Relationships among attitudes and subjective norms: Testing the theory of reasoned action across cultures
Park, Hee Sun
This study investigates the relationship between attitudes toward a behavior and subjective norms in the Theory of Reasoned Action (TRA). Unique to this research, attitudes toward a behavior are divided into social and personal attitudes in order to test an explanation for the moderate to high correlation between attitudes and subjective norms reported in previous research. As expected, only social attitudes toward a behavior are significantly related to subjective norms. The finding indicates that the overlap between attitudes and subjective norms in TRA research occurs when the attitudes studied are social in nature. Additionally, members ofa collectivistic culture tend to score higher on subjective norms and social attitudes, but the high score on subjective norms and social attitudes does not necessarily contribute to predicting behavioral intention. It is suggested that cross-cultural differences on the absolute strengths of attitudes and subjective norms may not translate to the differences in the relative weights of the two components in predicting behavioral intention.
According to the Theory of Reasoned Action (TRA), people intend to behave in ways that allow them to obtain favorable outcomes and meet the expectations of others (Ajzen & Fishbein, 1980). TRA has generated a great amount of research explaining and predicting a wide variety of behaviors in many contexts. More recent studies tend to focus more on socially-relevant behaviors such as AIDS-related sexual behaviors (e.g., Fishbein, Chan, O’Reilly, Schnell, & Wood, 1992), condom uses (e.g., Baker, Morrison, Carter, & Verson, 1996; Diaz-Loving & Villagran-Vazquez, 1999), continuing professional education (e.g., Becker & Gibson, 1998), and recycling (e.g., Park, Levine, & Sharkey, 1998), just to name a few.
Literature reviews on TRA have concluded that empirical studies have generally been supportive of the theory. For example, Ryan and Bonfield (1975) reviewed 10 studies in a marketing context and reported that the average multiple correlation of attitudes and subjective norms on behavioral intentions was .60. They concluded, “the model yields stable predictions” (p. 125). In several reviews of empirical studies, Fishbein and Ajzen maintained that intentions to engage in volitional acts were usually well predicted by the combination of attitudes toward a behavior and subjective norms (Ajzen & Fishbein, 1973, 1980; Fishbein, 1980). Indeed, Ajzen and Fishbein (1973) provided a detailed review of 10 empirical tests of the theory and reported a mean of multiple correlation (R) of .81. Finally, Sheppard, Hartwick, and Warshaw’s (1988) review applied meta-analytic techniques to a more extensive research literature that provided 87 estimates of the predictability of intention and behavior. These investigators reported a mean R of .66 for the prediction of intention from attitudes and subjective norms. In addition, a meta-analysis on the relationship between attitudes and behavior reported that not only are attitudes a significant predictor of behavior, but also the relationship between attitudes and behavior is mediated by behavioral intention (Kim & Hunter, 1993a, 1993b). The findings of this body of research provide general support for TRA.
In TRA, an individual’s behavioral intention (i.e., the subjective probability of performing a behavior) is the single best predictor of whether or not he/she will engage in a behavior. Behavioral intention is, in turn, determined by a person’s attitudes toward the behavior (i.e., positive or negative evaluation of the consequences of performing the behavior) and subjective norms (i.e., perception of social pressure on performing the behavior). A person’s attitudes toward a behavior are composed of two components: (a) his/her behavioral beliefs about the outcomes a behavior is believed to yield, and (b) his/her evaluation of these outcomes (i.e., whether the consequences of the outcomes are favorable or unfavorable). Favorable consequences for highly likely outcomes of a behavior increase a person’s intention to engage in the behavior. In contrast to attitudes toward a behavior, subjective norms are a function of normative beliefs about the social expectations of significant others (e.g., spouses, parents, close friends, etc.) and an individual’s motivation to comply with those significant others. In other words, subjective norms are the perceived social pressure an individual faces when deciding whether to behave in a certain way. Many studies report that a person’s attitudes toward a behavior and subjective norms are sufficient determinants of his/her behavioral intention (e.g., Bowman & Fishbein, 1978; Goldenhar & Connell, 1992; Jaccard & Davidson, 1972;Jones, 1990; Vinokur-Kaplan, 1978).
In TRA, both personal and social factors influence an individual’s likelihood of wanting to engage in a specific behavior. Behavioral intention, which is a predictor of engaging in a behavior, is directly affected by both personal and social variables (Ajzen, 1988), the individual’s attitudes toward a behavior and subjective norms, respectively. An individual’s attitudes toward a behavior are personal because they are internally generated, based only on the individual’s beliefs about potential outcomes of a certain behavior and his/her evaluation of the behavioral outcomes. An individual’s subjective norms are social in that they are based on information external to him/her (i.e., available only from people who surround an individual) and the individual’s perceived social pressure to engage in a behavior.
