Increasing the Salience of the Value-Expressive Function of Attitudes toward Exercise among Christians more

RUNNING HEAD: ATTITUDE FUNCTION SALIENCE 1 Increasing the Salience of the Value-Expressive Function of Attitudes toward Exercise among Christians Jenn Anderson 457 Communication Arts & Sciences Building Michigan State University East Lansing, MI 48824 517-355-4604 Mrs.Jenn.Anderson@gmail.com RUNNING HEAD: ATTITUDE FUNCTION SALIENCE ABSTRACT The functional theory of attitudes (Katz, 1960) provides an explanation for why people hold certain attitudes. Attitudes serve various functions: knowledge, ego-defensive, adjustive, and value-expressive. This proposal concentrates on the value-expressive function and proposes a measure for value-expressive attitude functionality to determine the degree to which a person perceives that their attitude expresses their personal values. The proposed study examines Christians’ attitudes toward exercise and those attitudes’ value-expressive functionality. It will 2 measure attitudes toward exercise and the value-expressive functionality of those attitudes before and after exposure to values-based persuasive messages about exercise. Results of the study will provide evidence of the values held by Christians, their attitudes toward exercise, the degree to which those attitudes express personal values, and the effects of persuasive messaging on attitudes and attitude functionality. Implications for health communication campaigns are discussed. Attitude Function Salience 3 Introduction More than two-thirds (78%) of adults in the United States consider themselves Christian (Pew Forum on Religion & Public Life, 2008). Around half (51.3%) of adults in the United States are affiliated with Protestant religions, i.e., Evangelical, Mainline, and Historically Black Churches, and a little less than a fourth (23%) are affiliated with the Catholic Church (Pew, 2008). It is no secret that religion plays a significant role in the lives of many Americans; what’s more, it is significantly related to health outcomes (Long, 2004; Koenig, McCullough, & Larson, 2001). As Parrot (2004) suggested, health communication researchers should increase their attention to, and understanding of, the relationship between religious faith and health behaviors, in order to more effectively communicate about health topics to these populations. In the past few years, health communication researchers have increasingly worked to incorporate religious or spiritual themes into targeted campaign messages (Adogame, 2007; Anderson, 2004; Boulay, Tweedie, Fiagbe, 2008; Moss, Gallaread, Siller, & Klausner, 2004). Targeted campaigns like these are useful because they respond directly to the needs, knowledge, experiences, and culture of a given population (Salmon & Atkin, 2003). In addition, these types of campaigns connect health communication research and theory to community-based projects to enhance health (Holt, Wynn, Southward, Litaker, Jeames, & Schulz, 2009). The proposed study provides a new way to think about framing health communication campaigns in the context of religious communities. It is based on a functional approach to attitudes (Katz, 1960; Smith, Bruner, & White, 1956), which is concerned with understanding why people hold the attitudes they do. According to the functional approach, one function that attitudes may serve is to express personal values (Katz, 1960; Hullett, 2001). Thus, attitudes toward various health behaviors have the potential to express some of the personal values that Attitude Function Salience 4 are shared among members of religious communities. If researchers can understand these personal values and the ways they are connected to attitudes toward health issues, they can develop better targeted and more effective health communication campaigns on the basis of those values. Value-expressive attitudes are especially useful for health communication researchers to consider because they are uniquely predictive of health behaviors, even when controlling for other factors like basic attitudes, subjective norms, and perceived behavioral control (Maio & Olson, 1995). The proposed study is an extension of a formative study with Christians from a variety of denominations that examined their personal values, their attitudes toward exercise, and the relationship between their personal values and those attitudes (Anderson, 2010). The current study is focused on exercise as a health behavior that can help prevent and decrease the prevalence of obesity. Christian churches, especially evangelical churches, are a critical place to examine and expand our understanding of health communication campaign effectiveness among what Dearing, Meyer, and Rogers (1994) call “unique populations,” who are less likely to attend to general health messages. Unique populations, like Evangelical Christians, feel ostracized by mainstream society as a result of their personal values and beliefs, and actively avoid mainstream media in order to pay attention to specialized (i.e., Christian) media. Thus, this population is important to reach with specialized persuasive health messages, because they have the same weight-related health risks as the general population (Gillum, 2006), but are less likely to attend to general health messaging (Gormly, 2003). This paper proposes using a functional approach to attitudes to develop targeted, persuasive, values-based health messages for a Christian population. Attitude Function Salience 5 Targeting Health Campaigns for Faith-Based Communities Previous faith-based health initiatives There are a number of ways to approach the relationship between religious faith and health. Some studies have examined how religious beliefs impact the way people understand various health issues, mostly with the goal of contrasting “religious understandings” with conventional understandings of health. These studies generally demonstrate that religious people report a greater sense that God has some degree of control over their health (either directly— through healing, or indirectly—through doctors) and that they consider their faith when making health decisions (Abrums, 2000; Harris, Parrott, & Dorgan, 2004; Mansfield, Mitchell, & King, 2002). Others focus on the relationship between religious behaviors, such as church attendance or membership, and various health behaviors or health outcomes. In general, church attendance is typically not significantly related to health behaviors, but is related to indirect influencers of health such as increased social support (Koenig et al., 2001). A third approach to the faith-health connection comes from health communication scholars who partner with faith communities to develop interventions targeted to members of local congregations. Some of these interventions utilize church organizations as a mechanism for delivery of health messages to specific populations (Catanzaro, Meador, Koenig, Kuchibhatla, & Clipp, 2006; Ferguson, Dortzbach, Dyrness, Dabir, & Sprujit-Metz, 2008). In particular, many interventions have been directed at the African American church community (see Wilcox et al., 2007 for a review of their effectiveness). Other campaigns focus more on the message features, rather than the message channels, that create uniquely persuasive messages for religious populations. In other words, these types of campaigns feature spiritual themes in health messages in an effort to create targeted messages. Compared with secular messages, spiritually-based Attitude Function Salience 6 messages are more effective at generating health attitude and behavior change (Holt, Lee, & Wright, 2008; Holt et al., 2009). Though great progress has been made since Parrott (2004) highlighted the lack of health communication research integrating religious beliefs, efforts in this area can be improved. One way to improve these efforts is to put a greater emphasis on understanding why religious beliefs might be related to a given health outcome. In other words, future campaigns need to go beyond utilizing faith-based organizations as an efficient way to deliver messages or a useful way to segment populations. Instead, the focus should be on understanding how religious beliefs and behaviors are related to specific health attitudes and behaviors. The current paper proposes that, unlike previous research examining this connection (Abrums, 2000; Harris et al., 2004; Mansfield et al., 2002), the relationship between faith and health may not be direct, but mediated by personal values. In other words, the personal values that develop as a result of religious faith, and participation in a faith-based organization, are the mechanism through which religious faith is connected to health attitudes and behaviors. Thus, rather than looking at the relationship between health behaviors and religious identification, or asking religious people to describe their own health beliefs and behaviors, this study proposes examining the ways that religious beliefs are expressed through personal values and how those personal values are linked to health attitudes and behaviors. A Functional Approach to Attitudes The functional approach to attitudes (Katz, 1960; Smith et al., 1956) posits that people hold attitudes for different reasons. In other words, attitudes serve various functions. In early work on the subject, Katz (1960) proposed four functions that attitudes serve; an earlier version of this theory presented by Smith et al. (1956) included similar functions. Katz’s four functions Attitude Function Salience 7 include the knowledge function, the adjustment function—or social-adjustive function (Smith et al., 1956), the value-expressive function, and the ego-defensive function. Attitudes that serve the knowledge function aid the individual in making sense of their world. In some ways, all other functions are a subset of this original function, because all attitudes help people make sense of their environment and experience. The adjustment function has to do with evaluating the rewards and punishments associated with holding a given attitude, or weighing the costs and benefits (Katz, 1960). Smith et al.’s (1956) social-adjustive function includes the social costs and benefits of holding a given attitude. The ego-defensive function, as described by Katz (1960), has to do with a person holding an attitude in response to attacks on his or her ego, and often includes the formation of attitudes designed to conceal one’s true nature from oneself. Finally, there is the value-expressive function. These attitudes provide allow a person to express his/her personal values and gain satisfaction from their expression (Katz, 1960). Values are relatively enduring abstract beliefs about achieving desirable end-states or goals (Rokeach, 1973; Schwartz, 1992). For example, one may value self-discipline because one believes that self-discipline is an ideal way to achieve a desirable end-state or goal such as earning a degree or winning a race. Self-discipline, therefore, could be a value one expresses through her/his positive attitude toward following an exercise plan. When this person expresses to a friend, “I think following a plan for exercise is a great way to stay on track and reach your goals,” the person is expressing, through his/her attitude toward exercise plans, that s/he values self-discipline. Value-expressive attitudes clarify one’s self-image and, at the same time, their expression crafts one’s self-image (Katz, 1960). They are strongly related to a person’s values (Maio & Olson, 2000) and allow a person to strive for or maintain his/her personal values (Hullett & Boster, 2001). Value-expressive attitudes need not be publicly expressed or observed by others Attitude Function Salience 8 in order to be considered value-expressive (Herek, 1986; Hullett & Boster, 2001). Valueexpressive attitudes are important for researchers because they are more resistant to change and better predictors of behavior than other types of attitudes (Shavitt, 1989). Maio and Olson (1995) experimentally tested the predictive capacity of both utilitarian and value-expressive attitudes and found that value-expressive attitudes predicted intentions to donate to cancer research even when controlling for basic attitudes, subjective norms, and perceived behavioral control. The empirical study of a functional approach to attitudes has been plagued by difficulties with conceptualization and measurement. One problematic approach to studying valueexpressive attitudes has been to treat them as by-products of individual differences in personality characteristics, such as self-monitoring (DeBono, 1987; DeBono & Harnish, 1988; see Shavitt, 1989 for a review). In these studies, general predispositions toward a given attitude function are inferred from personality characteristics. For example, high self-monitors are said to form more social-adjustive attitudes, whereas low self-monitors form more value-expressive attitudes (DeBono, 1987; DeBono & Harnish, 1988). Working with a similar conceptualization of attitude functionality, Murray, Haddock, and Zanna (1996) persuaded participants that they were the type of people who held either mostly value-expressive or mostly social-adjustive attitudes, and then measured the persuasive effect of value-expressive and social-adjustive messages. Not surprisingly, “functionally-matched” messages were most effective (Murray et al., 1996). But, both of these “personality” approaches to attitude functionality eschew the functional approach’s usefulness in terms of explaining the reasons that individual attitudes are held, rather than explaining individual predispositions to hold certain types of attitudes. For value-expressive attitudes in particular, Hullett and Boster Attitude Function Salience 9 (2001) argued that the personality-based approach removes the role of personal values in the effectiveness of values-based persuasive messages. Among studies that have attempted to measure attitude functionality, Herek (1986) and Shavitt (1990) provide examples of valid, but almost un-generalizable, methods. Herek’s (1987) work in the area demonstrates how time-consuming and situation-specific attitude function measurement can be. Looking at attitudes toward a particular outgroup (“homosexuals”), Herek (1987) coded undergraduate students’ open-ended responses to an essay prompt: “I have generally positive/negative attitudes toward lesbians and male homosexuals because…” (p. 287). Respondents selected either “positive” or “negative” and then wrote down the rest of their thoughts on the subject. After those responses were coded for functions, and a typology of functions was derived for the study, Herek (1987) developed a scale to measure the functions present in people’s attitudes toward gay men and lesbians, as well as toward individuals with stigmatizing illnesses. Herek (1987) successfully demonstrated that attitude functions could be measured with respect to those particular attitude objects, but unfortunately the measures are so object-specific that they have not been adapted or adopted by other researchers studying the functional approach to attitudes. Shavitt (1990) also started with attitude objects and worked