Does level of specificity affect measures of motivation to comply? A randomized evaluation

Does level of specificity affect measures of motivation to comply? A randomized evaluation Abstract The theory of planned behavior (TPB) is a popular value-expectancy model in social and behavioral health. Motivation to comply, one of the theory’s constructs, has not been well operationalized and measured in the past, and to date, there has been no assessment of whether level of specificity affects the measurement of the construct. The purpose of this study was to measure the motivation to comply construct across four domains (from general to TACT-behavior specific) and evaluate the potential impact the differences have when identifying determinants of generalized injunctive norms. Students (n = 234) attending a large southwestern university completed a TPB survey related to sleep and physical activity, and were randomized to one of four domains that measured motivation to comply (General domain, n = 58; Health domain, n = 60; Behavioral domain, n = 56; and TACT domain, n = 60). Across both behaviors, motivation to comply measurements did not appear to be affected by changing the level of specificity. Referents for sleep and physical activity were mostly significant, but the effects were small to medium. Future researchers should consider removing motivation to comply measures from TPB surveys to reduce respondent burden or find alternative ways of measuring the construct. Implications Practice: Public health practitioners should be cautioned when evaluating the construct “motivation to comply,” as current best practices are not clear. Policy: Before social norms are included in public health behavior change interventions, curriculum developers should understand that limitations exist in how this construct is operationalized and measured. Research: Future research is needed to explore alternative methods for evaluating the construct “motivation to comply.” INTRODUCTION Value-expectancy models are commonplace in the social and behavioral sciences. Likely the most popular value-expectancy model in health behavior research is the theory of planned behavior (TPB), which was recently updated to the reasoned action approach (RAA) [1, 2]. According to the TPB/RAA, behavioral intentions are the strongest antecedent toward performing a behavior (or action), barring any external barriers. Intentions are in turn determined by one’s attitudes toward a behavior, perceived norms about the behavior, and perceived behavioral control over a behavior [1]. Perceived norms, or the social pressure one feels to engage in (or not engage in) a behavior, consists of two types of normative pressure: injunctive norms, or the perception one has that significant individuals in their life want them to behave in a certain way, and descriptive norms, or the perception one has that a behavior is normal for people like themselves, and thus, should be performed. With regard to measuring injunctive norms (N1), Fishbein and Ajzen [1] recommend that items focus on a set of generalized social agents (i.e., most people). For example, the following item could be used to evaluate injunctive norms for getting an adequate amount of sleep each night: (N1) Most people who are important to me want me to sleep at least 7 hours per night … <Strongly agree/strongly disagree> As noted, while this level of measurement can ultimately indicate an individual’s overall social pressure to engage in behaviors, if a researcher is further interested in understanding which specific individuals or groups (often called referents) are responsible for creating this social pressure, injunctive norms are theorized to be determined by considering multiple injunctive normative beliefs (inbi), or beliefs that important referents want them to perform a specific behavior. For example, the following items could evaluate the strength of each injunctive normative belief (inbi) for the previously mentioned behavior: (inb1) My spouse wants me to sleep at least 7 hours per night … <Strongly agree/strongly disagree> (inb2) My boss wants me to sleep at least 7 hours per night … <Strongly agree/strongly disagree> (inb3) My doctor wants me to sleep at least 7 hours per night … <Strongly agree/strongly disagree> Although this measures the belief strength that one feels toward each referent to perform a specific behavior, it may also be the case that referents do not have an equal status in one’s life. That is to say, you may feel more compelled to listen to your spouse or doctor, compared with your boss. Therefore, it is also recommended to consider one’s motivation to comply (mtci), with the wishes of each referent. This can be evaluated for each referent as: (mtc1) I want to do what my spouse thinks I should do … <Strongly agree/strongly disagree> (mtc2) I want to do what my boss thinks I should do … <Strongly agree/strongly disagree> (mtc2) I want to do what my doctor thinks I should do … <Strongly agree/strongly disagree> As Fishbein and Ajzen [1] recommend, significant normative referents can be identified by evaluating a referents belief strength (inbi) and corresponding mtci, and then the product of these items (inb1 × mtc1; inb2 × mtc2; inb3 × mtc3) can be correlated (r) to a generalized injunctive normative scale (NI). This is illustrated in the following Fig. 1 (more information about this method can be found in Fishbein and Ajzen [1], pages 139–140): Fig 1 View largeDownload slide How determinants of generalized injunctive norms are identified. Fig 1 View largeDownload slide How determinants of generalized injunctive norms are identified. Although there is a great deal of empirical evidence that shows there are valid and reliable methods to evaluate the constructs of the TPB/RAA, one construct of the model, motivation to comply (mtc), has remained problematic. This has been demonstrated in a number of studies. Namely, motivation to comply was not useful for transforming injunctive normative belief strength items in studies that have evaluated women’s intentions to receive a mammogram [3], seat belt use [4], marijuana use [5], and sleep, physical activity, and sugary drink consumption [6]. What may be contributing to the error produced by mtci is the way the construct has been operationalized and measured. To operationalize generalized injunctive norms, Fishbein and Ajzen [1] purport that items follow the principle of compatibility by having items directed toward a TACT-specific behavior (target, action, context, and time). The same principle should also be followed when evaluating inbi. However, for mtci, rather than following the principle of compatibility they suggest evaluating the construct at the general level. We recommend staying at the general level of motivation to comply with a particular referent because once we have assessed a person’s normative belief and behavioral intention, a behavior-specific measure of motivation to comply becomes redundant, adding no unique information. [1] (p. 138) Fishbein and Ajzen [1] note that this recommendation is only based on theoretical reasons, and no study has demonstrated the need to measure mtci at the general level by comparing it with other levels of specificity (i.e., mtci measured at different domains). They also further concede that while measuring mtci at the general level is an indication of a referent’s overall normative power to influence a broad range of behaviors, in some cases, a domain specific measure of mtci might be preferable. However, because a given referent’s power base is likely to vary from domain to domain, it may be preferable to assess domain specific motivation to comply. For example, physicians have expertise in the health domain, but not necessarily in such domains as entertainment or cooking. We might thus be motivated to comply with physicians only when it comes to our health and asking about general motivation to comply might underestimate their influence on our behavior in the health domain. [1] (p. 138) Given the previous problems with evaluating mtci, research is needed to find alternative ways of measuring this construct to improve our understanding of which referents determine one’s injunctive norms toward performing a behavior. Such information will ultimately improve our understanding for how social pressure influences intentions, and ultimately the adoption of health behaviors. Specific information about which referents determine one’s generalized injunctive norms is also imperative for guiding the development of public health and health communication interventions. As mtci is oftentimes measured at the general level, as an alternative, it may be useful to consider evaluating the construct at different levels or domains. This has been proposed by Fishbein and Ajzen [1], but never formally evaluated. Therefore, the purpose of this study was to evaluate mtci measurements across four domains (from general to TACT-behavior specific), and evaluate the potential impact the differences have on belief strength (inbi) when identifying significant referents of generalized injunctive norms. METHODS Participants and procedures Participants attending a large southwestern university were recruited by a mass email sent to all students (18–30 years old) inviting them to participate in an online survey. As an incentive, participants could be entered into a random drawing to win one of five gift cards to a local retail chain in the amount of $10. All research activities were approved by the sponsoring university’s institutional review board (IRB #7862). Participants responded to measures online via Qualtrics. Upon visiting the link to the survey, participants completed an online consent form, and then were randomly assigned to complete one of four different sets of measures. Each set of measures was for two behaviors, sleep and physical activity, and for each behavior, intentions, generalized injunctive norms, injunctive normative belief strength, and a motivation to comply were evaluated. Measures for intentions, generalized injunctive norms, and belief strength were identical between groups. Motivation to comply was different, and was evaluated at either at the general domain, the health domain, the behavioral domain, or the TACT-specific behavioral domain. All measures were created using guidance from Fishbein and Ajzen’s [1] recommended steps for developing TPB/RAA questionnaires. It is recommended that mtc is evaluated in the general domain; therefore, this was considered the reference group. Measures of mtc at the other domains (health, behavioral, TACT) were considered experimental. The survey was