he/she is capable of, or has control over, performing a given behaviour, should be coded as dct:perceivedBehavioralControl_73dnt603.

People form representations of the world, and some of those representations pertain to the self. This set of attributes together form one’s general self-identity (i.e. unrelated to specific behaviors). However, most of these attributes are relatively trivial (e.g. all humans have a brain, but brains are usually not part of their self-identity). How central a given attribute is to one’s self-identity in general (its centrality) can depend on the degree to which it is perceived to distinguish one from others, confers positive characteristics, or is instrumental. For example, health is seen as desirable, so being ‘a healthy eater’ is desirable. Similarly, if sophistication is seen as desirable, and having well-developed tastes are seen as indicative of sophistication, then being somebody who is very picky about one’s coffee blends and preparation methods can be seen as desirable. As a final example of an instrumental function of self-identity, being ‘an addict’ can be used to excuse relapse.

Also, a given attribute can be more or less related to a given target behavior (e.g. being a ‘vegan’ is intrinsically related to eating behaviors). Given that one’s self-identity is defined as all these attributes that pertain to one’s self (weighted by their centrality), it follows that for different people, a given target behavior can be differentially relevant to their self-identity. In terms of operationalizations, this implies that this relevance of a behavior to one’s self-identity should be what is measured (note that this is not the same as general relevance of a behavior).

As a so-called “direct measure”, the items recommended by Snippe et al. (2020, https://doi.org/10.31234/osf.io/ygpmw) can be used.

According to the definition of self-identity in this DCT, if a target behavior is strongly associated to somebody’s self-identity, to the degree that this is not only a reflection of the frequency or intensity with which they engage in that behavior, this association reflects the fact that one or more attributes is simultaneously connected to the target behavior and central to one’s self-identity.

Therefore, to map the construct content for a given target behavior, for each attribute dimension along which self-identification can occur, two things have to be measured. First, that attribute’s centrality: the degree to which people consider that attribute a part of their self-identity. Second, the strength of that attribute’s association to the target behavior.

These instructions still have to be specified.

These instructions still have to be specified.

These instructions still have to be specified.

Tracking of one’s own actions with regard to a specific behaviour or goal.

Self-monitoring actions with respect to the target behaviour can be assessed in two ways:

1. An individual’s self-reported self-monitoring behaviour can be measured with a questionnnaire or interview. Questionnnaire items to measure self-monitoring with respect to the target behaviour can, for example, be formulated in the following way: “By counting the number of alcoholic drinks, I know exactly how much alcohol I consume during a night clubbing” or “I use a diary or mobile app to track my daily food consumption”. Response can be either dichotomous (yes/no) or three response options can be provided (“Does not apply”, “More or less”, “Does apply”). Social desirability and distorted remembrance of events may result in incongruences between actual and reported self-monitoring behaviour with this asssessment procedure.

2. An alternative way to assess an individual’s self-monitoring behaviour is to ask participants to keep a diary or use a mobile app (e.g. ecological momentary assessments (EMAs) ), internet program, or technical device to keep track of their self-monitoring actions with respect to the target behaviour, e.g. by asking youngsters to make a report each time they are aware to be exposed to music that is too loud (> 100 decibel) during a night out with the help of a mobile app. By asking people to report their self-monitoring behaviour, this behaviour, however, changes and is no longer a reflection of a person’s actual self-monitoring behaviour.

Questions or questionnaires that measure the degree to which an individual engages in self-monitoring with respect to the target behaviour. Items to measure self-reported self-monitoring with respect to a particular (health) behaviour are, for example: “By counting the number of alcoholic drinks, I know exactly how much alcohol I consume during a night clubbing” or “I use a diary or mobile app to track my daily food consumption”. Response can be either dichotomous (yes/no) or three response options can be provided (“Does not apply”, “More or less”, “Does apply”).

Participants may also be asked to keep a diary or use a mobile app, internet program, or technical device to keep track of their self-monitoring actions with respect to the target behaviour, e.g. by asking youngsters to make a report each time they are aware to be exposed to music that is too loud (> 100 decibel) during a night out with the help of a mobile app. However, by asking people to monitor their self-monitoring behaviour, this behaviour changes and is no longer a reflection of a person’s actual self-monitoring behaviour.