backward to identify the typical functions that attitudes toward those objects served. Instead of choosing a complex attitude object like gay men and lesbians, Shavitt (1990) used basic objects in the hopes that attitudes toward them would serve the same functions across participants. For example, Shavitt (1990) found that “coffee” typically only elicits attitudes with a utilitarian function, whereas “cars” can elicit utilitarian or social-expressive functions, or both. While Shavitt’s (1990) approach was innovative, it did little to illuminate the process by which attitude functions are Attitude Function Salience 10 formed or manipulated, and it did not illuminate how attitude functionality operates for novel, complex attitude objects that may elicit multiple functions within and across participants. Value-Matching A different way to approach the study of attitude functionality is to concentrate on only one attitude function at a time to learn how it operates and how to develop messages that are matched to those functions for maximum effectiveness (Hullett & Boster, 2001; Lapinski & Boster, 2001). Work with value-expressive attitudes in particular has moved away from attempting to measure or manipulate the multiple possible functions of attitudes, and has instead begun to focus on value-relevance in attitudes and messages. An unspoken assumption of this approach is that any attitude can be linked with personal values so that the value-expressiveness of an attitude can be increased. This work has shown that health messages matched to audiences’ values are more effective than messages that either do not contain audiences’ values, or are based on values not held by the audience (Hullett, 2002, 2004, 2006; Hullett & Boster, 2001). Hullett and Boster (2001) showed that when messages were matched to participants’ values, in this case conformity or self-directedness, the participants considered the message to be of higher quality, which in turn led them to agree with message recommendations, which in turn led them to indicate intention to perform the specified behavior. To develop and test the messages, Hullett and Boster (2001) first measured the extent to which the audience considered certain values important, and then developed messages based on those values. This offered the opportunity to test the direct relationship between personal values and perceptions of persuasive message quality, but did not provide a mechanism for understanding the degree to which participants’ attitudes toward the attitude object (in this case tuition increase for undergraduate Attitude Function Salience 11 students) served a value-expressive (or any other) function either before or after exposure to the message. Hullett (2002) addressed the issue of measuring the association of a particular attitude with personal values. In two studies, Hullett (2002) examined not only which values an attitude served, but also the value-relevance of said attitudes. Value-relevance is “the extent to which people perceive their attitudes as useful in attaining valued states” (Hullett, 2002, p. 161). Importantly, Hullett’s (2002) conceptualization of values differs slightly from many other studies of value-expressive attitudes, in that it focuses primarily on achieving desired end-states. However, this conceptualization of values more closely reflects Rokeach’s (1973) and Schwartz’s (1992) seminal work in human value theory and measurement. It also allows one to measure not simply whether the attitude is based on a personal value (Herek, 1987), but instead whether the attitude is useful in attaining the end state specified by a given value. If the attitude is useful in attaining a specific end-state, it is said to be value-relevant. In Study 1, Hullett (2002) postulated that the audience’s perception of a value-based message’s relevance would be dependent on the value-relevance of their attitude. If the audience considers their attitude to be value-relevant, and the message is matched to their values, the audience will consider the message relevant. A more relevant message will be judged as higher quality, and a high quality message should result in attitude change. Hullett (2002) found support for all the predicted relationships, and the overall causal model. In Study 2, Hullett (2002) focused on changing a negative attitude toward a particular object 1 to a positive attitude by demonstrating that the new, positive attitude better serves the important value than the current, negative attitude. In order to do this, Hullett (2002) surveyed the participants to assess their attitudes toward taking comprehensive exams, and also their 1 Taking comprehensive exams. Attitude Function Salience 12 personal values. Then, attitudes were correlated with value importance scores, to determine the value that was most negatively related to the participants’ current attitudes toward comprehensive exams. It was found that the value of wealth was most negatively correlated with attitudes toward comprehensive exams (r = -.26, p <.01) and was also considered to be a relatively important value to the participants. Based on this relationship, messages were developed to persuade the audience that holding a positive attitude toward comprehensive exams would be useful in attaining an end-state of being wealthy. Exposure to those messages, more than exposure to messages that focused on achieving an unimportant end-state, was positively associated with perceived value-relevance of attitudes toward comprehensive exams. In other words, participants that were exposed to messages claiming that positive attitudes toward comprehensive exams would be useful in becoming wealthy were more likely to consider their attitude toward comprehensive exams relevant to their personal value of wealth than were those exposed to a value-irrelevant message (Hullett, 2002). However, there was no test of the degree to which perceived value-relevance changed as a result of exposure to the message since value-relevance was not measured prior to message exposure. Perhaps it could be inferred that a negative relationship between wealth and attitudes toward comprehensive exams implies that attitude did not express that value. But the magnitude of the negative correlation suggests that there was a significant relationship between the value and the attitude, but the value prompted a negative attitude toward an object rather than a positive one. In other words, it could be that because the participants valued wealth, they had a negative attitude toward comprehensive exams. Perhaps they considered passing comprehensive exams to be a barrier to earning a degree and obtaining a high-paying job; thus, their value of Attitude Function Salience 13 wealth was relevant to their attitude toward comprehensive exams—it just did not produce a positive attitude. Thus, rather than developing a value-expressive attitude through highlighting the valuerelevance of a given attitude through messages, the preceding studies may have instead switched the valence of the attitude while maintaining the value-relevance of the attitude itself. Perhaps the effect was greater in the condition with a value-relevant message simply because it made the value more salient, and thus cued the participants to consider their attitude in terms of their values. In other words, it was a matter of making the dormant or subconscious link between the attitude and the value salient, rather than introducing it for the first time to the participants. In other expansions and tests of persuasion via values-matched messaging, Hullett (2004, 2006) targeted