also pilot tested with a small group of participants from the target population using Qualtrics (n = 8). Based on this feedback, the items were not significantly changed, as participants indicated that all of the items were clear, and that they were able to understand the differences in the mtc construct at the different domains. Behavioral intention Participants responded to three items to measure their intentions to either engage in the recommended amount of sleep per night (7–9 hr) or physical activity (at least 150 min of moderate activity, 75 min of vigorous activity, or a combination of the two per week). Items included, “I intend to…,” “I plan to…,” and “I will get the recommended amount of moderate or vigorous cardio exercise every week/sleep 7–9 hours every night” (strongly disagree [−3]/strongly agree [3]). The internal consistency reliability was acceptable for both behaviors, across all four groups (physical activity: general domain [α = .95], health domain [α = .90], general behavioral domain [α = .94], and TACT-specific behavioral domain [α = .94]); sleep: general domain [α = .90], health domain [α = .89], general behavioral domain [α = .88], and TACT-specific behavioral domain [α = .86]). Generalized injunctive norms Participants responded to three items to measure their generalized injunctive norms. Items included “Most people who are important to me think I should [do the behavior]” (strongly disagree [−3]/strongly agree [+3]), “Most people whose opinions I value would approve of me [doing the behavior]” (strongly disagree [−3]/strongly agree [+3]), and “It is expected of me that I [do the behavior]” (definitely false [−3]/definitely true [+3]). The internal consistency reliability was acceptable for both behaviors, across all four groups (physical activity: general domain [α = .60], health domain [α = .60], general behavioral domain [α = .60], and TACT-specific behavioral domain [α = .75]; sleep: general domain [α = .68], health domain [α = .70], general behavioral domain [α = .73], and TACT-specific behavioral domain [α = .60]). Injunctive normative belief strength Injunctive normative belief strength was measured for three referents for both physical activity and sleep. Belief strength items started with the phrase “My <referent> think(s) that I should [do the behavior]…” (strongly disagree [1]/strongly agree [7]). Participants were instructed to answer N/A for items not applicable to them. Beliefs for physical activity included (a) parents, (b) friends, and (c) significant others, and beliefs for sleep included (a) parents, (b) friends, and (c) my professors. Motivation to comply A corresponding item measuring motivation to comply was evaluated for each injunctive normative belief item for each condition (all scaled from strongly disagree [−3]/strongly agree [3]). In the first condition (general domain), mtci was evaluated as “In general, I want to do what my <referent> think(s) I should do”). In the second condition (health domain), mtci was evaluated as “For matters related to health, I want to do what my <referent> think(s) I should do”). In the third condition (behavioral domain), mtci was evaluated separately by behavior. For the sleep behavior, items were worded “For matters related to sleep, I want to do what my <referent> think(s) I should do” and for physical activity, items were worded “For matters related to exercise, I want to do what my <referent> think(s) I should do”). In the fourth condition (TACT-specific behavioral domain), mtci was again evaluated separately by behavior. For the sleep behavior, items were worded “When it comes to sleeping 7–9 hours every night, I want to do what my <referent> think(s) I should do” and for physical activity, items were worded “When it comes to getting the recommended amount of moderate or vigorous cardio exercise every week, I want to do what my <referent> think(s) I should do”). In each condition, inbi responses were matched, based on behavior if needed, to mtci responses creating a value-expectancy measure (inbi × mtci). Statistical analysis Intentions and generalized injunctive norms were summated and divided by the number items on the scale, resulting in a mean score between −3 and +3. For each construct, −3 indicated a strong negative or unfavorable predisposition toward the behavior, 0 was a neutral score, and +3 indicated a strong positive or favorable predisposition toward the behavior. Items measuring belief strength (inbi) were scaled from 1 to 7, and corresponding value-based measures (mtca–d) were scaled from −3 to +3. Following standard protocol, corresponding expectancy-based and value-based measures were multiplied to create a composite value-expectancy measure (inbi × mtci). Each value-expectancy based pair (inbi × mtci) was then correlated to the generalized injunctive norm of the corresponding behavior. As demonstrated by the varying sample sizes on Table 2, not all participants responded to each belief/value pair. RESULTS Overall, 234 students participated in this study (general domain [n = 58], health domain [n = 60], behavioral domain [n = 56], and TACT domain [n = 60]). To assure successful randomization, and equivalency between groups, an ANOVA was used to detect differences in key study variables between the four conditions. Table 1 shows that no variables, including age, physical activity intentions and generalized norms, and sleep intentions and generalized norms, were different across the four groups. Participants were traditionally aged college students (general [20.8 years ± 2.53], health [21.2 years ± 2.78], behavioral [20.3 years ± 1.71], and TACT [20.8 years ± 2.28]), and mostly female (general [71%], health [62%], behavioral [71%], and TACT [73%]), and Caucasian (general [68%], health [70%], behavioral [80%], and TACT [73%]). Table 1 A comparison of demographic and study variables Variable Observed range General domain (n = 58) Mean (SD) Health domain (n = 60) Mean (SD) Behavior domain (n = 56) Mean (SD) TACT domain (n = 60) Mean (SD) F statistic p-value Age (years) 18 to 30 20.9 (2.53) 21.2 (2.78) 20.3 (1.73) 20.8 (2.38) 1.473 .223 PA intentions −3 to +3 0.54 (1.88) 0.33 (1.55) 0.67 (1.59) 0.66 (1.63) 0.519 .670 PA injunctive norms −3 to +3 0.95 (1.21) 0.72 (1.19) 1.17 (1.06) 0.82 (1.23) 1.522 .210 Sleep intentions −3 to +3 0.45 (1.75) 0.62 (1.73) 1.22 (1.42) 0.84 (1.45) 2.469 .063 Sleep injunctive norms −3 to +3 1.49 (1.19) 1.20 (1.14) 1.45 (1.12) 1.64 (0.99) 1.642 .181 Variable Observed range General domain (n = 58) Mean (SD) Health domain (n = 60) Mean (SD) Behavior domain (n = 56) Mean (SD) TACT domain (n = 60) Mean (SD) F statistic p-value Age (years) 18 to 30 20.9 (2.53) 21.2 (2.78) 20.3 (1.73) 20.8 (2.38) 1.473 .223 PA intentions −3 to +3 0.54 (1.88) 0.33 (1.55) 0.67 (1.59) 0.66 (1.63) 0.519 .670 PA injunctive norms −3 to +3 0.95 (1.21) 0.72 (1.19) 1.17 (1.06) 0.82 (1.23) 1.522 .210 Sleep intentions −3 to +3 0.45 (1.75) 0.62 (1.73) 1.22 (1.42) 0.84 (1.45) 2.469 .063 Sleep injunctive norms −3 to +3 1.49 (1.19) 1.20 (1.14) 1.45 (1.12) 1.64 (0.99) 1.642 .181 PA physical activity. View Large Table 1 A comparison of demographic and study variables Variable Observed range General domain (n = 58) Mean (SD) Health domain (n = 60) Mean (SD) Behavior domain (n = 56) Mean (SD) TACT domain (n = 60) Mean (SD) F statistic p-value Age (years) 18 to 30 20.9 (2.53) 21.2 (2.78) 20.3 (1.73) 20.8 (2.38) 1.473 .223 PA intentions −3 to +3 0.54 (1.88) 0.33 (1.55) 0.67 (1.59) 0.66 (1.63) 0.519 .670 PA injunctive norms −3 to +3 0.95 (1.21) 0.72 (1.19) 1.17 (1.06) 0.82 (1.23) 1.522 .210 Sleep intentions −3 to +3 0.45 (1.75) 0.62 (1.73) 1.22 (1.42) 0.84 (1.45) 2.469 .063 Sleep injunctive norms −3 to +3 1.49 (1.19) 1.20 (1.14) 1.45 (1.12) 1.64 (0.99) 1.642 .181 Variable Observed range General domain (n = 58) Mean (SD) Health domain (n = 60) Mean (SD) Behavior domain (n = 56) Mean (SD) TACT domain (n = 60) Mean (SD) F statistic p-value Age (years) 18 to 30 20.9 (2.53) 21.2 (2.78) 20.3 (1.73) 20.8 (2.38) 1.473 .223 PA intentions −3 to +3 0.54 (1.88) 0.33 (1.55) 0.67 (1.59) 0.66 (1.63) 0.519 .670 PA injunctive norms −3 to +3 0.95 (1.21) 0.72 (1.19) 1.17 (1.06) 0.82 (1.23) 1.522 .210 Sleep intentions −3 to +3 0.45 (1.75) 0.62 (1.73) 1.22 (1.42) 0.84 (1.45) 2.469 .063 Sleep injunctive norms −3 to +3 1.49 (1.19) 1.20 (1.14) 1.45 (1.12) 1.64 (0.99) 1.642 .181 PA physical activity. View Large Level of specificity for motivation to comply Across both behaviors, motivation to comply measurements did not appear to be affected by changing the level of specificity. The sleep behavior referents were mostly significant, but effects were small to medium (Pearson’s r range across groups: parents [r = .25–.31], friends [r = .22–.41], and professors [r = .26–.41]). This was similar for physical activity (parents [r = .26–.44], friends [r = .26–.41], and significant other [r = .35–.48]). A post hoc analysis, using a dependent t-test, however showed that mtc was significantly different in some instances between the behavioral and TACT-behavior groups. Results showed that in the behavioral group, students had a stronger mtc to their parents for sleeping (0.54 ± 1.86) compared with exercising (0.15 ± 1.83; p = .05; d = 0.21); however, their mtc to their friends was not significantly different (sleep [0.15 ± 1.70]; exercising (−0.27 ± 1.64); p = .10). Similarly, in the TACT-behavioral group, results showed that students had a stronger mtc to their parents for meeting daily sleeping recommendations (1.27 ± 1.86) compared with meeting weekly physical activity recommendations (0.16 ± 1.91; p = .001; d = 0.59); however, their mtc to their friends was not significantly different (sleep [0.34 ± 1.65]; exercising [0.20 ± 1.73]; p = .57). DISCUSSION Understanding determinants of injunctive norms is a critical need for researchers when designing effective public health and health communication interventions that rely on changing social norms. The purpose of this study was to evaluate the measurement of mtci across four domains (from general to TACT-behavior specific) and evaluate the impact level of specificity had when attempting to identify salient determinants of injunctive norms. In this study, level of specificity did not appear to have an impact on identifying determinants of injunctive norms. Results from Table 2 show that overall, referents had a moderate association with generalized injunctive norms, and there were no trends based on how mtci was measured. Table 2 Value-expectancy models for sleep and physical activity Injunctive normative belief (sample size n) Belief strength (inbi  ) (range 1 to 7) Motivation to comply (mtci ) (range −3 to 3) Composite inbi × mtci (range −21 to 21) Correlation with INJ inbi × mtci M SD M SD M SD Sleep behavior Method 1 General domain Parents (58) 6.40 1.03 0.72 1.51 4.88 9.82 .17 Friends (58) 4.98 1.52 0.34 1.21 2.45 5.99 .28* Professor (58) 5.21 1.60 0.50 1.57 3.74 7.83 .41** Method 2 Health domain Parents (60) 6.27 0.94 0.80 1.39 5.18 9.13 .29* Friends (60) 5.13 1.26 0.23 1.24 1.57 6.54 .17 Professor (60) 5.08 1.84 -0.17 1.49 0.37 7.67 .35** Method 3 Behavioral domain Parents (56) 6.38 0.96 0.64 1.79 4.66 11.09 .27* Friends (56) 5.16 1.36 0.16 1.68 2.16 8.97 .41** Professor (56) 5.38 1.75 0.05 1.76 2.07 9.03 .27* Method 4 TACT domain Parents (60) 6.58 0.72 1.35 1.77 9.50 11.68 .29* Friends (60) 4.77 1.37 0.38 1.57 2.78 7.64 .21 Professor (60) 5.22 1.81 0.50 1.99 5.00 9.66 .15 Physical activity behavior Method 1 General domain Parents (58) 5.47 1.60 0.72 1.51 4.47 8.80 .30* Friends (58) 4.60 1.40 0.34 1.21 1.93 5.76 .41** Significant other (26) 5.04 1.61 0.65 1.62 4.85 6.79 .36 Method 2 Health domain Parents (60) 4.80 1.55 0.80 1.39 4.73 6.56 .28* Friends (60) 3.97 1.30 0.23 1.24 1.67 4.93 .25* Significant other (28) 4.89 1.69 1.50 1.34 8.54 8.04 .45** Method 3 Behavioral domain Parents (56) 5.34 1.70 0.14 1.79 2.50 9.90 .35** Friends (56) 4.59 1.35 -0.30 1.54 -0.55 7.60 .24 Significant other (32) 5.03 1.79 0.50 1.87 4.19 9.30 .48** Method 4 TACT domain Parents (60) 5.15 1.64 0.15 1.87 2.45 9.83 .44*** Friends (60) 4.33 1.42 0.18 1.67 1.90 8.00 .38** Significant other (36) 5.00 1.71 1.06 1.79 6.50 9.68 .37* Injunctive normative belief (sample size n) Belief strength (inbi  ) (range 1 to 7) Motivation to comply (mtci ) (range −3 to 3) Composite inbi × mtci (range −21 to 21) Correlation with INJ inbi × mtci M SD M SD M SD Sleep behavior Method 1 General domain Parents (58) 6.40 1.03 0.72 1.51 4.88 9.82 .17 Friends (58) 4.98 1.52 0.34 1.21 2.45 5.99 .28* Professor (58) 5.21 1.60 0.50 1.57 3.74 7.83 .41** Method 2 Health domain Parents (60) 6.27 0.94 0.80 1.39 5.18 9.13 .29* Friends (60) 5.13 1.26 0.23 1.24 1.57 6.54 .17 Professor (60) 5.08 1.84 -0.17 1.49 0.37 7.67 .35** Method 3 Behavioral domain Parents (56) 6.38 0.96 0.64 1.79 4.66 11.09 .27* Friends (56) 5.16 1.36 0.16 1.68 2.16 8.97 .41** Professor (56) 5.38 1.75 0.05 1.76 2.07 9.03 .27* Method 4 TACT domain Parents (60) 6.58 0.72 1.35 1.77 9.50 11.68 .29* Friends (60) 4.77 1.37 0.38 1.57 2.78 7.64 .21 Professor (60) 5.22 1.81 0.50 1.99 5.00 9.66 .15 Physical activity behavior Method 1 General domain Parents (58) 5.47 1.60 0.72 1.51 4.47 8.80 .30* Friends (58) 4.60 1.40 0.34 1.21 1.93 5.76 .41** Significant other (26) 5.04 1.61 0.65 1.62 4.85 6.79 .36 Method 2 Health domain Parents (60) 4.80 1.55 0.80 1.39 4.73 6.56 .28* Friends (60) 3.97 1.30 0.23 1.24 1.67 4.93 .25* Significant other (28) 4.89 1.69 1.50 1.34 8.54 8.04 .45** Method 3 Behavioral domain Parents (56) 5.34 1.70 0.14 1.79 2.50 9.90 .35** Friends (56) 4.59 1.35 -0.30 1.54 -0.55 7.60 .24 Significant other (32) 5.03 1.79 0.50 1.87 4.19 9.30 .48** Method 4 TACT domain Parents (60) 5.15 1.64 0.15 1.87 2.45 9.83 .44*** Friends (60) 4.33 1.42 0.18 1.67 1.90 8.00 .38** Significant other (36) 5.00 1.71 1.06 1.79 6.50 9.68 .37* INJ generalized injunctive norms. Significant at *p < .05; **p < .01; ***p < .001. View Large Table 2 Value-expectancy models for sleep and physical activity Injunctive normative belief (sample size n) Belief strength (inbi  ) (range 1 to 7) Motivation to comply (mtci ) (range −3 to 3) Composite inbi × mtci (range −21 to 21) Correlation with INJ inbi × mtci M SD M SD M SD Sleep behavior Method 1 General domain Parents (58) 6.40 1.03 0.72 1.51 4.88 9.82 .17 Friends (58) 4.98 1.52 0.34 1.21 2.45 5.99 .28* Professor (58) 5.21 1.60 0.50 1.57 3.74 7.83 .41** Method 2 Health domain Parents (60) 6.27 0.94 0.80 1.39 5.18 9.13 .29* Friends (60) 5.13 1.26 0.23 1.24 1.57 6.54 .17 Professor (60) 5.08 1.84 -0.17 1.49 0.37 7.67 .35** Method 3 Behavioral domain Parents (56) 6.38 0.96 0.64 1.79 4.66 11.09 .27* Friends (56) 5.16 1.36 0.16 1.68 2.16 8.97 .41** Professor (56) 5.38 1.75 0.05 1.76 2.07 9.03 .27* Method 4 TACT domain Parents (60) 6.58 0.72 1.35 1.77 9.50 11.68 .29* Friends (60) 4.77 1.37 0.38 1.57 2.78 7.64 .21 Professor (60) 5.22 1.81 0.50 1.99 5.00 9.66 .15 Physical activity behavior Method 1 General domain Parents (58) 5.47 1.60 0.72 1.51 4.47 8.80 .30* Friends (58) 4.60 1.40 0.34 1.21 1.93 5.76 .41** Significant other (26) 5.04 1.61 0.65 1.62 4.85 6.79 .36 Method 2 Health domain Parents (60) 4.80 1.55 0.80 1.39 4.73 6.56 .28* Friends (60) 3.97 1.30 0.23 1.24 1.67 4.93 .25* Significant other (28) 4.89 1.69 1.50 1.34 8.54 8.04 .45** Method 3 Behavioral domain Parents (56) 5.34 1.70 0.14 1.79 2.50 9.90 .35** Friends (56) 4.59 1.35 -0.30 1.54 -0.55 7.60 .24 Significant other (32) 5.03 1.79 0.50 1.87 4.19 9.30 .48** Method 4 TACT domain Parents (60) 5.15 1.64 0.15 1.87 2.45 9.83 .44*** Friends (60) 4.33 1.42 0.18 1.67 1.90 8.00 .38** Significant other (36) 5.00 1.71 1.06 1.79 6.50 9.68 .37* Injunctive normative belief (sample size n) Belief strength (inbi  ) (range 1 to 7) Motivation to comply (mtci ) (range −3 to 3) Composite inbi × mtci (range −21 to 21) Correlation with INJ inbi × mtci M SD M SD M SD Sleep behavior Method 1 General domain Parents (58) 6.40 1.03 0.72 1.51 4.88 9.82 .17 Friends (58) 4.98 1.52 0.34 1.21 2.45 5.99 .28* Professor (58) 5.21 1.60 0.50 1.57 3.74 7.83 .41** Method 2 Health domain Parents (60) 6.27 0.94 0.80 1.39 5.18 9.13 .29* Friends (60) 5.13 1.26 0.23 1.24 1.57 6.54 .17 Professor (60) 5.08 1.84 -0.17 1.49 0.37 7.67 .35** Method 3 Behavioral domain Parents (56) 6.38 0.96 0.64 1.79 4.66 11.09 .27* Friends (56) 5.16 1.36 0.16 1.68 2.16 8.97 .41** Professor (56) 5.38 1.75 0.05 1.76 2.07 9.03 .27* Method 4 TACT domain Parents (60) 6.58 0.72 1.35 1.77 9.50 11.68 .29* Friends (60) 4.77 1.37 0.38 1.57 2.78 7.64 .21 Professor (60) 5.22 1.81 0.50 1.99 5.00 9.66 .15 Physical activity behavior Method 1 General domain Parents (58) 5.47 1.60 0.72 1.51 4.47 8.80 .30* Friends (58) 4.60 1.40 0.34 1.21 1.93 5.76 .41** Significant other (26) 5.04 1.61 0.65 1.62 4.85 6.79 .36 Method 2 Health domain Parents (60) 4.80 1.55 0.80 1.39 4.73 6.56 .28* Friends (60) 3.97 1.30 0.23 1.24 1.67 4.93 .25* Significant other (28) 4.89 1.69 1.50 1.34 8.54 8.04 .45** Method 3 Behavioral domain Parents (56) 5.34 1.70 0.14 1.79 2.50 9.90 .35** Friends (56) 4.59 1.35 -0.30 1.54 -0.55 7.60 .24 Significant other (32) 5.03 1.79 0.50 1.87 4.19 9.30 .48** Method 4 TACT domain Parents (60) 5.15 1.64 0.15 1.87 2.45 9.83 .44*** Friends (60) 4.33 1.42 0.18 1.67 1.90 8.00 .38** Significant other (36) 5.00 1.71 1.06 1.79 6.50 9.68 .37* INJ generalized injunctive norms. Significant at *p < .05; **p < .01; ***p < .001. View Large Level of specificity is one of the most overlooked issues in the literature when measuring mtci. Examples show that researchers have operationalized the construct at many levels (i.e., the general [I want to do what my <referent> thinks I should do] [7]; the behavioral [for general dating behaviors] [8]; and at the TACT-behavioral [adherence to daily airway clearance treatments for cystic fibrosis patients]) [9]. In addition, for some studies, researchers do not give enough information to determine what level of domain mtc was evaluated (i.e., no example items) [10]. The issue we raised related to level of specificity for mtc has been hypothesized in the past by Fishbein and Ajzen [1, 11]; however, this hypothesis has not been empirically tested until now. Another problem with interpreting the existing evidence for mtc is that along with varying levels of specificity, researchers have evaluated the construct in a number of different ways. In 2004, a manual was developed to help health service researchers construct TPB surveys, and suggested an example mtc item as follows [12] (p. 19): My <Referent> approval of my medical practice is important to me. <Not at all/very much>. As an alternative, using Fishbein and Ajzen’s [1] recommendations for measuring mtc, a more appropriate item would have been: In general, I want to do what my <referent> think(s) I should do. From these two versions of the same item, it can be observed that the first item is a measure of one’s attitudes toward a referent’s approval (…is important to me), while the second item is a measure of one’s motivation to comply with a referent (I want to…). Variants of this approach have observed in other studies as well. In a study evaluating TPB measures related to breaking the speed limit, an example item for mtc was [13] (p. 255): I generally like to drive in the way that the police would approve of… <Strongly agree/strongly disagree> Again, this is more similar to one’s attitudes toward driving in a way the police would approve (I generally like to…), rather that one’s motivation to comply with the police’s wishes. Finally, in another study evaluating mothers on their infant-feeding intentions, mtc was evaluated as: In general, how much do you care about what each of the following thinks you should do? Participants were then given a list of referents to rate on a scale from “do not care at all” to “care very much” [14] (p. 661). In this example, the focus of item is how much an individual “cares” (or their attitudes) about complying with referents. Motivation to comply has also been operationalized in ways other than reflecting an attitude. A study evaluating how the TPB can explain sugar-sweetened beverage consumption provided an example item of mtc as: How important is it for you to drink the same amount of sugar-sweetened beverages as your friends do? [15] (p. 174) When examining this item closer, it appears to be measuring one’s attitudes (how important is it…) toward imitating or acting similar to a referent (…for you to drink the same amount of sugar-sweetened beverages as your friends do). The distinction between motivation and an attitude is important to clarify for the measurement of mtc. In Fishbein and Ajzen’s [11] original conceptualization of the TRA, they clearly stated that mtc is a measure of motivation, or intentions, and not attitudes. Perhaps of greater promise is an approach suggesting that motivation to comply can be interpreted as the person’s intention to comply with the referent in question. [11] (p. 366) Interestingly, they further note that mtc has the same determinants as intentions. That is, mtc with a referent is determined by one’s attitude toward complying with a referent, and injunctive norm concerning compliance with the referent [11]. An example measure for generalized measures of attitudes and injunctive norms in this context could be as follows, and future researchers could evaluate relationships among these constructs which could lead to new ways of conceptualizing mtc. However, if measures of attitudes toward compliance and injunctive norms concerning compliance are developed, proper pilot testing would be warranted. Motivation to comply In general, I want to do what my parents think I should do.