Note that self-monitoring as a personal capacity is the topic of interest here. Asking a person (e.g. a patient) to self-monitor his or her (health) behaviour may yield valuable information that other measurement methods (e.g. questionniare, interview) possibly can not provide, e.g. by keeping track of the activity and sleeping patterns of bipolar patients the development of the disease can be predicted and altered. Self-monitoring is also a valuable behaviour change technique; by increasing insight in their actions, people may be instigated to modify their (problem) behaviour. However, as said, self-monitoring as individual skill, and not self-monitoring as research method or behaviour change technique, is of interest here.

The term “self-monitoring” is also used in popular language and research to describe the personality trait along which people differ in their abilities and desires to change their self-presentations, expressive behaviour, and nonverbal affective displays. This is a different construct from the one discussed here and should not be coded as such. The Self Monitoring Scale (Snyder et al, 1974) is a questionnaire that is used to measure this construct.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use the following introduction and question(s): “People differ in the degree they monitor their behaviour”, followed by: “Do you keep an eye on your engagement in TARGET BEHAVIOUR?” (if response is “Yes”:) “How do you keep track of your engagement in TARGET BEHAVIOUR?” (e.g. “How do you keep track of your smoking behaviour?").

Expressions that demonstrate or imply the presence or absence of self-monitoring with regard to the target behaviour. For example: “I weigh my body weight every day” or “I keep a diary to help me understand why I feel stressed at certain moments”.

Note that the term “self-monitoring” is also used in popular language and research to describe the personality trait along which people differ in their abilities and desires to change their self-presentations, expressive behaviour, and nonverbal affective displays. This is a different construct from the one discussed here and should not be coded as such.

Self-standards are preferential self-beliefs (as opposed to descriptive and evaluative ones), which consist of self-imposed criteria for judging oneself. They are self-views that refer to some desired or undesired end-states.

Whether a participant’s self-standards in- or decrease the likelihood the target behaviour is performed, can be assessed by presenting the following two questions (each consisting of six subquestions) to participants:

1. What kind of person would you like to be? Assign a score from 1 to 6 to each of the following six descriptions. Each score is assigned once (i.e. a score of 6 for the description you would most want to be like, and a score of 1 for the description you would least want to be like). [Six descriptions or traits are given - the choice and length of the descriptions is determined by the target behaviour and target population in question. For example: Assertive, Brilliant, Eloquent, Unpredictable, Athletic, and Resourceful.]

2. “Do you think a person that [include one of the six previous descriptions] would engage in TARGET BEHAVIOUR?” For each of the six descriptions the response options are: “Very likely”(=2), “Likely” (=1), “Not Likely or Unlikely” (=0), “Unlikely” (-1), “Very unlikely” (-2).

The answers to question 1 and 2 are multiplied. This results in a maximum score of 42 and a minimum score of -42.

A high score indicates that a participant’s self-standards increase the likelihood that the TARGET BEHAVIOUR is performed, while a low score indicates that a participant’s self-standards decrease the likelihood that the TARGET BEHAVIOUR is performed.

Questions or questionnaires that measure the degree a participant’s self-standards in- or decrease the likelihood of engagement in the target behaviour.

This can, for example, be assessed by presenting the following two questions (each consisting of six subquestions) to participants:

1. What kind of person would you like to be? Assign a score from 1 to 6 to each of the following six descriptions. Each score is assigned once (i.e. a score of 6 for the description you would most want to be like and a score of 1 for the description you would least want to be like). [Six descriptions or traits are given - the choice and length of the descriptions is determined by the target behaviour and target population in question. For example: Assertive, Brilliant, Eloquent, Unpredictable, Athletic, and Resourceful.]

2. “Do you think a person that [include one of the six previous descriptions] would engage in TARGET BEHAVIOUR?” For each of the six descriptions the response options are: “Very likely”(=2), “Likely” (=1), “Not Likely or Unlikely” (=0), ““Unlikely” (-1), “Very unlikely” (-2).

The answers to question 1 and 2 are multiplied. This results in a maximum score of 42 and a minimum score of -42.

A high score indicates that a participant’s self-standards increase the likelihood that the TARGET BEHAVIOUR is performed, while a low score indicates that a participant’s self-standards decrease the likelihood that the TARGET BEHAVIOUR is performed).

Note that most instruments that measure self-standards are not suited to use for the design of behaviour change interventions, as they assess self-standards in relation to self-discrepancy (i.e. the degree of dissimilarity between self-standards and the actual self). Outcomes of these instruments reflect the perceived degree of actualization of an individual’s self-standards. Although self-discrepancy may operate as an incentive for engagement in the target behaviour, the self-standards they relate to are trait-like and, thus, hard to change.