health beliefs and attitudes. In these studies, Hullett (2004, 2006) added additional predictors such as goal-relevance and negative affective states (i.e., guilt, fear, and uncertainty) to the causal model linking value-expressive attitudes to behavioral intentions. In all cases, the value-matched messages produced the greatest changes in attitudes and behavioral intentions when compared to control or value-mismatched messages (Hullett 2002, 2004, & 2006; Hullett & Boster, 2001). In conclusion, previous work on the formation and use of value-expressive attitudes has provided evidence of the usefulness of the functional approach to attitudes for persuasive health communication. First, it has shown that people do conceive of their attitudes as serving various functions. Second, it has shown that people believe their attitudes can express their personal values, i.e., they hold value-expressive attitudes. Third, it has shown that the relevance of personal values plays an important role in the processing of values-based messaging. Fourth, it has shown that general attitude functionality can be manipulated, i.e., the functions of attitudes Attitude Function Salience 14 can be made more salient. Finally, it has shown that functionally-matched messages, particularly values-based messages, are more effective than mis-matched messages. Attitude Function Salience There is room for expansion and improvement in the measurement and application of functional theory. First, there are few studies with direct measures of the extent to which a subject perceives a specific attitude to be serving a given function. For example, Hullett (2002) measured the degree to which a subject considered an attitude to be relevant to their values. But, value-relevance of an attitude was measured only after exposure to the message, and the implications of that measurement design are discussed above. Second, the issue of measuring attitude functionality only after exposure to experimental conditions is an important weakness across studies. Because it is measured only after exposure to experimental stimuli, it is unclear the degree to which perceived attitude functionality actually changed in any given condition. Third, given the lack of pretest-posttest comparisons, there is a possibility that these studies are not forming or changing attitude functions as much as they are making those functions more salient. This paper adds another dimension to the understanding of attitude functionality, and its use in the persuasion domain: attitude function salience. Perhaps simply prompting a participant to think about an attitude in terms of the function it can serve can have an effect on a number of cognitive processes: message processing (Chaiken, 1980; Petty, Cacioppo, & Goldman, 1981), a drive to reduce cognitive dissonance (Festinger & Carlsmith, 1959), and in the case of valueexpressive attitudes, a strengthening of connections between attitudes and self-concept or personal values (Roskos-Ewoldsen, Arpan-Ralstin, & St. Pierre, 2002). Attitude Function Salience 15 Attitude function salience borrows conceptually from work with attitude accessibility (Fazio, Powell, & Williams, 1989; Roskos-Ewoldsen, et al., 2002). Attitude accessibility has to do with the ease of retrieving an attitude from memory (Roskos-Ewoldson, et al., 2002), or the “strength of association in memory between the object and the individual’s evaluation of the object,” (Fazio, et al., 1989, p. 280). Similarly, attitude function salience deals with the strength of the association between a person’s attitude toward an object and the function that attitude serves, or the ease with which a person can consider an attitude’s function when prompted to think about an attitude in reference to an object. Attitude accessibility moderates the relationship between attitude and behavior such that a more accessible attitude is more predictive of behavior than a less accessible attitude (Fazio, et al., 1989). In the same way, it is hypothesized that attitude function salience moderates the attitude-behavior link such that greater attitude function salience results in attitudes that are more predictive of behaviors than attitudes with low function salience. Consider, for example, one’s attitude toward exercise. If the function of the attitude is not made salient, one might simply think: “I like exercise” or “I don’t care for exercise.” However, if the value-expressive function of such an attitude were made salient through a persuasive message that, for example, showed that developing an exercise program shows others that you value self-discipline, a person might think, “I like exercise because it lets others know I value self-discipline.” This second type attitude is value-expressive; it is held because it allows the person to express his or her values. Producing the second type of attitude should be the aim of health communication campaign designers since value-expressive attitudes are more predictive of behavior and more resistant to change than basic attitudes (Maio & Olson, 1995). Thus, a message that can increase the salience of attitudes’ value-expressive function will generate Attitude Function Salience 16 attitudes that are more predictive of behavior and resistant to change than messages that do not produce such functional salience. The proposed study is an attempt to manipulate attitude function salience by creating values-based messaging to both enhance attitude function salience and generate more positive attitudes toward exercise. Messages will incorporate values-based content derived from formative work with a Christian sample that assessed personal values, attitudes toward exercise, and exercise behavior, as well as the relationships between those variables (Anderson, 2010). Anderson (2010) found that the following values 2 from Schwartz’s (1992) value schema were rated highly by participants and were also related to exercise attitudes: responsible, a spiritual life, devout, obedient, ambitious, self-discipline, humble, and successful. Among those, a spiritual life, devout, obedient, and self-discipline formed a “spiritual value-cluster,” or grouping of values that are highly interrelated both conceptually and empirically. Values that were ranked quite low and were unrelated to exercise attitudes were: wealth, success, and social recognition; this grouping could be called the “success value-cluster” (Anderson, 2010). The proposed message-testing study will use three conditions: a spiritual values message, a success values message, and an informational (or no values) message. The following hypotheses are proposed with respect to the effectiveness of each condition. H1: Positive change in attitudes toward exercise will be greatest among participants exposed to the value-matched message. H2: Participants in the value-matched message condition will report the most positive attitudes toward exercise. H3: Positive change in value-expressive functionality will be greatest among participants exposed to the value-matched message. 