Controversy exists, however, over the discriminant validity of the attitudinal and normative components. Several investigators argue that the attitudinal and normative components are not conceptually distinct, and that it is not possible to distinguish between personal and social influences on an individual’s behavioral intention (see O’Keefe, 1990, for a brief review). The argument is that the effect of an individual’s attitudes on his/her intention to undertake a behavior is contingent on subjective norms, and vice versa (e.g., Acock & DeFleur, 1972; Andrews & Kandel, 1979; Warner & DeFleur, 1969). For example, the normative component exerts a direct effect on attitudes as well as its expected direct effect on intention (Smetana & Adler, 1980). Experimental manipulation of either component has been shown to impact the other component. For example, Miniard and Cohen (1979) found that their intended manipulation of the normative component affected both the normative and attitudinal components; their intended manipulation of the attitudinal component also affected both components of the theory. Furthermore, there exists evidence of multicollinearity in measures of attitudinal and normative components (Miniard & Cohen, 1979; Warshaw, 1980). As O’Keefe (1990) pointed out, significant and large positive correlations between the attitudinal and normative components are not uncommon, ranging from .40 to .91 (e.g., Greene, Hale, & Rubin, 1997; Miniard & Cohen, 1981; Shepherd & O’Keefe, 1984). Thus, the overlap between the attitudinal and normative components of the theory has been suggested, the two components are often highly correlated, and the influences of personal and social elements on behavioral intentions are not easily differentiated (e.g., Miniard & Cohen, 1981; O’Keefe, 1990).
In refuting to these criticisms, especially with regard to the high correlation between attitudes and subjective norms, Fishbein and Ajzen (1981) contend that “attitudes and subjective norms are highly predictive of intentions and they correlated more strongly with the criterion than with each other” (p. 341). Further, Ajzen and Fishbein’s (1972) findings indicate that a manipulation designed to influence the attitudinal component has a strong effect on the attitudes measure and no significant effect on the normative measure, and vice versa. Fishbein and Ajzen (1975) conclude “it is useful to maintain the distinction between beliefs about the consequences of performing a behavior and beliefs about expectations of relevant referents” (p. 304).
One approach to resolving this controversy on the distinctiveness of the attitudinal and normative components of TRA is to analyze the nature of “attitudes” as defined in the theory. Theoretically, the attitudinal component of the theory is conceived as personal (i.e., internal) in nature while subjective norms reflect external social influence (Fishbein & Ajzen, 1975). TRA implicitly conceives all behavioral outcomes as being personal outcomes of engaging in a behavior. For example, in understanding women’s occupational orientations (i.e., pursuing a career or becoming a homemaker), Sperber, Fishbein, and Ajzen (1980) used behavioral outcomes that were mostly personal in nature (e.g., “being financially secure ..being burdened with a lot of responsibilities,” “having a glamorous life,” “doing what I want,” “enjoying life,” “having time to devote to my own goals and plans,” “using my talents wisely”). Individuals, however, may also consider social (i.e., external) outcomes. Consider, for example, the behavior of a husband taking out the garbage. He might take the garbage out to avoid smelling the bad odor coming out of the garbage bag (personal outcome) or to avoid an argument with his wife (social outcome). Likewise, some of behavioral outcomes for women’s occupation orientation may be social in nature (e.g., becoming a homemaker pleases her family, pursuing a career makes her parents proud of her, etc.), even though Sperber et al. (1980) did not include the social outcomes in their study. Behavioral outcomes can be social in addition to personal because an individual’s behavior often has consequences for other people. For some people, the personal outcomes that the behavior brings about may be more important for the decision to engage in the behavior in question. On the other hand, other people may pay more attention to behavioral outcomes that benefit others more than themselves.
Conceptualizing personal attitudes separately from social attitudes and considering them as two different components of attitudes toward a behavior may be useful for resolving the controversy over the separability of attitudinal and normative determinants. Social attitudes are determined by a person’s belief about social aspects of behavioral outcomes and likelihood of the outcomes, while subjective norms are a function of a person’s perception about other people’s approval of his/her engagement in the behavior and willingness to comply with those persons. Put differently, social attitudes involve specific behavioral outcomes that affect others, while subjective norms deal with what reference groups think about the behavior itself. However, the point that both social attitudes and subjective norms involve a person’s behavior in relation to other people may show why social attitudes should be related to subjective norms. Stated differently, the reason for the high correlation between attitudinal and normative determinants reported in some studies (e.g., Miniard & Cohen, 1981; Shepherd & O’Keefe, 1984) may possibly be because a person may consider outcomes involving others when inferring how other people think about his/her behavior or vice versa.