In contrast, an individual’s perceptions of how his or her self-standards relate to the target behaviour may be easier to change and are, thus, a more appropriate goal for behaviour change interventions. Bear in mind that in this latter case the self-standard itself is not changed, ‘solely’ an individual’s perceptions how the self-standard relates to the particular behaviour. For example, the use of role models and other BCTs may help to change these perceptions.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this two consecutive questions: “What kind of person do you want to be?” and “Does this kind of person engage in TARGET BEHAVIOUR?”.

Expressions that specify the relationship between self-standards and the target behaviour. For example: “I think it is very important to be honest. Me secretly drinking does not fit with that ideal” or “I don’t have a problem being the only one wearing ear plugs. Being independent and doing what I believe in, has always been important to me”.

A situation or stimulus that creates a desire to carry out the problem behaviour(s).

Use Likert scales that measure the degree to which participants are temped by a situation or stimulus to carry out the behaviour of interest. For example, participants are asked to indicate if they are temped to take drugs in the following situations: “When I am feeling depressed”, “When I am being offered drugs in a social situation”, “When I am physically tired”, “When I see others using drugs at a bar or a party” with responses “Not at all”, “Not very”, “Moderately”, “Very”, and “Extremely” for each item on a 5-point Likert scale. The sample items are from the Temptation to Use Drugs Scale (Di Clemente et al., 1994).

Another example is provided by Lazuras et al. (2010). They measured the situational tempation among athletes to use doping substances with the stem proposition “How much would you be tempted to use prohibited doping substances to enhance your performance this season,” followed by four items reflecting temptations to use prohibited performance-enhancing substances under different circumstances that potentially induce normative influence (“when your coach suggests so,” “when you believe that most colleagues of yours use prohibited substances,” “when you were told to enhance your performance,” and “when you prepare for an important game/competition”). Responses were recorded on a 5-point Likert scale (1 = not at all tempted, 5 = very much tempted).

Note that the instructions and examples given above apply to the measurement of situational temptation. For the measurement of the susceptibility to tempations in general (i.e. assessing the presence of a personality trait as the causal factor) other instructions and instruments apply, e.g. the Susceptibility to Temptation Scale can be used.

Questions or questionnaires that measure the degree to which participants are temped by a situation or stimulus to carry out the behaviour of interest. For example, participants are asked to indicate if they are temped to take drugs in the following situations: “When I am feeling depressed”, “When I am being offered drugs in a social situation”, “When I am physically tired”, “When I see others using drugs at a bar or a party” with responses “Not at all”, “Not very”, “Moderately”, “Very”, and “Extremely” for each item on a 5-point Likert scale. The sample items are from the Temptation to Use Drugs Scale (Di Clemente et al., 1994).

Another example is provided by Lazuras et al. (2010). They measured the situational temptation among athletes to use doping substances with the stem proposition “How much would you be tempted to use prohibited doping substances to enhance your performance this season,” followed by four items reflecting temptations to use prohibited performance-enhancing substances under different circumstances that potentially induce normative influence (“when your coach suggests so,” “when you believe that most colleagues of yours use prohibited substances,” “when you were told to enhance your performance,” and “when you prepare for an important game/competition”). Responses were recorded on a 5-point Likert scale (1 = not at all tempted, 5 = very much tempted).

Note that the instructions and examples given above apply to the measurement of situational temptation. For the measurement of the susceptibility to temptations in general (i.e. as a trait) other instructions and instruments apply, e.g. the Susceptibility to Temptation Scale can be used.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use this questions: “What situations or circumstances create a desire to perform PROBLEM BEHAVIOUR?”

For example, “What situations or circumstances create a desire to drink alcohol?”

Expressions that describe situations or circumstances that arouse a desire to engage in the behaviour of interest, e.g. expressions such as “when I see somebody drink a beer it is impossible to resist” and “when I see that a bike is not locked, I can not walk by without taking it with me”.

Expressions that describe a susceptibility to temptations in general (i.e. not refering to situations or circumstances but to a personality trait as causal factor), do not describe dct:temptation_79n2fh4w and should not be coded as such.

Beliefs about the significance or magnitude of the threat.

Use 5- or 7-point Likert scales that measure the severity of the (health) threat in question, e.g. " I believe that HEALTH THREAT is severe” and “I believe that HEALTH THREAT has serious negative consequences” with “Strongly disagree” and “Strongly agree” as anchors.