<Strongly agree/strongly disagree> Attitude toward compliance Doing what my parents think I should do is <good/bad; important/not important> Injunctive norms concerning compliance Most people who are important to me want me to do what my parents think I should do. <Strongly agree/strongly disagree> Motivation to comply In general, I want to do what my parents think I should do.<Strongly agree/strongly disagree> Attitude toward compliance Doing what my parents think I should do is <good/bad; important/not important> Injunctive norms concerning compliance Most people who are important to me want me to do what my parents think I should do. <Strongly agree/strongly disagree> View Large Motivation to comply In general, I want to do what my parents think I should do.<Strongly agree/strongly disagree> Attitude toward compliance Doing what my parents think I should do is <good/bad; important/not important> Injunctive norms concerning compliance Most people who are important to me want me to do what my parents think I should do. <Strongly agree/strongly disagree> Motivation to comply In general, I want to do what my parents think I should do.<Strongly agree/strongly disagree> Attitude toward compliance Doing what my parents think I should do is <good/bad; important/not important> Injunctive norms concerning compliance Most people who are important to me want me to do what my parents think I should do. <Strongly agree/strongly disagree> View Large Perhaps one factor that could be causing researchers and practitioners confusion is that textbook authors that describe the TPB define and operationalize mtc in different ways. For example, while Sharma [16] (p. 104) defines mtc as the “degree to which a person wants to act in accordance with the perceived wishes of those significant in his or her life,” Simons–Morton and colleagues [16] (p. 108) defines it as “how much the actor values the opinions of particular referents with respect to a particular object or behavior.” Simons–Morton and colleagues [17] also label mtc an “ill-named” concept, because they assert mtc should represent how one values the wishes of a referent (attitude), rather than how much an individual wants to comply with a referent (motivation/intention). Clearly, more work is needed to better operationalize mtc, and find best methods for measurement and evaluation. Limitations There are a few limitations to this study that should be addressed. All case studies were based on self-reported data, and therefore have the potential for social desirability and other biases inherent in self-reported data. Data presented were also based on the TPB/RAA as operationalized for health behaviors only, and therefore, results on the usefulness of mtci should not be generalized to all behaviors. Data from this study were also based on a convenience sample; therefore, the generalizability of our results may be limited. It should also be noted that while we recruited from the university mass email system, our sample was largely similar to the student population for race (71.7% of the student body was Caucasian) and age (the average age of the student body was 21.3 years); however, our sample contained more female than male students, while the university had approximately equal representations for gender (50.1% of the student body is female) [18]. Finally, all of the case studies used a cross-sectional design; therefore, nothing can be concluded about the causality between constructs. CONCLUSIONS “Social norms” is one of the most popular, and well-documented, constructs for understanding and explaining human behavior. This is especially true in today’s society with the advent of social media platforms (i.e., Facebook, Twitter, Snapchat) that have revolutionized the way that individuals, communities, and organization communicate and share information, and in turn, influence one another [19]. Many studies have demonstrated the impact social media (through social networking and creating new social norms) has on health behaviors [20–22]; however, as reviewed by Ngia and colleagues [19], more research is needed to understand the broader impact social media has, with one specific focus being on social power and influence. It is therefore paramount that researchers understand how social norms are related to health behaviors, and as such, best methods for evaluating norms should be a priority. In Fishbein and Ajzen’s first book published in 1975, Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research, they called mtc the “least understood” construct of the TRA, and conceded a problem with the construct was that it could be interpreted in different ways [11]. As demonstrated in this article, their notions were correct, in that since the introduction of the TRA, measures of mtc have not been well standardized, and vary from study to study. Clearly, more work is needed to better operationalize mtc and find best methods for measurement and evaluation. Compliance with Ethical Standards Conflicts of Interest: There are no conflicts of interest to report for either author. Ethical Approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors. Informed Consent: Informed consent was obtained from all individual participants included in the study. Acknowledgment Researchers did not receive funding for this study. References 1. Fishbein M , Ajzen I. Predicting and Changing Behavior: The Reasoned Action Approach . New York, NY : Psychology Press ; 2010 . 2. McEachan R , Taylor N , Harrison R , Lawton R , Gardner P , Conner M . Meta-analysis of the Reasoned Action Approach (RAA) to understanding health behaviors . Ann Behav Med . 2016 ; 50 ( 4 ): 592 – 612 . Google Scholar CrossRef Search ADS PubMed 3. Montaño DE , Thompson B , Taylor VM , Mahloch J . Understanding mammography intention and utilization among women in an inner city public hospital clinic . Prev Med . 1997 ; 26 ( 6 ): 817 – 824 . Google Scholar CrossRef Search ADS PubMed 4. Budd RJ , North D , Spencer C . Understanding seat-belt use: a test of Bentler and Speckart’s extension of the “theory of reasoned action.” Eur J Soc Psychol . 1984 ; 14 ( 1 ): 69 – 78 . Google Scholar CrossRef Search ADS 5. Sayeed S , Fishbein M , Hornik R , Cappella J , Kirkland Ahern R . Adolescent marijuana use intentions: using theory to plan an intervention . Drugs (Abingdon Engl) . 2005 ; 12 ( 1 ): 19 – 34 . 6. Branscum P , Collado Rivera M , Fairchild G , Qualls Fay K . Do injunctive and descriptive normative beliefs need a value-laden multiplier in value expectancy models? A case series across multiple health behaviors . Health Behav Res . 2017 ; 1 ( 1 ): 1 – 15 . 7. Blue CL , Wilbur J , Marston-Scott M . Exercise among blue-collar workers: application of the theory of planned behavior . Res Nurs Health . 2001 ; 24 ( 6 ): 481 – 493 . Google Scholar CrossRef Search ADS PubMed 8. Etcheverry PE , Agnew CR . Predictors of motivation to comply with social referents regarding one’s romantic relationship . J Soc Pers Relat . 2016 ; 23 ( 2 ): 214 – 233 . Google Scholar CrossRef Search ADS 9. Grossoehme DH , Szczesniak RD , Mrug S , Dimitriou SM , Marshall A , McPhail GL . Adolescents’ spirituality and cystic fibrosis airway clearance treatment adherence: examining mediators . J Pediatr Psychol . 2016 ; 41 ( 9 ): 1022 – 1032 . Google Scholar CrossRef Search ADS PubMed 10. Ickes MJ , Sharma M . Does behavioral intention predict nutrition behaviors related to adolescent obesity ? ICAN: Infant, Child, & Adolescent Nutrition . 2011 ; 3 ( 1 ): 38 – 48 . Google Scholar CrossRef Search ADS 11. Fishbein M , Ajzen I. Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research . Reading, MA : Addison-Wesley ; 1975 . 12. Francis J , Eccles MP , Johnston M , et al. Constructing questionnaires based on the theory of planned behaviour: a manual for health services researchers . Newcastle upon Tyne, UK : Centre for Health Services Research, University of Newcastle ; 2004 . 13. Conner M , Smith N , McMillan B . Examining normative pressure in the theory of planned behaviour: impact of gender and passengers on intentions to break the speed limit . Curr Psychol . 2003 ; 22 ( 3 ): 252 – 263 . Google Scholar CrossRef Search ADS 14. Manstead AS , Proffitt C , Smart JL . Predicting and understanding mothers’ infant-feeding intentions and behavior: testing the theory of reasoned action . J Pers Soc Psychol . 1983 ; 44 ( 4 ): 657 – 671 . Google Scholar CrossRef Search ADS PubMed 15. Zoellner J , Estabrooks PA , Davy BM , Chen YC , You W . Exploring the theory of planned behavior to explain sugar-sweetened beverage consumption . J Nutr Educ Behav . 2012 ; 44 ( 2 ): 172 – 177 . Google Scholar CrossRef Search ADS PubMed 16. Sharma , M. Theoretical Foundations of Health Education and Health Promotion . Burlington, MA : Jones & Bartlett Publishers ; 2017 . 17. Simons-Morton B , McLeroy KR , Wendel ML. Behavior Theory in Health Promotion Practice and Research . Burlington, MA : Jones & Bartlett Publishers ; 2012 . 18. University of Oklahoma . 2018 . University of Oklahoma Fact Book . Available at http://www.ou.edu/irr/fact-books#norman. Accessibility verified May 2, 2018. 19. Ngai EW , Tao SS , Moon KK . Social media research: theories, constructs, and conceptual frameworks . Inter J of Info Manag . 2015 ; 35 ( 1 ): 33 – 44 . Google Scholar CrossRef Search ADS 20. Perkins HW , Linkenbach JW , Lewis MA , Neighbors C . Effectiveness of social norms media marketing in reducing drinking and driving: a statewide campaign . Addict Behav . 2010 ; 35 ( 10 ): 866 – 874 . Google Scholar CrossRef Search ADS PubMed 21. Vaterlaus JM , Patten EV , Roche C , Young JA . #Gettinghealthy: the perceived influence of social media on young adult health behaviors . Computers in Human Behav . 2015 ; 45 : 151 – 157 . Google Scholar CrossRef Search ADS 22. Hoffman EW , Pinkleton BE , Weintraub Austin E , Reyes-Velázquez W . Exploring college students’ use of general and alcohol-related social media and their associations with alcohol-related behaviors . J Am Coll Health . 2014 ; 62 ( 5 ): 328 – 335 . Google Scholar CrossRef Search ADS PubMed © Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Translational Behavioral Medicine Oxford University Press

Does level of specificity affect measures of motivation to comply? A randomized evaluation

Loading next page...