2 These values were also significantly correlated with strength of religious faith, indicating that they are uniquely religious values that can be linked to exercise. Attitude Function Salience 17 H4: Participants in the value-matched message condition will report their attitudes as most value-expressive. H5: Intention to exercise will be greatest among participants exposed to the valuematched message condition. Given that previous studies have not explicitly examined value-expressive functionality of attitudes, the relationship between value-expressive attitude functionality and other variables of interest remains unclear. Therefore, rather than hypothesizing about possible relationships between those variables, the following research question is advanced: RQ1: What is the relationship between attitudes toward exercise and perceived valueexpressive functionality of attitude prior to message exposure in each condition? Since increased value-expressive functionality does not necessarily guarantee attitude change in the direction desired (i.e., more positive attitudes toward exercise), the two need not be positively correlated. In fact, they may be negatively correlated; perhaps upon greater reflection on one’s values, one determines that his/her positive attitude toward exercise was misplaced, and therefore develops a more negative attitude toward exercise. Attitude functionality and attitude valence may also be uncorrelated. It is entirely possible that the messages could generate greater perceived attitude functionality, but not convince the participants to hold positive attitudes toward exercise. On the contrary, the messages might convince participants to hold more positive attitudes toward exercise, while failing to induce greater perceptions of value-expressive attitude functionality. The relationships between attitudes, attitude change, attitude functionality, and attitude functionality change, then, are unknown. Thus, the following research questions are advanced: RQ2: What is the relationship between attitudes toward exercise and perceived valueexpressive functionality of attitude after message exposure in each condition? Attitude Function Salience 18 Method Participants Based on the study design, 150 participants will be recruited from ten churches representing at least five different denominations. Participants will be over 18, and must be affiliated with a local church. Procedure Data will be collected at local churches after Sunday services at three time points across eight weeks. At Time 1, participants will respond to a survey measuring demographics, religiosity, exercise attitudes, exercise behavior, and personal values (See Appendix A for a copy of all measures). Based on the values reported in this survey, exercise messages will be refined for three conditions: value-matched message, value-mismatched message, and a control message (i.e., informational/non-values-based message). Appendix B includes example messages that were developed using data obtained from an earlier study of Christians’ values (Anderson, 2010). Participants are expected to report value ratings similar to those reported by Christians in an earlier study (Anderson, 2010), but between Time 1 and Time 2, the messages will be modified based on value ratings from Time 1. The value-expressive functionality scale will also be refined based on the data from Time 1. At Time 2, six weeks after Time 1, participants will take a brief pre-test survey that will measure their attitudes toward exercise, exercise behavior 3, and value-expressive attitude function salience. One week later, at Time 3, participants will be randomly assigned to message conditions. At Time 3, participants will read one of the three messages, and respond to a post-test that will include measures of exercise attitudes, value-expressive attitude function salience, Exercise attitudes and behaviors will be re-measured at Time 2 to account for changes in those variables that might occur naturally due to summer weather and schedules offering more opportunities to exercise. 3 Attitude Function Salience 19 exercise behavior and exercise behavioral intentions. Behavioral intentions will be measured with a modified version of the exercise behavior scale. Participants’ data will be matched at all three time points so that individual change scores can be calculated. Measures Values. Schwartz’s (1992) universal values scale will be used to verify the most important values in the sample. Based on evidence from Anderson (2010) the current value measure will only include the top ten most important values for this population: meaning in life, responsible, family security, a spiritual life, devout, obedient, healthy, ambitious, self-discipline, and humble. As Schwartz (1992, p. 17) and Hullett and Boster (2001, p. 143) recommend, respondents will indicate the degree to which each value acts as a “guiding principle in my life” on a 11-point scale ranging from 0 (not at all important) to 10 (of extreme importance) rather than rank-ordering all the values. This study will use Hullett and Boster’s (2001) scaling rather than Schwartz’s (1992, -1 to 7) scaling, since it seems more intuitive for both participants and readers. A copy of this measure can be found in Appendix A. Attitudes toward exercise. Attitudes toward exercise will be measured using the attitude toward exercise scale (ATE, Anderson, 2010). The scale is comprised of ten questions asking participants to indicate their level of agreement with each statement using a Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree), with higher scores indicating a more positive attitude toward physical activity. Anderson (2010) found the scale to be unidimensional (CFI = .86, RMR = .03) and reliable, Cronbach’s α = .85. Responses were also consistent across race, gender, age, marital status, number of children, state of residence, and educational attainment. A copy of this measure can be found in Appendix A. Attitude Function Salience 20 Exercise behavior. A measure of exercise behavior that taps into three key elements of exercise (frequency, duration, and intensity) was developed for this study. In addition to closedended self-report of exercise frequency, duration, and intensity, the measure allows for openended report of the specific type of exercise a participant may engage in. A copy of this measure can be found in Appendix A. Strength of religious faith. The Santa Clara Strength of Religious Faith Questionnaire (SCSORF, Plante & Boccaccini, 1997) will be included since it measures the extent to which a person has integrated his or her faith in daily life, and is therefore a useful measure to determine the overall religiousness of the sample studied. The SCSORF Questionnaire has been shown to be a reliable and valid unidimensional measure (Sherman & Simonton, 2001). A copy of this measure can be found in Appendix A. Attitude functionality. The degree to which participants’ attitudes toward exercise are functioning as value-expressive attitudes will be measured with 10 items designed by the researcher for this study. Participants will rate their agreement with each of the 10 statements on a 5-point Likert-type scale (1 = strongly disagree, 5 = strongly agree), with higher scores representing greater perception of value-expressive attitude functionality. Some items include a blank, each of which will be filled in with a specific value that was determined to be important to the participants during study 1. A copy of this measure can be found in Appendix A. Planned Analysis Reliability analysis (Cronbach, 1951) and confirmatory factor analysis (Hunter & Gerbing, 1982) will be conducted on all scales to ensure that they are unidimensional and reliable. Differences in variables of interest, (i.e., value ratings, attitudes toward exercise, attitude functionality, exercise behavior, and intention to exercise) with respect to denomination, gender, Attitude Function Salience 21 race, age, education, and religiosity will be analyzed. If differences are observed with respect to denomination, or other demographic variables, the following analyses will be modified to account for those