The distinction between social attitudes and personal attitudes also may help explain why attitudinal component is usually found to be more important for behavioral intention even when normative component is expected to be more important. For example, subjective norms can be an important factor for individual and behavioral characteristics involving other people. A small number of studies have identified individual characteristics for which subjective norms are more important than attitudes toward a behavior. Examples of such individual characteristics include male adolescents (for behavioral intention to use condoms, Greene et al., 1997); people from a collectivistic culture such as Koreans (for behavioral intention to buy sneakers, Lee & Green, 1991); people with low private self-consciousness (for behavioral intention to vote, Echabe & Garate, 1994); people with higher attention to the collective self (for multiple types of behavioral intentions, Trafimow & Finlay, 1996; Ybarra & Trafimow, 1998). However, regarding intention of condom use, attitudes were more important than subjective norms in predicting behavioral intention for South Asian males (Godin et al., 1996). Likewise, and contrary to a TRA-based expectation, for people who emphasize interdependent self-construal, attitudes were also more important for behavioral intention to recycle (Park et al., 1998). Therefore, when the expectation for the importance of normative component is not supported, it can be expected that people with higher concern for others take social attitudes seriously instead. In other words, even for people who are highly concerned with what their significant others think, the attitudinal determinant still may be more important in deciding whether they would engage in a certain behavior if some behavioral outcomes have social aspects. For example, for people high in self-consciousness and self-monitoring, social attitudes rather than nonsocial attitudes are significantly and positively correlated to behavioral measures (Miller & Grush, 1986).
In the current study, an individual’s personal attitudes toward a behavior are considered to be a function of behavioral beliefs about personal outcomes (i.e., behavioral outcomes that affect self, rather than others) and the evaluation of these personal outcomes. An individual’s social attitudes toward a behavior are determined by behavioral beliefs about social outcomes (i.e., behavioral outcomes that affect others) and the evaluation of these social outcomes. It is anticipated that when attitudes toward a behavior are divided into social and personal attitudes, it will be social attitudes toward a behavior that are responsible for the correlation between attitudinal and normative components of TRA. Therefore, it is hypothesized that
H1: Social attitudes toward a behavior are more highly correlated with subjective norms than are personal attitudes toward a behavior.
Cross- Cultural Differences in Personal and Social Attitudes Toward a Behavior
Interesting variation in behaviors occurs in cross-cultural contexts in so far as individualistic and collectivistic cultures highlight differences in personal and social attitudes. Individualism is characterized as prioritizing one’s own goal over group goals (Triandis, 1995), and members tend to decide and act on the basis of whether an action will produce personal gain (Hui & Triandis, 1986). In individualistic cultures, behavioral beliefs about outcomes that have implications for personal gain or loss may be more salient than behavioral beliefs about social outcomes. Collectivistic cultures are characterized as emphasizing group goals over personal goals (Triandis, 1989, 1995), and members prefer doing what is “right” from the perspective of the collective (Triandis, 1995). In collectivistic cultures, behavioral beliefs about outcomes that have implications for others may be more salient than behavioral beliefs about personal outcomes. For example, among Koreans, individuals’ social attitudes toward a behavior are a stronger predictor of intention to study for a final exam than personal attitudes toward the behavior (Park, 1999). Thus,
H2a: for members of a collectivistic culture, social attitudes toward a behavior will be weighted more heavily than their personal attitudes toward a behavior in predicting behavioral intention.
H2b: for members of an individualistic culture, personal attitudes toward a behavior will be weighted more heavily than their social attitudes toward a behavior in predicting behavioral intention.
METHOD
The study investigated the importance of personal versus social attitudes in predicting behavioral intention. The behavioral context involved studying for a final exam because studying behavior was a volitional behavior across different cultures. Based on survey procedure suggested by Ajzen and Fishbein (1980), a questionnaire was designed to measure each participant’s behavioral intention to study for a final exam, attitudes toward a behavior, and subjective norms. The questionnaire included both personal and social outcomes of the behavior.
Cultures Selected
Korea and the mainland U.S. were selected to test the relationships of social and personal attitudes with subjective norms and cross-cultural comparisons of personal and social attitudes. Korea has been described as one of the most collectivistic cultures and the culture of the mainland U.S. is almost always considered to be individualistic (e.g., Hofstede, 1980; Kim, 1994).
Participants
One-hundred-forty-eight undergraduate students (99 males, 48 females, and 1 unclassifiable) enrolled in various classes from Chung-Ang University in Seoul, Korea, and 150 undergraduate students (67 males, 80 males, and 3 unclassifiable) enrolled in Communication classes at Michigan State University participated in this study. The mean age was 21.96 (with roughly the same age distribution in both groups, Korea, M = 22.61; the mainland U.S., M = 21.32) and ages ranged from 18 to 40.