 
/lp/ou_press/does-level-of-specificity-affect-measures-of-motivation-to-comply-a-zWFcEaosKm
Copyright
© Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
1869-6716
eISSN
1613-9860
D.O.I.
10.1093/tbm/iby056
Publisher site
See Article on Publisher Site

Abstract

Abstract The theory of planned behavior (TPB) is a popular value-expectancy model in social and behavioral health. Motivation to comply, one of the theory’s constructs, has not been well operationalized and measured in the past, and to date, there has been no assessment of whether level of specificity affects the measurement of the construct. The purpose of this study was to measure the motivation to comply construct across four domains (from general to TACT-behavior specific) and evaluate the potential impact the differences have when identifying determinants of generalized injunctive norms. Students (n = 234) attending a large southwestern university completed a TPB survey related to sleep and physical activity, and were randomized to one of four domains that measured motivation to comply (General domain, n = 58; Health domain, n = 60; Behavioral domain, n = 56; and TACT domain, n = 60). Across both behaviors, motivation to comply measurements did not appear to be affected by changing the level of specificity. Referents for sleep and physical activity were mostly significant, but the effects were small to medium. Future researchers should consider removing motivation to comply measures from TPB surveys to reduce respondent burden or find alternative ways of measuring the construct. Implications Practice: Public health practitioners should be cautioned when evaluating the construct “motivation to comply,” as current best practices are not clear. Policy: Before social norms are included in public health behavior change interventions, curriculum developers should understand that limitations exist in how this construct is operationalized and measured. Research: Future research is needed to explore alternative methods for evaluating the construct “motivation to comply.” INTRODUCTION Value-expectancy models are commonplace in the social and behavioral sciences. Likely the most popular value-expectancy model in health behavior research is the theory of planned behavior (TPB), which was recently updated to the reasoned action approach (RAA) [1, 2]. According to the TPB/RAA, behavioral intentions are the strongest antecedent toward performing a behavior (or action), barring any external barriers. Intentions are in turn determined by one’s attitudes toward a behavior, perceived norms about the behavior, and perceived behavioral control over a behavior [1]. Perceived norms, or the social pressure one feels to engage in (or not engage in) a behavior, consists of two types of normative pressure: injunctive norms, or the perception one has that significant individuals in their life want them to behave in a certain way, and descriptive norms, or the perception one has that a behavior is normal for people like themselves, and thus, should be performed. With regard to measuring injunctive norms (N1), Fishbein and Ajzen [1] recommend that items focus on a set of generalized social agents (i.e., most people). For example, the following item could be used to evaluate injunctive norms for getting an adequate amount of sleep each night: (N1) Most people who are important to me want me to sleep at least 7 hours per night … <Strongly agree/strongly disagree> As noted, while this level of measurement can ultimately indicate an individual’s overall social pressure to engage in behaviors, if a researcher is further interested in understanding which specific individuals or groups (often called referents) are responsible for creating this social pressure, injunctive norms are theorized to be determined by considering multiple injunctive normative beliefs (inbi), or beliefs that important referents want them to perform a specific behavior. For example, the following items could evaluate the strength of each injunctive normative belief (inbi) for the previously mentioned behavior: (inb1) My spouse wants me to sleep at least 7 hours per night … <Strongly agree/strongly disagree> (inb2) My boss wants me to sleep at least 7 hours per night … <Strongly agree/strongly disagree> (inb3) My doctor wants me to sleep at least 7 hours per night … <Strongly agree/strongly disagree> Although this measures the belief strength that one feels toward each referent to perform a specific behavior, it may also be the case that referents do not have an equal status in one’s life. That is to say, you may feel more compelled to listen to your spouse or doctor, compared with your boss. Therefore, it is also recommended to consider one’s motivation to comply (mtci), with the wishes of each referent. This can be evaluated for each referent as: (mtc1) I want to do what my spouse thinks I should do … <Strongly agree/strongly disagree> (mtc2) I want to do what my boss thinks I should do … <Strongly agree/strongly disagree> (mtc2) I want to do what my doctor thinks I should do … <Strongly agree/strongly disagree> As Fishbein and Ajzen [1] recommend, significant normative referents can be identified by evaluating a referents belief strength (inbi) and corresponding mtci, and then the product of these items (inb1 × mtc1; inb2 × mtc2; inb3 × mtc3) can be correlated (r) to a generalized injunctive normative scale (NI). This is illustrated in the following Fig. 1 (more information about this method can be found in Fishbein and Ajzen [1], pages 139–140): Fig 1 View largeDownload slide How determinants of generalized injunctive norms are identified. Fig 1 View largeDownload slide How determinants of generalized injunctive norms are identified. Although there is a great deal of empirical evidence that shows there are valid and reliable methods to evaluate the constructs of the TPB/RAA, one construct of the model, motivation to comply (mtc), has remained problematic. This has been demonstrated in a number of studies. Namely, motivation to comply was not useful for transforming injunctive normative belief strength items in studies that have evaluated women’s intentions to receive a mammogram [3], seat belt use [4], marijuana use [5], and sleep, physical activity, and sugary drink consumption [6]. What may be contributing to the error produced by mtci is the way the construct has been operationalized and measured. To operationalize generalized injunctive norms, Fishbein and Ajzen [1] purport that items follow the principle of compatibility by having items directed toward a TACT-specific behavior (target, action, context, and time). The same principle should also be followed when evaluating inbi. However, for mtci, rather than following the principle of compatibility they suggest evaluating the construct at the general level. We recommend staying at the general level of motivation to comply with a particular referent because once we have assessed a person’s normative belief and behavioral intention, a behavior-specific measure of motivation to comply becomes redundant, adding no unique information. [1] (p. 138) Fishbein and Ajzen [1] note that this recommendation is only based on theoretical reasons, and no study has demonstrated the need to measure mtci at the general level by comparing it with other levels of specificity (i.e., mtci measured at different domains). They also further concede that while measuring mtci at the general level is an indication of a referent’s overall normative power to influence a broad range of behaviors, in some cases, a domain specific measure of mtci might be preferable. However, because a given referent’s power base is likely to vary from domain to domain, it may be preferable to assess domain specific motivation to comply. For example, physicians have expertise in the health domain, but not necessarily in such domains as entertainment or cooking. We might thus be motivated to comply with physicians only when it comes to our health and asking about general motivation to comply might underestimate their influence on our behavior in the health domain. [1] (p. 138) Given the previous problems with evaluating mtci, research is needed to find alternative ways of measuring this construct to improve our understanding of which referents determine one’s injunctive norms toward performing a behavior. Such information will ultimately improve our understanding for how social pressure influences intentions, and ultimately the adoption of health behaviors. Specific information about which referents determine one’s generalized injunctive norms is also imperative for guiding the development of public health and health communication interventions. As mtci is oftentimes measured at the general level, as an alternative, it may be useful to consider evaluating the construct at different levels or domains. This has been proposed by Fishbein and Ajzen [1], but never formally evaluated. Therefore, the purpose of this study was to evaluate mtci measurements across four domains (from general to TACT-behavior specific), and evaluate the potential impact the differences have on belief strength (inbi) when identifying significant referents of generalized injunctive norms. METHODS Participants and procedures Participants attending a large southwestern university were recruited by a mass email sent to all students (18–30 years old) inviting them to participate in an online survey. As an incentive, participants could be entered into a random drawing to win one of five gift cards to a local retail chain in the amount of $10. All research activities were approved by the sponsoring university’s institutional review board (IRB #7862). Participants responded to measures online via Qualtrics. Upon visiting the link to the survey, participants completed an online consent form, and then were randomly assigned to complete one of four different sets of measures. Each set of measures was for two behaviors, sleep and physical activity, and for each behavior, intentions, generalized injunctive norms, injunctive normative belief strength, and a motivation to comply were evaluated. Measures for intentions, generalized injunctive norms, and belief strength were identical between groups. Motivation to comply was different, and was evaluated at either at the general domain, the health domain, the behavioral domain, or the TACT-specific behavioral domain. All measures were created using guidance from Fishbein and Ajzen’s [1] recommended steps for developing TPB/RAA questionnaires. It is recommended that mtc is evaluated in the general domain; therefore, this was considered the reference group. Measures of mtc at the other domains (health, behavioral, TACT) were considered experimental. The survey was also pilot tested with a small group of participants from the target population using Qualtrics (n = 8). Based on this feedback, the items were not significantly changed, as participants indicated that all of the items were clear, and that they were able to understand the differences in the mtc construct at the different domains. Behavioral intention Participants responded to three items to measure their intentions to either engage in the recommended amount of sleep per night (7–9 hr) or physical activity (at least 150 min of moderate activity, 75 min of vigorous activity, or a combination of the two per week). Items included, “I intend to…,” “I plan to…,” and “I will get the recommended amount of moderate or vigorous cardio exercise every week/sleep 7–9 hours every night” (strongly disagree [−3]/strongly agree [3]). The internal consistency reliability was acceptable for both behaviors, across all four groups (physical activity: general domain [α = .95], health domain [α = .90], general behavioral domain [α = .94], and TACT-specific behavioral domain [α = .94]); sleep: general domain [α = .90], health domain [α = .89], general behavioral domain [α = .88], and TACT-specific behavioral domain [α = .86]). Generalized injunctive norms Participants responded to three items to measure their generalized injunctive norms. Items included “Most people who are important to me think I should [do the behavior]” (strongly disagree [−3]/strongly agree [+3]), “Most people whose opinions I value would approve of me [doing the behavior]” (strongly disagree [−3]/strongly agree [+3]), and “It is expected of me that I [do the behavior]” (definitely false [−3]/definitely true [+3]). The internal consistency reliability was acceptable for both behaviors, across all four groups (physical activity: general domain [α = .60], health domain [α = .60], general behavioral domain [α = .60], and TACT-specific behavioral domain [α = .75]; sleep: general domain [α = .68], health domain [α = .70], general behavioral domain [α = .73], and TACT-specific behavioral domain [α = .60]). Injunctive normative belief strength Injunctive normative belief strength was measured for three referents for both physical activity and sleep. Belief strength items started with the phrase “My <referent> think(s) that I should [do the behavior]…” (strongly disagree [1]/strongly agree [7]). Participants were instructed to answer N/A for items not applicable to them. Beliefs for physical activity included (a) parents, (b) friends, and (c) significant others, and beliefs for sleep included (a) parents, (b) friends, and (c) my professors. Motivation to comply A corresponding item measuring motivation to comply was evaluated for each injunctive normative belief item for each condition (all scaled from strongly disagree [−3]/strongly agree [3]). In the first condition (general domain), mtci was evaluated as “In general, I want to do what my <referent> think(s) I should do”). In the second condition (health domain), mtci was evaluated as “For matters related to health, I want to do what my <referent> think(s) I should do”). In the third condition (behavioral domain), mtci was evaluated separately by behavior. For the sleep behavior, items were worded “For matters related to sleep, I want to do what my <referent> think(s) I should do” and for physical activity, items were worded “For matters related to exercise, I want to do what my <referent> think(s) I should do”). In the fourth condition (TACT-specific behavioral domain), mtci was again evaluated separately by behavior. For the sleep behavior, items were worded “When it comes to sleeping 7–9 hours every night, I want to do what my <referent> think(s) I should do” and for physical activity, items were worded “When it comes to getting the recommended amount of moderate or vigorous cardio exercise every week, I want to do what my <referent> think(s) I should do”). In each condition, inbi responses were matched, based on behavior if needed, to mtci responses creating a value-expectancy measure (inbi × mtci). Statistical analysis Intentions and generalized injunctive norms were summated and divided by the number items on the scale, resulting in a mean score between −3 and +3. For each construct, −3 indicated a strong negative or unfavorable predisposition toward the behavior, 0 was a neutral score, and +3 indicated a strong positive or favorable predisposition toward the behavior. Items measuring belief strength (inbi) were scaled from 1 to 7, and corresponding value-based measures (mtca–d) were scaled from −3 to +3. Following standard protocol, corresponding expectancy-based and value-based measures were multiplied to create a composite value-expectancy measure (inbi × mtci). Each value-expectancy based pair (inbi × mtci) was then correlated to the generalized injunctive norm of the corresponding behavior. As demonstrated by the varying sample sizes on Table 2, not all participants responded to each belief/value pair. RESULTS Overall, 234 students participated in this study (general domain [n = 58], health domain [n = 60], behavioral domain [n = 56], and TACT domain [n = 60]). To assure successful randomization, and equivalency between groups, an ANOVA was used to detect differences in key study variables between the four conditions. Table 1 shows that no variables, including age, physical activity intentions and generalized norms, and sleep intentions and generalized norms, were different across the four groups. Participants were traditionally aged college students (general [20.8 years ± 2.53], health [21.2 years ± 2.78], behavioral [20.3 years ± 1.71], and TACT [20.8 years ± 2.28]), and mostly female (general [71%], health [62%], behavioral [71%], and TACT [73%]), and Caucasian (general [68%], health [70%], behavioral [80%], and TACT [73%]). Table 1 A comparison of demographic and study variables Variable Observed range General domain (n = 58) Mean (SD) Health domain (n = 60) Mean (SD) Behavior domain (n = 56) Mean (SD) TACT domain (n = 60) Mean (SD) F statistic p-value Age (years) 18 to 30 20.9 (2.53) 21.2 (2.78) 20.3 (1.73) 20.8 (2.38) 1.473 .223 PA intentions −3 to +3 0.54 (1.88) 0.33 (1.55) 0.67 (1.59) 0.66 (1.63) 0.519 .670 PA injunctive norms −3 to +3 0.95 (1.21) 0.72 (1.19) 1.17 (1.06) 0.82 (1.23) 1.522 .210 Sleep intentions −3 to +3 0.45 (1.75) 0.62 (1.73) 1.22 (1.42) 0.84 (1.45) 2.469 .063 Sleep injunctive norms −3 to +3 1.49 (1.19) 1.20 (1.14) 1.45 (1.12) 1.64 (0.99) 1.642 .181 Variable Observed range General domain (n = 58) Mean (SD) Health domain (n = 60) Mean (SD) Behavior domain (n = 56) Mean (SD) TACT domain (n = 60) Mean (SD) F statistic p-value Age (years) 18 to 30 20.9 (2.53) 21.2 (2.78) 20.3 (1.73) 20.8 (2.38) 1.473 .223 PA intentions −3 to +3 0.54 (1.88) 0.33 (1.55) 0.67 (1.59) 0.66 (1.63) 0.519 .670 PA injunctive norms −3 to +3 0.95 (1.21) 0.72 (1.19) 1.17 (1.06) 0.82 (1.23) 1.522 .210 Sleep intentions −3 to +3 0.45 (1.75) 0.62 (1.73) 1.22 (1.42) 0.84 (1.45) 2.469 .063 Sleep injunctive norms −3 to +3 1.49 (1.19) 1.20 (1.14) 1.45 (1.12) 1.64 (0.99) 1.642 .181 PA physical activity. View Large Table 1 A comparison of demographic and study variables Variable Observed range General domain (n = 58) Mean (SD) Health domain (n = 60) Mean (SD) Behavior domain (n = 56) Mean (SD) TACT domain (n = 60) Mean (SD) F statistic p-value Age (years) 18 to 30 20.9 (2.53) 21.2 (2.78) 20.3 (1.73) 20.8 (2.38) 1.473 .223 PA intentions −3 to +3 0.54 (1.88) 0.33 (1.55) 0.67 (1.59) 0.66 (1.63) 0.519 .670 PA injunctive norms −3 to +3 0.95 (1.21) 0.72 (1.19) 1.17 (1.06) 0.82 (1.23) 1.522 .210 Sleep intentions −3 to +3 0.45 (1.75) 0.62 (1.73) 1.22 (1.42) 0.84 (1.45) 2.469 .063 Sleep injunctive norms −3 to +3 1.49 (1.19) 1.20 (1.14) 1.45 (1.12) 1.64 (0.99) 1.642 .181 Variable Observed range General domain (n = 58) Mean (SD) Health domain (n = 60) Mean (SD) Behavior domain (n = 56) Mean (SD) TACT domain (n = 60) Mean (SD) F statistic p-value Age (years) 18 to 30 20.9 (2.53) 21.2 (2.78) 20.3 (1.73) 20.8 (2.38) 1.473 .223 PA intentions −3 to +3 0.54 (1.88) 0.33 (1.55) 0.67 (1.59) 0.66 (1.63) 0.519 .670 PA injunctive norms −3 to +3 0.95 (1.21) 0.72 (1.19) 1.17 (1.06) 0.82 (1.23) 1.522 .210 Sleep intentions −3 to +3 0.45 (1.75) 0.62 (1.73) 1.22 (1.42) 0.84 (1.45) 2.469 .063 Sleep injunctive norms −3 to +3 1.49 (1.19) 1.20 (1.14) 1.45 (1.12) 1.64 (0.99) 1.642 .181 PA physical activity. View Large Level of specificity for motivation to comply Across both behaviors, motivation to comply measurements did not appear to be affected by changing the level of specificity. The sleep behavior referents were mostly significant, but effects were small to medium (Pearson’s r range across groups: parents [r = .25–.31], friends [r = .22–.41], and professors [r = .26–.41]). This was similar for physical activity (parents [r = .26–.44], friends [r = .26–.41], and significant other [r = .35–.48]). A post hoc analysis, using a dependent t-test, however showed that mtc was significantly different in some instances between the behavioral and TACT-behavior groups. Results showed that in the behavioral group, students had a stronger mtc to their parents for sleeping (0.54 ± 1.86) compared with exercising (0.15 ± 1.83; p = .05; d = 0.21); however, their mtc to their friends was not significantly different (sleep [0.15 ± 1.70]; exercising (−0.27 ± 1.64); p = .10). Similarly, in the TACT-behavioral group, results showed that students had a stronger mtc to their parents for meeting daily sleeping recommendations (1.27 ± 1.86) compared with meeting weekly physical activity recommendations (0.16 ± 1.91; p = .001; d = 0.59); however, their mtc to their friends was not significantly different (sleep [0.34 ± 1.65]; exercising [0.20 ± 1.73]; p = .57). DISCUSSION Understanding determinants of injunctive norms is a critical need for researchers when designing effective public health and health communication interventions that rely on changing social norms. The purpose of this study was to evaluate the measurement of mtci across four domains (from general to TACT-behavior specific) and evaluate the impact level of specificity had when attempting to identify salient determinants of injunctive norms. In this study, level of specificity did not appear to have an impact on identifying determinants of injunctive norms. Results from Table 2 show that overall, referents had a moderate association with generalized injunctive norms, and there were no trends based on how mtci was measured. Table 2 Value-expectancy models for sleep and physical activity Injunctive normative belief (sample size n) Belief strength (inbi  ) (range 1 to 7) Motivation to comply (mtci ) (range −3 to 3) Composite inbi × mtci (range −21 to 21) Correlation with INJ inbi × mtci M SD M SD M SD Sleep behavior Method 1 General domain Parents (58) 6.40 1.03 0.72 1.51 4.88 9.82 .17 Friends (58) 4.98 1.52 0.34 1.21 2.45 5.99 .28* Professor (58) 5.21 1.60 0.50 1.57 3.74 7.83 .41** Method 2 Health domain Parents (60) 6.27 0.94 0.80 1.39 5.18 9.13 .29* Friends (60) 5.13 1.26 0.23 1.24 1.57 6.54 .17 Professor (60) 5.08 1.84 -0.17 1.49 0.37 7.67 .35** Method 3 Behavioral domain Parents (56) 6.38 0.96 0.64 1.79 4.66 11.09 .27* Friends (56) 5.16 1.36 0.16 1.68 2.16 8.97 .41** Professor (56) 5.38 1.75 0.05 1.76 2.07 9.03 .27* Method 4 TACT domain Parents (60) 6.58 0.72 1.35 1.77 9.50 11.68 .29* Friends (60) 4.77 1.37 0.38 1.57 2.78 7.64 .21 Professor (60) 5.22 1.81 0.50 1.99 5.00 9.66 .15 Physical activity behavior Method 1 General domain Parents (58) 5.47 1.60 0.72 1.51 4.47 8.80 .30* Friends (58) 4.60 1.40 0.34 1.21 1.93 5.76 .41** Significant other (26) 5.04 1.61 0.65 1.62 4.85 6.79 .36 Method 2 Health domain Parents (60) 4.80 1.55 0.80 1.39 4.73 6.56 .28* Friends (60) 3.97 1.30 0.23 1.24 1.67 4.93 .25* Significant other (28) 4.89 1.69 1.50 1.34 8.54 8.04 .45** Method 3 Behavioral domain Parents (56) 5.34 1.70 0.14 1.79 2.50 9.90 .35** Friends (56) 4.59 1.35 -0.30 1.54 -0.55 7.60 .24 Significant other (32) 5.03 1.79 0.50 1.87 4.19 9.30 .48** Method 4 TACT domain Parents (60) 5.15 1.64 0.15 1.87 2.45 9.83 .44*** Friends (60) 4.33 1.42 0.18 1.67 1.90 8.00 .38** Significant other (36) 5.00 1.71 1.06 1.79 6.50 9.68 .37* Injunctive normative belief (sample size n) Belief strength (inbi  ) (range 1 to 7) Motivation to comply (mtci ) (range −3 to 3) Composite inbi × mtci (range −21 to 21) Correlation with INJ inbi × mtci M SD M SD M SD Sleep behavior Method 1 General domain Parents (58) 6.40 1.03 0.72 1.51 4.88 9.82 .17 Friends (58) 4.98 1.52 0.34 1.21 2.45 5.99 .28* Professor (58) 5.21 1.60 0.50 1.57 3.74 7.83 .41** Method 2 Health domain Parents (60) 6.27 0.94 0.80 1.39 5.18 9.13 .29* Friends (60) 5.13 1.26 0.23 1.24 1.57 6.54 .17 Professor (60) 5.08 1.84 -0.17 1.49 0.37 7.67 .35** Method 3 Behavioral domain Parents (56) 6.38 0.96 0.64 1.79 4.66 11.09 .27* Friends (56) 5.16 1.36 0.16 1.68 2.16 8.97 .41** Professor (56) 5.38 1.75 0.05 1.76 2.07 9.03 .27* Method 4 TACT domain Parents (60) 6.58 0.72 1.35 1.77 9.50 11.68 .29* Friends (60) 4.77 1.37 0.38 1.57 2.78 7.64 .21 Professor (60) 5.22 1.81 0.50 1.99 5.00 9.66 .15 Physical activity behavior Method 1 General domain Parents (58) 5.47 1.60 0.72 1.51 4.47 8.80 .30* Friends (58) 4.60 1.40 0.34 1.21 1.93 5.76 .41** Significant other (26) 5.04 1.61 0.65 1.62 4.85 6.79 .36 Method 2 Health domain Parents (60) 4.80 1.55 0.80 1.39 4.73 6.56 .28* Friends (60) 3.97 1.30 0.23 1.24 1.67 4.93 .25* Significant other (28) 4.89 1.69 1.50 1.34 8.54 8.04 .45** Method 3 Behavioral domain Parents (56) 5.34 1.70 0.14 1.79 2.50 9.90 .35** Friends (56) 4.59 1.35 -0.30 1.54 -0.55 7.60 .24 Significant other (32) 5.03 1.79 0.50 1.87 4.19 9.30 .48** Method 4 TACT domain Parents (60) 5.15 1.64 0.15 1.87 2.45 9.83 .44*** Friends (60) 4.33 1.42 0.18 1.67 1.90 8.00 .38** Significant other (36) 5.00 1.71 1.06 1.79 6.50 9.68 .37* INJ generalized injunctive norms. Significant at *p < .05; **p < .01; ***p < .001. View Large Table 2 Value-expectancy models for sleep and physical activity Injunctive normative belief (sample size n) Belief strength (inbi  ) (range 1 to 7) Motivation to comply (mtci ) (range −3 to 3) Composite inbi × mtci (range −21 to 21) Correlation with INJ inbi × mtci M SD M SD M SD Sleep behavior Method 1 General domain Parents (58) 6.40 1.03 0.72 1.51 4.88 9.82 .17 Friends (58) 4.98 1.52 0.34 1.21 2.45 5.99 .28* Professor (58) 5.21 1.60 0.50 1.57 3.74 7.83 .41** Method 2 Health domain Parents (60) 6.27 0.94 0.80 1.39 5.18 9.13 .29* Friends (60) 5.13 1.26 0.23 1.24 1.57 6.54 .17 Professor (60) 5.08 1.84 -0.17 1.49 0.37 7.67 .35** Method 3 Behavioral domain Parents (56) 6.38 0.96 0.64 1.79 4.66 11.09 .27* Friends (56) 5.16 1.36 0.16 1.68 2.16 8.97 .41** Professor (56) 5.38 1.75 0.05 1.76 2.07 9.03 .27* Method 4 TACT domain Parents (60) 6.58 0.72 1.35 1.77 9.50 11.68 .29* Friends (60) 4.77 1.37 0.38 1.57 2.78 7.64 .21 Professor (60) 5.22 1.81 0.50 1.99 5.00 9.66 .15 Physical activity behavior Method 1 General domain Parents (58) 5.47 1.60 0.72 1.51 4.47 8.80 .30* Friends (58) 4.60 1.40 0.34 1.21 1.93 5.76 .41** Significant other (26) 5.04 1.61 0.65 1.62 4.85 6.79 .36 Method 2 Health domain Parents (60) 4.80 1.55 0.80 1.39 4.73 6.56 .28* Friends (60) 3.97 1.30 0.23 1.24 1.67 4.93 .25* Significant other (28) 4.89 1.69 1.50 1.34 8.54 8.04 .45** Method 3 Behavioral domain Parents (56) 5.34 1.70 0.14 1.79 2.50 9.90 .35** Friends (56) 4.59 1.35 -0.30 1.54 -0.55 7.60 .24 Significant other (32) 5.03 1.79 0.50 1.87 4.19 9.30 .48** Method 4 TACT domain Parents (60) 5.15 1.64 0.15 1.87 2.45 9.83 .44*** Friends (60) 4.33 1.42 0.18 1.67 1.90 8.00 .38** Significant other (36) 5.00 1.71 1.06 1.79 6.50 9.68 .37* Injunctive normative belief (sample size n) Belief strength (inbi  ) (range 1 to 7) Motivation to comply (mtci ) (range −3 to 3) Composite inbi × mtci (range −21 to 21) Correlation with INJ inbi × mtci M SD M SD M SD Sleep behavior Method 1 General domain Parents (58) 6.40 1.03 0.72 1.51 4.88 9.82 .17 Friends (58) 4.98 1.52 0.34 1.21 2.45 5.99 .28* Professor (58) 5.21 1.60 0.50 1.57 3.74 7.83 .41** Method 2 Health domain Parents (60) 6.27 0.94 0.80 1.39 5.18 9.13 .29* Friends (60) 5.13 1.26 0.23 1.24 1.57 6.54 .17 Professor (60) 5.08 1.84 -0.17 1.49 0.37 7.67 .35** Method 3 Behavioral domain Parents (56) 6.38 0.96 0.64 1.79 4.66 11.09 .27* Friends (56) 5.16 1.36 0.16 1.68 2.16 8.97 .41** Professor (56) 5.38 1.75 0.05 1.76 2.07 9.03 .27* Method 4 TACT domain Parents (60) 6.58 0.72 1.35 1.77 9.50 11.68 .29* Friends (60) 4.77 1.37 0.38 1.57 2.78 7.64 .21 Professor (60) 5.22 1.81 0.50 1.99 5.00 9.66 .15 Physical activity behavior Method 1 General domain Parents (58) 5.47 1.60 0.72 1.51 4.47 8.80 .30* Friends (58) 4.60 1.40 0.34 1.21 1.93 5.76 .41** Significant other (26) 5.04 1.61 0.65 1.62 4.85 6.79 .36 Method 2 Health domain Parents (60) 4.80 1.55 0.80 1.39 4.73 6.56 .28* Friends (60) 3.97 1.30 0.23 1.24 1.67 4.93 .25* Significant other (28) 4.89 1.69 1.50 1.34 8.54 8.04 .45** Method 3 Behavioral domain Parents (56) 5.34 1.70 0.14 1.79 2.50 9.90 .35** Friends (56) 4.59 1.35 -0.30 1.54 -0.55 7.60 .24 Significant other (32) 5.03 1.79 0.50 1.87 4.19 9.30 .48** Method 4 TACT domain Parents (60) 5.15 1.64 0.15 1.87 2.45 9.83 .44*** Friends (60) 4.33 1.42 0.18 1.67 1.90 8.00 .38** Significant other (36) 5.00 1.71 1.06 1.79 6.50 9.68 .37* INJ generalized injunctive norms. Significant at *p < .05; **p < .01; ***p < .001. View Large Level of specificity is one of the most overlooked issues in the literature when measuring mtci. Examples show that researchers have operationalized the construct at many levels (i.e., the general [I want to do what my <referent> thinks I should do] [7]; the behavioral [for general dating behaviors] [8]; and at the TACT-behavioral [adherence to daily airway clearance treatments for cystic fibrosis patients]) [9]. In addition, for some studies, researchers do not give enough information to determine what level of domain mtc was evaluated (i.e., no example items) [10]. The issue we raised related to level of specificity for mtc has been hypothesized in the past by Fishbein and Ajzen [1, 11]; however, this hypothesis has not been empirically tested until now. Another problem with interpreting the existing evidence for mtc is that along with varying levels of specificity, researchers have evaluated the construct in a number of different ways. In 2004, a manual was developed to help health service researchers construct TPB surveys, and suggested an example mtc item as follows [12] (p. 19): My <Referent> approval of my medical practice is important to me. <Not at all/very much>. As an alternative, using Fishbein and Ajzen’s [1] recommendations for measuring mtc, a more appropriate item would have been: In general, I want to do what my <referent> think(s) I should do. From these two versions of the same item, it can be observed that the first item is a measure of one’s attitudes toward a referent’s approval (…is important to me), while the second item is a measure of one’s motivation to comply with a referent (I want to…). Variants of this approach have observed in other studies as well. In a study evaluating TPB measures related to breaking the speed limit, an example item for mtc was [13] (p. 255): I generally like to drive in the way that the police would approve of… <Strongly agree/strongly disagree> Again, this is more similar to one’s attitudes toward driving in a way the police would approve (I generally like to…), rather that one’s motivation to comply with the police’s wishes. Finally, in another study evaluating mothers on their infant-feeding intentions, mtc was evaluated as: In general, how much do you care about what each of the following thinks you should do? Participants were then given a list of referents to rate on a scale from “do not care at all” to “care very much” [14] (p. 661). In this example, the focus of item is how much an individual “cares” (or their attitudes) about complying with referents. Motivation to comply has also been operationalized in ways other than reflecting an attitude. A study evaluating how the TPB can explain sugar-sweetened beverage consumption provided an example item of mtc as: How important is it for you to drink the same amount of sugar-sweetened beverages as your friends do? [15] (p. 174) When examining this item closer, it appears to be measuring one’s attitudes (how important is it…) toward imitating or acting similar to a referent (…for you to drink the same amount of sugar-sweetened beverages as your friends do). The distinction between motivation and an attitude is important to clarify for the measurement of mtc. In Fishbein and Ajzen’s [11] original conceptualization of the TRA, they clearly stated that mtc is a measure of motivation, or intentions, and not attitudes. Perhaps of greater promise is an approach suggesting that motivation to comply can be interpreted as the person’s intention to comply with the referent in question. [11] (p. 366) Interestingly, they further note that mtc has the same determinants as intentions. That is, mtc with a referent is determined by one’s attitude toward complying with a referent, and injunctive norm concerning compliance with the referent [11]. An example measure for generalized measures of attitudes and injunctive norms in this context could be as follows, and future researchers could evaluate relationships among these constructs which could lead to new ways of conceptualizing mtc. However, if measures of attitudes toward compliance and injunctive norms concerning compliance are developed, proper pilot testing would be warranted. Motivation to comply In general, I want to do what my parents think I should do.