Examples of items measuring the perceived severity of a health threat are: “I believe that HIV infection has serious negative consequences” and “I believe that HIV infection is extremely harmful”, with “Strongly disagree” and “Strongly agree” as anchors on a 5-point Likert scale (examples are from the Risk Behaviour Diagnosis Scale [Witte et al., 1995]).

Questions or questionnaires that measure the perceived severity of the (health) threat in question, e.g. “I believe that HEALTH THREAT is severe” and “I believe that HEALTH THREAT has serious negative consequences” with “Strongly disagree” and “Strongly agree” as anchors on a 5-point Likert scale.

Examples of items measuring the perceived severity of a health threat are: “I believe that HIV infection has serious negative consequences” and “I believe that HIV infection is extremely harmful”, with “Strongly disagree” and “Strongly agree” as anchors on a 5-point Likert scale (examples are from the Risk Behaviour Diagnosis Scale [Witte et al., 1995]).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use the questions: “How would you rate the risk associated with HEALTH THREAT (e.g. TARGET BEHAVIOUR)?” or “Can you describe your personal ideas of the risk associated with HEALTH THREAT (e.g. TARGET BEHAVIOUR)?”.

Expressions that demonstrate or imply the degree of perceived threat severity, containing verbs such as “very/not very severe”, “malicious/benign”, and “great disturbance”/“minor annoyance”.

Expressions that demonstrate or imply the degree of perceived threat susceptibility, containing phrases and words such as “high/low risk”, “high/low prevalence”, “likely/unlikely”, “nobody I know”/“a lot of people I know” (the latter two responses refering to adverse consequences of engaging in the target [risk] behaviour) should be coded as dct:threat_susceptibility_79n2fh4s.

Expressions that illustrate the participant’s risk perception with regard to the health threat without a reference to susceptibility or severity should be coded as dct:risk_perception_79n2fh4t. For example, if the question is “Can you describe your personal ideas of the risk associated with smoking cigarettes?” and the response is “I know there is a risk associated with smoking, so I try not to smoke to much”, no clear reference is made to to the severity of the threat or to personal susceptibility.

Beliefs about one’s risk of experiencing the threat.

Use 5- or 7-point Likert scales that measure the perceived susceptibility to the (health) threat in question, e.g. “I am at risk for getting HEALTH THREAT” and “It is possible that I get HEALTH THREAT” with “Strongly disagree” and “Strongly agree” as anchors.

Examples of items measuring the perceived susceptibility to a health threat are: “It is likely that I will get infected with HIV” and “It is possible that I will get infected with HIV”, with “Strongly disagree” and “Strongly agree” as anchors on a 5-point Likert scale (examples are from the Risk Behaviour Diagnosis Scale [Witte et al., 1995]).

Questions or questionnaires that measure the perceived susceptibility to the (health) threat in question, e.g. “I am at risk for getting HEALTH THREAT” and “It is possible that I get HEALTH THREAT” with “Strongly disagree” and “Strongly agree” as anchors on a 5-point Linkert scale.

Examples of items measuring the perceived susceptibility to a health threat are: “It is likely that I will get infected with HIV” and “It is possible that I will get infected with HIV”, with “Strongly disagree” and “Strongly agree” as anchors on a 5-point Likert scale (examples are from the Risk Behaviour Diagnosis Scale [Witte et al., 1995]).

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, use the questions: “How would you rate the risk associated with HEALTH THREAT (e.g. TARGET BEHAVIOUR)?” or “Can you describe your personal ideas of the risk associated with HEALTH THREAT (e.g. TARGET BEHAVIOUR)?”.

Expressions that demonstrate or imply the degree of perceived threat susceptibility, containing phrases and words such as “high/low risk”, “high/low prevalence”, “likely/unlikely”, and “nobody I know”/“a lot of people I know” (the latter two responses refering to adverse consequences of engaging in the target [risk] behaviour).

Expressions that demonstrate or imply the degree of perceived threat severity, containing verbs such as “very/not very severe”, “malicious/benign”, and “great disturbance”/“minor annoyance” should be coded as dct:threat_severity_79n2fh4r.

Expressions that illustrate the participant’s risk perception with regard to the health threat without a reference to susceptibility or severity should be coded as dct:risk_perception_79n2fh4t. For example, if the question is “Can you describe your personal ideas of the risk associated with smoking cigarettes?” and the response is “I know there is a risk associated with smoking, so I try not to smoke to much” no clear reference is made to to the severity of the threat or to personal susceptibility.