All of the participants from Korea were Koreans culturally as well as ethnically. Participants from the mainland U.S. included 125 Caucasians (83.3%), 13 African Americans (8.7%), 2 Filipinos (1.3%), 1 Japanese (.7%), 1 Korean (.7%), 5 other (3.30/o), and 3 unclassifiable (2.0%). Culturally, 117 participants (78.0%) indicated that they belong to the mainland U.S. culture, 15 African American (10.0%), 4 Filipino (2.7%), 2 Hispanic (1.3%), 2 Korean (1.3%), 8 other (5.3%), and 1 unclassifiable (.7%).
Procedures
In Korea, two teaching assistants recruited undergraduate university students to participate in this research on a voluntary basis. In Michigan, participation in the research was also solicited on a voluntary basis. The research was completed during regularly scheduled classes and all participants received extra credit in exchange for their participation. All participants were told that the study was investigating people’s attitudes toward studying for an exam. The questionnaires were distributed to students a few weeks prior to their final exams.
Measuring Components of TRA
The focus of the current study was the participants’ intentions to study for their upcoming final exams. The behavior of studying was chosen because studying was personally relevant to the participants (college students), and it is volitional in nature. TRA applies to volitional behaviors that people can easily perform if they decide to (Ajzen & Fishbein, 1980). In Korea and the mainland U.S., final exams usually comprise a significant portion of grades. As suggested by Fishbein and Ajzen (1975), instead of measuring intention to study in general, a specific behavior, studying more than usual for the upcoming exam, was selected to measure behavioral intention. All items used the response options recommended by Ajzen and Fishbein (1980) and a 7-point scale format.
Behavioral intention. Following the procedure suggested by Ajzen and Fishbein (1980), behavioral intention was measured with the item “I intend to study more than usual for my upcoming final exam.” Participants responded to this item on a scale bounded by extremely likely and extremely unlikely (1 = extremely unlikely, 7 = extremely likely).
Attitudinal component. Attitudes toward a behavior are a function of behavioral beliefs and outcome evaluations. Behavioral beliefs were assessed by asking participants to rate the likelihood that their studying would truly lead to each outcome (0 = extremely unlikely, 6 = extremely likely). Specific behavioral beliefs were derived by the author, based on pilot testing to obtain undergraduate students’ general beliefs about the outcomes of studying. The behavioral beliefs were divided into social outcomes and personal outcomes. Social outcomes of behavior indicated that the outcomes of behavior had something to do with others, and included getting recognition from others, cutting down on time with friends, making others proud of me, getting good grades that reflect positively on people important to me, and getting a good job that reflects positively on people important to me. Personal outcomes of behavior referred to the behavior that impacted only oneself, not others, including personally learning more, getting stressed, cutting down on time for myself, getting good grades because it is good for me, and getting a job that is good for me. Outcome evaluations were assessed by asking the participants to rate each consequence on a scale ranging from extremely good to extremely bad (scored + 3 to – 3). Social outcomes of behavior and evaluation of the outcomes were used to calculate social attitudes toward a behavior, and personal outcomes of behavior and evaluation of the outcomes were used to obtain personal attitudes toward a behavior. Social attitudes toward a behavior and personal attitudes toward a behavior were summed and used as attitudes toward a behavior.
Normative component. Subjective norms are a function of normative beliefs and motivations to comply with important others. Nine reference groups or individuals were used for determining normative beliefs and motivation to comply, as derived from previous research (Ajzen & Fishbein, 1980;jones, 1990) and by the author. The groups or individuals included parents, siblings, close friends, romantic partner, co-workers, roommates, classmates, relatives, and teachers. Normative beliefs were assessed by having the participants rate whether or not each reference group or individual thought that the participant should study for his/her upcoming exam. The motivation to comply items asked the extent to which participants wanted to do what each reference group or individual thought they should do. Both normative beliefs and motivations to comply were scored from 0 to 6 (0 = extremely unlikely, 6 = extremely likely). The normative belief for each reference group or individual and the corresponding motivation to comply rating were multiplied, and the products were summed across reference groups as subjective norms. Since only 71 participants from Korea responded to questions regarding coworkers and only 63 participants from Korea responded to questions regarding roommates, these two reference groups were omitted from the calculation of the subjective norms. This procedure was used because missing values dramatically lowered valid cases for the subjective norms (e.g., valid n = 52 for Korea and n = 141 for the mainland U.S.).
Questionnaire Preparation
The questionnaire was translated from English to Korean and back translated from Korean to English by two bilinguals. Then, a native Korean graduate student in Korea checked the Korean version of the questionnaire for its fluency and semantics. The Korean version of the questionnaire was distributed to Korean students, and the English version of the questionnaire was distributed to the participants from Michigan.