<Strongly agree/strongly disagree> Attitude toward compliance Doing what my parents think I should do is <good/bad; important/not important> Injunctive norms concerning compliance Most people who are important to me want me to do what my parents think I should do. <Strongly agree/strongly disagree> Motivation to comply In general, I want to do what my parents think I should do.<Strongly agree/strongly disagree> Attitude toward compliance Doing what my parents think I should do is <good/bad; important/not important> Injunctive norms concerning compliance Most people who are important to me want me to do what my parents think I should do. <Strongly agree/strongly disagree> View Large Motivation to comply In general, I want to do what my parents think I should do.<Strongly agree/strongly disagree> Attitude toward compliance Doing what my parents think I should do is <good/bad; important/not important> Injunctive norms concerning compliance Most people who are important to me want me to do what my parents think I should do. <Strongly agree/strongly disagree> Motivation to comply In general, I want to do what my parents think I should do.<Strongly agree/strongly disagree> Attitude toward compliance Doing what my parents think I should do is <good/bad; important/not important> Injunctive norms concerning compliance Most people who are important to me want me to do what my parents think I should do. <Strongly agree/strongly disagree> View Large Perhaps one factor that could be causing researchers and practitioners confusion is that textbook authors that describe the TPB define and operationalize mtc in different ways. For example, while Sharma [16] (p. 104) defines mtc as the “degree to which a person wants to act in accordance with the perceived wishes of those significant in his or her life,” Simons–Morton and colleagues [16] (p. 108) defines it as “how much the actor values the opinions of particular referents with respect to a particular object or behavior.” Simons–Morton and colleagues [17] also label mtc an “ill-named” concept, because they assert mtc should represent how one values the wishes of a referent (attitude), rather than how much an individual wants to comply with a referent (motivation/intention). Clearly, more work is needed to better operationalize mtc, and find best methods for measurement and evaluation. Limitations There are a few limitations to this study that should be addressed. All case studies were based on self-reported data, and therefore have the potential for social desirability and other biases inherent in self-reported data. Data presented were also based on the TPB/RAA as operationalized for health behaviors only, and therefore, results on the usefulness of mtci should not be generalized to all behaviors. Data from this study were also based on a convenience sample; therefore, the generalizability of our results may be limited. It should also be noted that while we recruited from the university mass email system, our sample was largely similar to the student population for race (71.7% of the student body was Caucasian) and age (the average age of the student body was 21.3 years); however, our sample contained more female than male students, while the university had approximately equal representations for gender (50.1% of the student body is female) [18]. Finally, all of the case studies used a cross-sectional design; therefore, nothing can be concluded about the causality between constructs. CONCLUSIONS “Social norms” is one of the most popular, and well-documented, constructs for understanding and explaining human behavior. This is especially true in today’s society with the advent of social media platforms (i.e., Facebook, Twitter, Snapchat) that have revolutionized the way that individuals, communities, and organization communicate and share information, and in turn, influence one another [19]. Many studies have demonstrated the impact social media (through social networking and creating new social norms) has on health behaviors [20–22]; however, as reviewed by Ngia and colleagues [19], more research is needed to understand the broader impact social media has, with one specific focus being on social power and influence. It is therefore paramount that researchers understand how social norms are related to health behaviors, and as such, best methods for evaluating norms should be a priority. In Fishbein and Ajzen’s first book published in 1975, Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research, they called mtc the “least understood” construct of the TRA, and conceded a problem with the construct was that it could be interpreted in different ways [11]. As demonstrated in this article, their notions were correct, in that since the introduction of the TRA, measures of mtc have not been well standardized, and vary from study to study. Clearly, more work is needed to better operationalize mtc and find best methods for measurement and evaluation. Compliance with Ethical Standards Conflicts of Interest: There are no conflicts of interest to report for either author. Ethical Approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors. Informed Consent: Informed consent was obtained from all individual participants included in the study. Acknowledgment Researchers did not receive funding for this study. References 1. Fishbein M , Ajzen I. Predicting and Changing Behavior: The Reasoned Action Approach . New York, NY : Psychology Press ; 2010 . 2. McEachan R , Taylor N , Harrison R , Lawton R , Gardner P , Conner M . Meta-analysis of the Reasoned Action Approach (RAA) to understanding health behaviors . Ann Behav Med . 2016 ; 50 ( 4 ): 592 – 612 . Google Scholar CrossRef Search ADS PubMed 3. Montaño DE , Thompson B , Taylor VM , Mahloch J . Understanding mammography intention and utilization among women in an inner city public hospital clinic . Prev Med . 1997 ; 26 ( 6 ): 817 – 824 . Google Scholar CrossRef Search ADS PubMed 4. Budd RJ , North D , Spencer C . Understanding seat-belt use: a test of Bentler and Speckart’s extension of the “theory of reasoned action.” Eur J Soc Psychol . 1984 ; 14 ( 1 ): 69 – 78 . Google Scholar CrossRef Search ADS 5. Sayeed S , Fishbein M , Hornik R , Cappella J , Kirkland Ahern R . Adolescent marijuana use intentions: using theory to plan an intervention . Drugs (Abingdon Engl) . 2005 ; 12 ( 1 ): 19 – 34 . 6. Branscum P , Collado Rivera M , Fairchild G , Qualls Fay K . Do injunctive and descriptive normative beliefs need a value-laden multiplier in value expectancy models? A case series across multiple health behaviors . Health Behav Res . 2017 ; 1 ( 1 ): 1 – 15 . 7. Blue CL , Wilbur J , Marston-Scott M . Exercise among blue-collar workers: application of the theory of planned behavior . Res Nurs Health . 2001 ; 24 ( 6 ): 481 – 493 . Google Scholar CrossRef Search ADS PubMed 8. Etcheverry PE , Agnew CR . Predictors of motivation to comply with social referents regarding one’s romantic relationship . J Soc Pers Relat . 2016 ; 23 ( 2 ): 214 – 233 . Google Scholar CrossRef Search ADS 9. Grossoehme DH , Szczesniak RD , Mrug S , Dimitriou SM , Marshall A , McPhail GL . Adolescents’ spirituality and cystic fibrosis airway clearance treatment adherence: examining mediators . J Pediatr Psychol . 2016 ; 41 ( 9 ): 1022 – 1032 . Google Scholar CrossRef Search ADS PubMed 10. Ickes MJ , Sharma M . Does behavioral intention predict nutrition behaviors related to adolescent obesity ? ICAN: Infant, Child, & Adolescent Nutrition . 2011 ; 3 ( 1 ): 38 – 48 . Google Scholar CrossRef Search ADS 11. Fishbein M , Ajzen I. Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research . Reading, MA : Addison-Wesley ; 1975 . 12. Francis J , Eccles MP , Johnston M , et al. Constructing questionnaires based on the theory of planned behaviour: a manual for health services researchers . Newcastle upon Tyne, UK : Centre for Health Services Research, University of Newcastle ; 2004 . 13. Conner M , Smith N , McMillan B . Examining normative pressure in the theory of planned behaviour: impact of gender and passengers on intentions to break the speed limit . Curr Psychol . 2003 ; 22 ( 3 ): 252 – 263 . Google Scholar CrossRef Search ADS 14. Manstead AS , Proffitt C , Smart JL . Predicting and understanding mothers’ infant-feeding intentions and behavior: testing the theory of reasoned action . J Pers Soc Psychol . 1983 ; 44 ( 4 ): 657 – 671 . Google Scholar CrossRef Search ADS PubMed 15. Zoellner J , Estabrooks PA , Davy BM , Chen YC , You W . Exploring the theory of planned behavior to explain sugar-sweetened beverage consumption . J Nutr Educ Behav . 2012 ; 44 ( 2 ): 172 – 177 . Google Scholar CrossRef Search ADS PubMed 16. Sharma , M. Theoretical Foundations of Health Education and Health Promotion . Burlington, MA : Jones & Bartlett Publishers ; 2017 . 17. Simons-Morton B , McLeroy KR , Wendel ML. Behavior Theory in Health Promotion Practice and Research . Burlington, MA : Jones & Bartlett Publishers ; 2012 . 18. University of Oklahoma . 2018 . University of Oklahoma Fact Book . Available at http://www.ou.edu/irr/fact-books#norman. Accessibility verified May 2, 2018. 19. Ngai EW , Tao SS , Moon KK . Social media research: theories, constructs, and conceptual frameworks . Inter J of Info Manag . 2015 ; 35 ( 1 ): 33 – 44 . Google Scholar CrossRef Search ADS 20. Perkins HW , Linkenbach JW , Lewis MA , Neighbors C . Effectiveness of social norms media marketing in reducing drinking and driving: a statewide campaign . Addict Behav . 2010 ; 35 ( 10 ): 866 – 874 . Google Scholar CrossRef Search ADS PubMed 21. Vaterlaus JM , Patten EV , Roche C , Young JA . #Gettinghealthy: the perceived influence of social media on young adult health behaviors . Computers in Human Behav . 2015 ; 45 : 151 – 157 . Google Scholar CrossRef Search ADS 22. Hoffman EW , Pinkleton BE , Weintraub Austin E , Reyes-Velázquez W . Exploring college students’ use of general and alcohol-related social media and their associations with alcohol-related behaviors . J Am Coll Health . 2014 ; 62 ( 5 ): 328 – 335 . Google Scholar CrossRef Search ADS PubMed © Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Translational Behavioral MedicineOxford University Press

Published: May 30, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off