Openness to engaging in a risky behaviour, should the opportunity arise: i.e. it represents a propensity to engage in a risky behaviour without necessarily indicating a plan to engage in that behaviour.

Willingness to engage in a risky behaviour can be measured by asking participants to consider a hypothetical scenario and indicate their willingness to engage in the risk behaviour of interest. Responses on these items range from “Very unwilling” to “Very willing” on a 5- or 7-point Likert scale.

Alternatively, following the scenario description, participants are asked to indicate the likelihood they would engage in the risk behaviour, with repsonses ranging from “Very unlikely” to “Very likely” on a 5- or 7-point Likert scale.

Scores on the different items measuring behavioural willingness can subsequently be summed or averaged.

For example, Molloy et al. (2019) measured willingness to use Nonmedical Prescription Stimulants (NPS) for academic reasons by asking participants to consider various hypothetical scenarios. The first scenario described a friend offering free prescription stimulants to study for a big, upcoming exam. Participants indicated how willing they would be to: (1) Take the pills, (2) Take the pills and ask for extra to use for another exam or paper, or (3) Tell the friend no thank you (reverse-coded).

Davies et al. (2017) measured willingness to drink in a similar fashion, and used three hypothetical scenarios; a family wedding, a house party, and at the park. Participants were asked how likely it was that they would (1) take a drink and drink it, from 1 (“Unlikely”) to 7 (“Highly likely”); and (2) say no thanks, from 1 (“Unlikely”) to 7 (“Likely”) (reverse scored).

Howell et al (2017) measured willingness to engage in risky sun behaviour by asking participants to assume there is no sunscreen, and then to respond how willing they would be to: (1) go to the beach with friends, (2) decline an invitation to go to the beach with friends (reverse coded). Participants responded on a scale ranging from “Very unwilling” (=1) to “Very willing” (=7).

Davies et al. (2017) expressed that a pilot study revealed that use of the verb “willing to” was difficult to understand for young adults. Therefore, (young) participants were asked to indicate the “likelihood” they would engage in a behaviour. Ideally thus the phrasing of items using “willing to” (or similar) should be first tested among the target population in a pilot study.

Willingness to engage in a risky behaviour is usually measured by asking participants to consider a hypothetical scenario and then to indicate their willingness to engage in a particular risk behaviour. Alternatively, following the scenario description, participants are asked to indicate the likelihood they would engage in the risk behaviour. Responses can range from “Very unwilling” to “Very willing” or “Very unlikely” to “Very likely” on a 5- or 7-point Likert scale. Scores on the different items measuring behavioural willingness can subsequently be summed or averaged.

For example, Howell et al (2017) measured willingness to engage in risky sun behaviour by asking participants to assume there is no sunscreen, and then to respond how willing they would be to: (1) go to the beach with friends, (2) decline an invitation to go to the beach with friends (reverse coded). Participants responded on a scale ranging from “Very unwilling” (=1) to “Very willing” (=7).

Fishbein & Ajzen (2011) criticize the use and measurement of behavioural willingness and state that asking people whether they would or would not engage in a certain risk behaviour (e.g. unprotected sex) under certain circumstances may simply measure a more specific intention than the intention to perform these behaviours in general.

Conduct a qualitative study where participants are interviewed, and the interviews are either recorded and transcribed, or notes are kept. These sources are then coded using the instruction for aspect coding.

In this qualitative study, decribe a scenario in which the participant has the opportunity to engage in the risk behaviour of interest and, subsequently, ask the participant how willing he/she would be to engage in the risk behaviour.

For example, if the risk behaviour of interest is “exposure to loud noise (>100 decibel) without hearing protection” the scenario and question can be: “Imagine you are going to a pop concert and you notice the music is very loud, and you forgot to bring ear plugs. How willing would you be to stay at the concert?”.

Expressions that demonstrate or imply the presence or absence of willingness to engage in the risk behaviour. For example using verbs and phrases such as “Very willing”, “Very likely”, “I would probably perform RISK BEHAVIOUR”, etc.

Expressions that refer to explicit intentions or specific plans to engage in (subbehaviours of) the risk behaviour should not be coded as willingness (dct: willingness) but as intention (dct:intention_73dnt604) or action plan (dct:action_plan_79n2w1bh).