RESULTS
The purpose of this study was to explore correlation between attitudes toward a behavior and subjective norms by differentiating personal attitudes and social attitudes. The first hypothesis predicted that individuals’ social attitudes toward a behavior would be more highly correlated with their subjective norms than would their personal attitudes toward a behavior. Before testing this hypothesis, the correlation between the attitudinal and normative components was computed to ensure that the components were significantly related to each other. The correlation between attitudes toward a behavior and subjective norms was significantly positive and moderate in its magnitude overall and in both cultures, as is reported in Table 1. Table 1 also reports zero-order correlations among all the variables.
Hypothesis 1
In order to determine if the moderate correlation between the attitudinal and normative components was due to individuals’ personal attitudes toward their behavior, their social attitudes toward their behavior, or both, correlation analyses were conducted. First, the relationship between personal attitudes and social attitudes needed to be examined. If an association exists between personal and social attitudes, semi-partial correlation instead of zero-order correlation is the more appropriate way to test the relationship between social attitudes and subjective norms. If personal attitudes and social attitudes are correlated, it is not possible to see whether both are needed to predict subjective norms using zero-order correlation because there would be a spurious component in the zero-order correlation. Semi-partial correlation will reveal the unique impact that social attitudes may have on subjective norms. Personal attitudes and social attitudes were significantly correlated overall, r(291) = .47, p
Cross- Cultural Tests of TRA
TRA proposes that both individuals’ attitudes and subjective norms are important in predicting their behavioral intention. While it was not the main focus of this study, regression analysis showed that, when entered simultaneously, attitudes toward a behavior and subjective norms contributed significantly in predicting behavioral intention, F(2, 270) = 29.08, adjusted R^sup 2^ = .17, p
The findings indicated no cross-cultural differences in the relative weights of attitudes toward a behavior and subjective norms in predicting behavioral intention. Attitudes toward a behavior were the only significant predictor of behavioral intention in both cultures. Thus, three t-tests were undertaken to see if there were cross-cultural differences regarding the absolute strengths of personal attitudes, social attitudes, and subjective norms. As Table 3 indicates, t-tests showed that Koreans scored significantly higher on social attitudes toward a behavior and subjective norms than did participants in the mainland U.S., while Koreans and participants in the U.S. sample did not differ in their ratings of personal attitudes toward a behavior. Therefore, cross-cultural differences were observed in terms of the strengths of each component, not in terms of the relationship between the two components and behavioral intention.
Hypothesis 2
Hypothesis 2 predicted that personal attitudes would be the more important predictor of behavioral intention for people in the mainland U.S., while social attitudes would be the more important predictor for Koreans. To test the cross-cultural differences, three simultaneously-entered multiple regression analyses (one for all the participants and one each for the two cultures) were conducted with personal attitudes and social attitudes as predictor variables and behavioral intention as a dependent variable. As Table 4 indicates, when all the participants from the two cultures were included in the analysis, both social and personal attitudes were significant predictors of behavioral intention. When the relative predictive powers of social and personal attitudes were examined within each culture, however, only personal attitudes toward a behavior carried statistically significant variance in behavioral intention in each culture. Therefore, the data were not consistent with hypothesis 2.
In summary, while social attitudes toward a behavior are related to subjective norms, only personal attitudes toward a behavior, in this instance, predicts behavioral intention. While Koreans score higher on social attitudes than Americans, the relationship between social attitudes and behavioral intention is the same in each culture.
DISCUSSION
In the current data, when attitudinal components were divided into social attitudes toward a behavior and personal attitudes toward a behavior, only social attitudes were significantly related to subjective norms, while personal attitudes were not. Culture, as an external variable to the TRA, affected the absolute strengths of attitudinal and normative components of the theory, but not the relative weights of attitudinal and normative components in predicting behavioral intention.
Implications for the Relationship between Attitudinal and Normative Components
These results shed light on the controversy surrounding separability of attitudes toward a behavior and subjective norms. Behavioral outcomes that contained social aspects were correlated with individuals’ subjective norms, while behavioral outcomes with personal aspects failed to account for unique variance in their subjective norms. Thus, the separation of social attitudes from personal attitudes offers an explanation for the inconsistent past findings. The distinction of social attitudes from personal attitudes should be considered as one of the factors that explain the relationship between attitudes toward a behavior and subjective norms. In general, attitudes toward a behavior and subjective norms are separate. But only personal attitudes toward a behavior and subjective norms are orthogonal, while social attitudes toward a behavior and subjective norms share a social component.
Social and personal attitudes also need to be taken into consideration when constructing persuasive messages. People can change their intentions to perform a given behavior by changing the attitudinal component, the normative component, or the relative weight of the two components (see Ajzen & Fishbein, 1972; Fishbein, Ajzen, & McArdle, 1980; McCarty, 1981). Identifying conditions under which either the attitudinal or the normative component is more significantly weighted provides a guide for effective persuasion messages: one can construct persuasive messages targeted toward the more significantly weighted component. Similarly, finding conditions or variables that affect the degree of attitudinal and normative influences on action would certainly inform the study of persuasion with respect to the circumstances under which an influence strategy designed to alter the relative weights of the two components is likely to be appropriate. When the attitudinal component contains mostly social outcomes, however, an attempt to target separately either the attitudinal or normative components may not be fruitful since both of the components will share large variance. Besides looking into normative components for the effects of social influence, it might be helpful to look deep into attitudinal components for social aspects of behavioral outcomes.