differences. To test H1, a 1 x 3 ANOVA using attitude change scores (calculated by subtracting the pretest attitude score from the posttest attitude score) as the dependent variable will be conducted. The three message conditions will serve as the independent variables. To test the comparison of matched against mismatched and control messages, the following contrast coefficients will be assigned to the matched, mismatched and control messages respectively: 2, -1, and 1. To test H2, a 1 x 3 ANOVA using posttest attitude scores as the dependent variable will be conducted. The three message conditions will serve as the independent variables. To test the comparison of matched against mismatched and control messages the following contrast coefficients will be assigned to the matched, mismatched and control messages respectively: 2, -1, and 1. To test H3, a 1 x 3 ANOVA using value-expressive functionality change scores (calculated by subtracting pretest from posttest scores on the value-expressive functionality measure) as the dependent variable will be conducted. The three message conditions will serve as the independent variables. To test the comparison of matched against mismatched and control messages the following contrast coefficients will be assigned to the matched, mismatched and control messages respectively: 2, -1, and 1. To test H4, a 1 x 3 ANOVA using value-expressive functionality posttest scores as the dependent variable will be conducted. The three message conditions will serve as the independent variables. To test the comparison of matched against mismatched and control Attitude Function Salience 22 messages the following contrast coefficients will be assigned to the matched, mismatched and control messages respectively: 2, -1, and 1. To test H5, a 1 x 3 ANOVA using intention to exercise as the dependent variable will be conducted. The three message conditions will serve as the independent variables. To test the comparison of matched against mismatched and control messages the following contrast coefficients will be assigned to the matched, mismatched and control messages respectively: 2, 1, and 1. Discussion The proposed study will provide evidence of the extent to which Christians consider their attitudes toward exercise to be expressive of their personal values. One interesting outcome will be documenting the degree to which Christians perceive of their attitudes as value-expressive prior to messaging. If they already have a strong sense of their attitudes as serving a valueexpressive function, then the matched values-based messaging may not strengthen the perceived attitude functionality of their attitudes toward exercise, but might still have an effect on their attitude toward exercise. Conversely, if Christians do not perceive that their attitudes toward exercise serve a value-expressive function, this study will test whether a value-matched persuasive message will be effective in prompting the participants perceive a greater valueexpressive functionality of their attitudes toward exercise. If the study can successfully create positive change in attitudes toward exercise by using values-based messages and value-linking exercises, this same approach could be adopted in future “real-world” campaigns. A values-based campaign strategy could be useful across a range of health behaviors and target populations, but the success of such a strategy rests on the quality of formative research that reveals the current values of the target population. Attitude Function Salience 23 If the study can not only create positive change in attitudes toward exercise, but also increase the perception of value-expressive functionality of those particular attitudes, then it will have major implications for the use of functional theory in persuasion research. First, it will be the first measure of change in perceived attitude functionality. Second, it will provide a validation of a scale used to measure value-expressive attitude functionality. Third, it will provide the groundwork for the development of additional scales that could be used to assess perceived attitude functionality for different functions. Fourth, it shows the ability of functional theory to be used to target specific—rather than general—attitudes and attitude functions. Fifth, it will model the co-processes of attitude change and attitude function salience change. In other words, it will show that attitude change can occur simultaneously with a change in the perception of attitude function (or the salience of that attitude’s function), while also noting that they two are separate processes that need not occur in tandem. Even though this proposed study can provide very useful practical and theoretical information, it does have some limitations. First, the proposed study, like previous work in functional theory, does not provide the participants with the opportunity to report which functions their attitudes serve. Instead, this study only looks at the extent to which a particular attitude serves the value-expressive function. As mentioned above, this study can serve as a first step in the development of a broader measurement designed to tap into the degree to which an attitude serves any and all of the functions, but as it is designed, the proposed experiment does not allow for this. Another limitation to the study is that, like many experiments, this study does not include a measurement of sustained attitude change. In other words, it will be unknown whether changes observed in this experiment are maintained after the participants leave. Attitude Function Salience 24 Future studies should examine the value-expressiveness of other health attitudes, such as attitudes toward organ donation or attitudes about alcohol consumption. If key values are attached to these types of attitudes, then values-based messaging in these campaigns could be quite effective. Future studies should also be designed to tease out the differential processes of attitude change and attitude function change. These studies could examine what components of persuasive messages cause change in attitude but not function, vice versa, and in both. Finally, future studies should validate the value-expressive functionality scale in other populations, and across different attitudinal domains. In conclusion, the role of values in health attitudes is important to consider because it can impact the effectiveness of a health communication campaign. Matching health messages to audience values can increase their effectiveness, and also enhance the degree to which health attitudes express personal values. The more value-expressive attitudes are perceived to be, the stronger they become and the more predictive they are of behavior. Proposing a method for measuring value-expressive attitude functionality will allow researchers to better understand the degree to which attitudes are value-expressive and therefore design more effective persuasive messages and better understand attitude functions. Attitude Function Salience 25 References Abrums, M. (2000). ‘Jesus will fix it after a while’: Meanings and health. Social Science & Medicine, 50, 89-105. Adogame, A. (2007). HIV/AIDS support and African Pentecostalism: The case of the Redeemed Christian Church of God (RCCG). Journal of Health Psychology, 12(3), 475-484. Anderson, J. (2010, Nov.). Building bridges between faith and health for faith-based community campaigns: Linking Christians’ personal values to exercise attitudes and behaviors. Paper accepted for presentation at the meeting of the National Communication Association, San Francisco, CA. Anderson, J. (2009, Nov.). When health meets holiness: The impact of spiritual community norms on personal health behaviors. Paper presented as part of the panel “Sustaining and Transforming the Communication Discipline: Examining the Relationship of Spirituality and Communication” at the meeting of the National Communication Association, Chicago, IL. Boulay, M., Tweedie, L., & Fiagbe, E. (2008). The effectiveness of a national communication campaign using religious leaders to reduce HIV-related stigma in Ghana. African Journal of AIDS Research, 7(1), 133-141. Catanzaro, A. M., Meador, K. G., Koenig, H. G., Kuchibhatla, M., & Clipp, E. C. (2006). Congregational health ministries: A national study of pastor’s wives. Public Health Nursing, 24(1), 6-17. Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology, 39, 752-766. Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16(3), 297-334. DeBono, K. G. (1987). Investigating the social-adjustive and value-expressive functions of attitudes: Implications for persuasion. Journal of Personality and Social Psychology, 52(2), 279-287. DeBono, K. G., & Harnish, R. J. (1988). Source expertise, source attractiveness, and the processing of persuasive information: A functional approach. Journal of Personality and Social Psychology, 55(4), 541-546. DeHaven, M. J., Hunter, I. B., Wilder, L., Walton, J., & Berry, J. (2004). Health programs in faith-based organizations: Are they effective? American Journal of Public Health, 94(6), 1030-1036. Ferguson, K. M., Dortzbach, K., Dyrness, G. R., Dabir, N., & Sprujit-Metz, D. (2008). Faithbased programs and outcomes for street-living youth in Los Angeles, Mumbai, and Nairobi: A comparative study. International Social Work, 51, 159-177. Festinger, L., & Carlsmith, J.M. (1959). Cognitive consequences of forced compliance. Journal of Abnormal and Social Psychology, 58, 203-210. Gillum, R. F. (2006). Frequency of attendance at religious services, overweight, and obesity in American women and men: the Third National Health and Nutrition Examination Survey. Annals of Epidemiology, 16(9), 655-660. Gormly, E. (2003). Evangelizing through appropriation: Toward a cultural theory of growth of contemporary Christian music. Journal of Media and Religion, 2 (4), 251-265. Attitude Function Salience 26 Harris, T. M., Parrott, R., & Dorgan, K. A. (2004). Talking about human genetics within religious frameworks. Health Communication, 16, 105-116. Herek, G. M. (1986). The instrumentality of attitudes: Toward a neofunctional theory. Journal of Social Issues, 42(2), 99-114. Herek, G. M. (1987). Can functions be measured? A new perspective on the functional approach to attitudes. Social Psychology Quarterly, 50(4), 285-303. Holt, C. L., Roberts, C., Scarinci, I., Wiley, S., Eloubeidi, M., Crowther, M., et al. (2009). Development of a spiritually based educational program to increase colorectal cancer screening among African American men and women. Health Communication, 24, 400412. Holt, C. L., Lee, C., & Wright, K. (2008). A spiritually based approach to breast cancer awareness: Cognitive response analysis of communication effectiveness. Health Communication, 23 (1), 13-22. Holt, C. L., Wynn, T. A., Southward, P., Litaker, M. S., Jeames, S., Schulz, E. (2009). Development of a spiritually based educational intervention to increase informed decision making for prostate cancer screening among church-attending African American men. Journal of Health Communication, 14 (6), 590-604. Hullett, C., & Boster, F. (2001). Matching messages to the values underlying value-expressive and social-adjustive attitudes: Reconciling an old theory with a contemporary measurement approach. Communication Monographs, 68(2), 133-153. Hullett, C. (2002). Charting the process underlying the change of value-expressive attitudes: The importance of value-relevance in predicting the message effect. Communication Monographs, 69(2), p. 158-178. Hullett, C. R. (2004). Using functional theory to promote sexually transmitted disease (STD) testing. Communication Research, 31(4), 363-396. Hullett, C. R. (2006). Using functional theory to promote HIV testing: The impact of valueexpressive messages, uncertainty, and fear. Health Communication, 20(1), 57-67. Hunter, J., & Gerbing, D. (1982). Unidimensional measurement, second order factor analysis, and causal models. Research in Organizational Behavior, 4, 267-320. Katz, D. (1960). The functional approach to the study of attitudes. Public Opinion Quarterly, 24, 163-204. Koenig, H. G., McCullough, M. E., & Larson, D. B. (2001). Handbook of religion and health. Oxford: Oxford University Press. Lapinski, M., & Boster, F. (2001). Modeling the ego-defensive function of attitudes. Communication Monographs, 68(3), 314-324. Long, M. (2004). The Church’s role in health and wholeness. Health Communication, 16(1), 129-130. Maio, G. R., & Olson, J. M. (1995). Relations between values, attitudes, and behavioral intentions: The moderating role of attitude function. Journal of Experimental Psychology, 31, 266-285. Maio, G. R., & Olson, J. M. (2000). What is a "value-expressive" attitude? In G.R. Maio & J. M. Olson (Eds.). Why we evaluate: Functions of attitudes (pp. 249-270). Mahwah, NJ: Lawrence Earlbaum Associates. Mansfield, C. J., Mitchell, J., & King, D. E. (2002). The doctor as God’s mechanic? Beliefs in the Southeastern United States. Social Science & Medicine, 54, 399-409. Attitude Function Salience 27 Moss, N. J., Gallaread, A., Siller, J., & Klausner, J. D. (2004). “Street medicine”: Collaborating with a faith-based organization to screen at-risk youths for sexually transmitted diseases. American Journal of Public Health, 94(7), 1081-1084. Murray, S.L., Haddock, G., & Zanna, M. P. (1996). On creating value-expressive attitudes: An experimental approach. In C. Seligman, J. M. Olson, & M. P. Zanna (Eds.). The psychology of values. (107-133). Mahwah, N.J.: L. Erlbaum Associates. Parrott, R. (2004). "Collective Amnesia:" The Absence of Religious Faith and Spirituality in Health Communications Research and Practice. Health Communication, 16, 1-5. Petty, R.E., Cacioppo, J.T., & Goldman, R. (1981). Personal involvement as a determinant of argument-based persuasion. Journal of Personality and Social Psychology, 41, 847-855. Rokeach, M. (1973). The nature of human values. New York: Macmillan Publishing Co., Inc. Roskos-Ewoldsen, D.R., Arpan-Ralstin, L.A., St. Pierre, J. (2002). Attitude accessibility and persuasion: The quick and the strong. In J.P. Dillard & M. Pfau (Eds.), The persuasion handbook: Developments in theory and practice (pp. 39-61). Thousand Oaks, CA: Sage. Salmon, C., & Atkin, C. (2003). Using media campaigns for health promotion. In T. L. Thompson, A. M. Dorsey, K. I. Miller, & R. Parrott (Eds.). Handbook of health communication, (pp. 449-472). Mahwah: Earlbaum. Schwartz, S. H. (1992). Universals in the content and structure of values: Theoretical advances and empirical tests in 20 countries. Advances in Experimental Psychology, 25, 1-65. Shavitt, S. (1989). Operationalizing functional theories of attitude. In A. R. Pratkanis, S. J. Breckler, & A. G., Greenwald (Eds.), Attitude Structure and Function (pp. 311-338). Hillsdale, NJ: Lawrence Earlbaum Associates. Smith, M. B., Bruner, J. S., & White, R. (1956). Opinions and Personality. New York: Wiley & Sons. Wilcox, S., Laken, M., Anderson, T., Bopp, M., Bryant, D., Carter, R., et al. (2007). The Healthe-AME faith-based physical activity initiative: Description and baseline findings. Health Promotion Practice, 8(1), 69-78. Attitude Function Salience 28 Appendix A Santa Clara Strength of Religious Faith Scale All questions will be answered using a 1 to 5 Likert-type scale (1=strongly disagree, 5=strongly agree). 