Implications for Cross-Cultural Research
Cross-cultural differences were observed in terms of the absolute strengths of individuals’ social attitudes and subjective norms, but not in terms of the relative weights of their attitudes toward a behavior and subjective norms in predicting their behavioral intention. The Koreans more strongly believed that their significant others thought that they should study than did people in the mainland U.S. The fact that Koreans scored significantly higher on subjective norms than did people from the mainland U.S. may indicate that Koreans consider highly what other people think about their behavior. Contrary to the expectations, however, it was attitudes toward a behavior that had a heavier weight in determining behavioral intention for Koreans. Likewise, even though Koreans possessed stronger social attitudes toward a behavior than did people from the mainland U.S., personal attitudes were the only significant predictor of behavioral intention for people from both cultures. Thus, the current finding indicates that the increased consideration of other people’s perceptions does not necessarily influence one’s decision to engage in certain behaviors, (in this case, studying).
Interestingly, despite Koreans’ significantly greater awareness of social aspects of their behavior, there were no cross-cultural differences in what determines behavioral intention. Among the personal attitudes toward a behavior and social attitudes toward a behavior, it was personal attitudes toward a behavior that had a significant effect on predicting behavioral intention. As the results of the ANOVA indicated, Koreans had significantly higher social attitudes toward a behavior than Americans, while there was no difference on the strength of personal attitudes toward a behavior among the members of different cultures.
These findings pose implications for cross-cultural studies of communication. Many of previous cross-cultural studies relied on ANOVA or t-test analyses, which involve comparing the mean scores of responses on some type of scales given to the members of different cultures. The findings from those types of studies may indicate that there are cross-cultural differences in the absolute size of variables. For example, while Koreans may be reportedly more polite than Americans, a relative relationship between politeness and some other variables might be similar across cultures. Then, it is doubtful that the differences in self-reported perception always lead to differences in decision-making. Instead, cross-cultural studies should provide a close examination of the patterns of relationships among variables, because even when there are no cross-cultural differences in a certain variable, how the variable interacts with other variable may vary cross-culturally (e.g., Brockner & Chen, 1996).
Limitations of the Current Study and Directions for Future Research
It was expected that, even when subjective norms are not an important predictor for behavioral intention, social attitudes may be an important predictor, thereby explaining why attitudinal component is more important for people with higher concern for others (i.e., people from a collectivistic culture). The current finding did not show that social attitudes are important for people from a collectivistic culture. The lack of cross-cultural differences in the relative strengths of the components may stem from the type of behavior investigated in the current study. Studying behavior may be a behavior that is usually done alone, even though students sometimes form a study group or study in public places (e.g., study hall or library). Trafimow and Fishbein (1994) report that an individual takes subjective norms more seriously for behaviors that involve others (e.g., going to a Korean restaurant with a date) rather than only him/herself (e.g., going to a Korean restaurant alone). Considering that previous research has reported that behaviors involving others are influenced by normative components, it is not obvious whether the nature of studying behavior inherently requires personal attitudes to be more important for behavioral intention. However, questions still remain regarding whether studying behavior is exclusively personal behavior. For example, Koreans gave higher ratings on social outcomes of studying. Studying behavior can be done alone, but outcomes of the behavior can influence the relationship between a person and his/her important others (e.g., making others proud or disappointed, etc.). Future research should examine behaviors that are more social in nature (e.g., attending a protest rally, going to an extended family gathering, etc.) in order to find the role of social attitudes as well as subjective norms.
Even when a majority of people are under the influence of attitudinal components, a small of number of people are constantly under the influence of normative components for their behavioral intention (Trafimow & Finlay, 1996). Usually, a stronger relationship between the normative component and behavioral intention is found for people with higher concern for the collective self (Trafimow & Finlay, 1996) and people with stronger interdependent self-construal (Park & Levine, 1999). Thus, even though the cultural differences did not affect the relative importance of social and personal attitudes in predicting behavioral intention, there might be individual and/or personality characteristics that influence social and personal attitudes. Furthermore, future research can look into how cultural differences interact with individual characteristics for affecting the relationship between the two different types of attitudinal components and behavioral intention.
Another limitation of the current study is that the relationship between individuals’ behavioral intention and their actual behavior is not examined cross-culturally. Even though Kim and Hunter’s (1993b) meta-analysis revealed that behavioral intention mediates attitudes on behaviors and previous research indicates that behavioral intention predicts actual behavior (e.g., Bowman & Fishbein, 1978), it is not well known whether behavioral intention is related to actual behavioral pattern in cultures other than the North America. For the cross-cultural validation of the TRA, future research should examine the link between behavioral intention and behavior and the relative dominance of attitudes and subjective norms in various types of behaviors across different cultures.
Conclusion
This study investigated the relationship between attitudinal and normative components of TRA by introducing the concepts of social attitudes toward a behavior and personal attitudes toward a behavior, tested the theory cross-culturally, and compared cultural differences in each component of the theory. The findings indicate that different types of attitudes toward a behavior (social versus personal) may be able to shed light on the controversy surrounding the separability of the attitudinal and normative components. Additionally, the strengths of attitudes toward a behavior (social versus personal) and subjective norms may vary cross-culturally while the process by which behavior is engaged can remain the same. Closer examination of processes involved in communication will benefit understanding of cross-cultural differences and similarities.
REFERENCES
Acock, A. C., & DeFleur, M. L. (1972). A configurational approach to contingency consistency. American Sociological Review, 37,714-726.
Ajzen, I. (1988). Attitudes, personality and behavior. Chicago: Dorsey Press.
Ajzen, I., & Fishbein, M. (1972). Attitudes and normative beliefs as factors influencing behavioral intentions. Journal ofPersonality and Social Psychology, 21, 1-19.
Ajzen, I., & Fishbein, M. (1973). Attitudinal and normative variables as predictors of specific behaviors. Journal ofPersonality and Social Psychology, 27,41-57.
Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall.
Andrews, P., & Kandel, D. B. (1979). Attitude and behavior: A specification of the contingency consistency hypothesis. American Sociological Review, 44,298-310.
Baker, S. A., Morrison, D. M., Carter, W. B., & Verdon, M. S. (1996). Using the theory of reasoned action (TRA) to understand the decision to use condoms in an STD clinic population. Health Education Quarterly, 23, 528-542.
Becker, E. A., & Gibson, C. C. (1998). Fishbein and Ajzen’s theory of reasoned action: Accurate prediction of behavioral intentions for enrolling in distance education courses. Adult Education Quarterly, 49, 43-55.
Bowman, C. H., & Fishbein, M. (1978). Understanding public reaction to energy proposals: An application of the Fishbein model. Journal ofApplied Social Psychology, 9,319-340.
Brockner, J., & Chen, Y. R (1996). The moderating roles of self-esteem and self-construal in reaction to the self: Evidence from the People’s Republic of China and the United States. Journal of Personality and Social Psychology, 71,603-615.
Diaz-Loving, R., & Villagran-Vazquez, G. (1999). The theory of reasoned action applied to condom use and request of condom use in Mexican government workers. Applied Psychology: An International Review, 48, 139-151. Echabe, A. E., & Garate,J. F. V. (1994). Private self-consciousness as moderators of the importance of attitude and subjective norms: The prediction of voting. European Journal ofSocial Psychology, 24,285-293.
Fishbein, M. (1980). A theory of reasoned action: Some applications and implication. In H. E. Howe, Jr. & M. M. Page (Eds.), Nebraska Symposium on Motivation (Vol. 27, pp. 65-116). Lincoln, NE: University of Nebraska Press.
Fishbein, M., & Ajzen, I. (1975). Beliefs, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley.
Fishbein, M., & Ajzen, I. (1981). On construct validity: A critique of Miniard and Cohen’s paper. Journal of Experimental Social Psychology, 17,340-350.
Fishbein, M., Ajzen, I., & McArdle, J. (1980). Changing the behavior of alcoholics: Effects of persuasive communication. In I. Ajzen & M. Fishbein (Eds.), Understanding attitudes and predicting social behavior (pp. 217-242). Englewood Cliffs, NJ: Prentice-Hall.
Fishbein, M., Chan, D. K., O’Reilly, K., Schnell, D., & Wood, C. (1992). Attitudinal and normative factors as determinants of gay men’s intentions to perform AIDS-related sexual behaviors: A multisite analysis. Journal of Applied Social Psychology, 22, 999-1011.
Godin, G., Maticka-Tynadale, E., Adrien, A., Manson-Singer, S., Williams, D., & Cappon, P. (1996). Cross-cultural testing of three social cognitive theories: An application to condom use. Journal of Applied Social Psychology, 26, 1556-1586.
Goldenhar, L. M., & Connell, C. M. (1993). Understanding and predicting recycling behavior: An application of the theory of reasoned action. Journal ofEnvironmental Systems, 22,91-103.
Greene, K., Hale, J. L., & Rubin, D. L. (1997). A test of the Theory of Reasoned Action in the context of condom use and AIDS. Communication Reports, 10, 21-33.