1) My religious faith is extremely important to me. 2) I pray often. 3) I look to my faith as a source of inspiration. 4) I look to my faith as providing meaning and purpose in my life. 5) I consider myself active in my faith or church. 6) My faith is an important part of who I am as a person. 7) My relationship with God is extremely important to me. 8) I enjoy being around others who share my faith. 9) I look to my faith as a source of comfort. 10) My faith impacts many of my decisions. Personal Values Survey All items will be rated using a 0 to 10 scale (0 = not at all important, 5 = somewhat important, 10 = of supreme importance) to indicate the degree to which each of these values acts as a guiding principle in the participant’s life. 1. SOCIAL POWER (control over others) 2. A SPIRITUAL LIFE (emphasis on spiritual not material matters) 3. AN EXCITING LIFE (stimulating experiences) 4. MEANING IN LIFE (a purpose in life) 5. WEALTH (material possessions, money) 6. SELF-DISCIPLINE (self-restraint, resistance to temptation) 7. FAMILY SECURITY (safety for loved ones) 8. SOCIAL RECOGNITION (respect, approval by others) 9. INDEPENDENT (self-reliant, self-sufficient) 10. MODERATE (avoiding extremes of feeling and action) 11. AMBITIOUS (hardworking, aspiring) 12. HUMBLE (modest, self-effacing) 13. CHOOSING OWN GOALS (selecting own purposes) 14. HEALTHY (not being sick physically or mentally) 15. ACCEPTING MY PORTION IN LIFE (submitting to life’s circumstances) 16. PRESERVING MY PUBLIC IMAGE (protecting my “face”) 17. OBEDIENT (dutiful, meeting obligations) 18. DEVOUT (holding to religious faith and belief) 19. RESPONSIBLE (dependable, reliable) 20. SUCCESSFUL (achieving goals) Attitude Function Salience 29 Attitude toward Exercise All questions will be answered using a 1 to 5 Likert-type scale (1=strongly disagree, 5=strongly agree). 1. 2. 3. 4. 5. 6. 7. 8. Exercise is important for maintaining good health. Exercise is too time-consuming. [Reverse code] Exercise is enjoyable. Exercise is worth the energy/effort. Exercise is something that can improve self-esteem. Exercise is only for people who need to lose weight. [Reverse code] Exercise is a good stress reliever. Exercise is only for people who need to improve their health for medical reasons. [Reverse code] 9. Exercise is an activity everyone can benefit from doing. 10. Exercise is only for people who are concerned with looking good. [Reverse code] Exercise Behavior In the past week, how many times did you exercise? In the past month, about how many times per week did you exercise? When you exercise, how many minutes do you typically spend? When you exercise, what level of intensity do you usually reach? (circle one choice) a. Low intensity (walking at a relaxed pace, gardening, cleaning house) b. Moderate intensity (walking at a fast pace, riding a bike, playing with young children) c. High intensity (jogging, swimming, playing soccer or tennis) 5. What type of exercise do you typically engage in? 1. 2. 3. 4. Value-Expressive Functionality All questions will be answered using a 1 to 5 Likert-type scale (1=strongly disagree, 5=strongly agree). 1. My attitude toward exercise reflects my personal values. 2. My attitude toward exercise tells other people a lot about the kind of person I am. 3. My attitude toward exercise expresses what I value most. 4. The fact that I value [obedience to God] is reflected in my attitude toward exercise.* 5. My attitude toward exercise shows that I value [obedience to God]. * 6. My attitude toward exercise allows me to demonstrate to others that I value [obedience to God].* 7. I gain satisfaction from knowing that my attitude toward exercise reflects that I value [obedience to God].* 8. Even if no one is aware of it, my attitude toward exercise is based on the fact that I value [obedience to God]. * 9. My attitude toward exercise is based on things other than my personal values. [recode] 10. There is no connection between my attitude toward exercise and the fact that I value [obedience to God]. [recode]* *Items contain specific value in the brackets; this material may be modified in response to data from Study 1 Attitude Function Salience 30 Appendix B Control Message (Basic Informational) [161 words] Exercise is recommended for almost everyone; because it produces many benefits that help you lead a healthy life. Exercise gives you more energy to engage in activities you enjoy. It boosts your immune system so that you can fight off sickness and disease that would keep you from living your life. Exercise strengthens your heart so that you can live life with all your heart. It can help prevent health problems like adult onset diabetes, heart disease, or obesity that would keep you from reaching your goals—whatever they may be. When you exercise, you release endorphins that put you in a good mood, and you can also shed excess weight. Both of these things help you feel better about yourself. Long story short, exercise can help you feel better inside and out. That is why exercise is recommended for almost everyone, but you should check with a physician before beginning an exercise plan to make sure it’s right for you. Value-Matched Message (Spirituality/Obedience Values Highlighted) [186 words] Exercise is recommended for almost everyone; because it produces many benefits that help you lead a healthy, spiritual life. Exercise gives you more energy to engage in spiritual disciplines like prayer, Bible reading, & serving others. It boosts your immune system so that you can fight off sickness and disease that would keep you from doing your spiritual work. Exercise strengthens your heart so that you can obey the Lord with all your heart. It can help prevent health problems like adult onset diabetes, heart disease, or obesity that would keep you from obeying God’s call on your life—wherever He might lead you. When you exercise, you release endorphins that put you in a good mood, and you can also shed excess weight. Both of these things help you feel better about yourself and therefore obey God by valuing yourself as much as He does. Long story short, exercise can help you feel better inside and out. That is why exercise is recommended for almost everyone, but you should check with a physician before beginning an exercise plan to make sure it’s right for you. Value-Mismatched Message (Wealth/Success Values Highlighted) [186 words] Exercise is recommended for almost everyone; because it produces many benefits that help you lead a healthy, wealthy life. Exercise gives you more energy to engage in exciting things like beating someone out for a promotion or getting a big raise. It boosts your immune system so that you can fight off sickness and disease that would keep you from increasing your earning potential. Exercise strengthens your heart so that you can pursue your financial goals with all your heart. It can help prevent health problems like adult onset diabetes, heart disease, or obesity that would keep you from reaching your full earning potential—whatever that might be. When you exercise, you release endorphins that put you in a good mood, and you can also shed excess weight. Both of these things help you feel better about yourself and therefore treat yourself like a million bucks. Long story short, exercise can help you feel better inside and out. That is why exercise is recommended for almost everyone, but you should check with a physician before beginning an exercise plan to make sure it’s right for you.