Hofstede, G. (1980). Culture’s consequences. Beverly Hills, CA: Sage.
Hui, C. H., & Triandis, H. C. (1986). Individualism-collectivism: A study of cross-cultural researchers.Journal of Cross-Cultural Psychology 17,225-248.
Jaccard, J. J., & Davidson, A. R. (1972). Toward an understanding of family planning behaviors: An initial investigation. Journal ofApplied Social Psychology, 2, 228-235.
Jones, R. E. (1990). Understanding paper recycling in an institutionally supportive setting: An application of the theory of reasoned action.Journal ofEnvironmental Systems, 19,307-321.
Kim, M. S., & Hunter,J. E. (1993a). Attitude-behavior relations: A meta-analysis of attitudinal relevance and topic.Journal of Communication, 43, 101-142.
Kim, M. S., & Hunter, J. E. (1993b). Relationships among attitudes, behavioral intentions, and behavior: A meta-analysis of past research, part 2. Communication Research, 20,331-364.
Kim, U. (1994). Significance of paternalism and communalism in the occupational welfare system of Korean firms: A national survey. In U. Kim, H. C. Triandis, C. Kagitcibasi, S-C. Choi, & G. Yoon (Eds.). Individualism and collectivism: Theory, method, and applications (pp. 251-266). Newbury Park, CA: Sage.
Lee, C., & Green, R T. (1991). Cross-cultural examination of the Fishbein behavioral intention model.Journal ofinternational Business Studies, 22, 289-305.
McCarty, D. (1981). Changing contraceptive usage intentions: A test of the Fishbein model of intention. Journal ofApplied Social Psychology, 11, 192-211.
Miller, L. E., & Grush, J. E. (1986). Individual differences in attitudinal versus normative determination of behavior. Journal of&perimental Social Psychology, 22, 190-202.
Miniard, P. W., & Cohen,J. B. (1979). Isolating attitudinal and normative influences in behavioral intention model.Journal ofMarketing Research, 16, 102-110.
Miniard, P. W., & Cohen,J. B. (1981). An examination of the Fishbein-Ajzen behavioral-intentions model’s concepts and measure. Journal ofExperimental Social Psychology, 17,309-339.
O’Keefe, D.J. (1990). Persuasion: Theory and research. Newbury Park, CA: Sage.
Park, H. S. (1999). The theory of reasoned action and self-construal in predicting intention of studying among Korean college students. Communication Research Reorts, 13 267-279.
Park, H. S., & Levine, T. R. (1999). The theory of reasoned action and self-construal: Evidence from three cultures. Communication Monographs, 66, 199-218.
Park, H. S., Levine, T. R., & Sharkey, W. F. (1998). Understanding behavioral intention to recycle in Hawaii: The Theory of Reasoned Action and self-construals. Communication Studies, 49, 196-208.
Ryan, M. J., & Bonfield, E. H. (1975). The Fishbein extended model and consumer behavior. Journal of Consumer Research, 2, 118-136.
Shepherd, G.J., & O’Keefe, D.J. (1984). Separability of attitudinal and normative influence on behavioral intentions in the Fishbein-Ajzen model. TheJournal ofSocial Psychology, 122,287-288.
Sheppard, B. H., Hartwick,J., & Warshaw, P. RL (1988). The theory of reasoned action: A meta-analysis of past research with recommendations for modification and future research. Journal of Consumer Research, 15, 325-343.
Smetana, J. G., & Adler, N. E. (1980). Fishbein’s value x expectancy model: An examination of some assumptions. Personality and Social Psychology Bulletin, 6, 89-96.
Sperber, B. M., Fishbein, M., & Ajzen, I. (1980). Predicting and understanding women’s occupational orientations: Factors underlying choice intentions. In I. Ajzen & M. Fishbein (Eds.), Understanding attitudes and predicting social behavior (pp. 113-129). Englewood Cliffs, NJ: Prentice-Hall.
Trafimow, D., & Finlay, K. A. (1996). The importance of subjective norms for a minority of people: Between-subjects and within-subjects analysis. Personality and Social Psychology Bulletin, 22,820-828.
Trafimow, D., & Fishbein, M. (1994). The moderating effect of behavior type on the subjective normsbehavior relationship.Journal of Social Psychology, 134, 755-763.
Triandis, H. C. (1989). The self and social behavior in differing cultural contexts. Psychological Review, 96, 506-520.
Triandis, H. C. (1995). Individualism and collectivism. Boulder, CO: Westview Press.
Vinokur-Kaplan, D. (1978). To have-or not to have-another child: Family planning attitudes, intentions, and behavior.Journal ofApplied Social Psychology, 8,29-46.
Warner, L. G., & DeFleur, M. L. (1969). Attitude as an interactional concept: Social constraints and social distance as intervening variables between attitudes and action. American Sociological Reviews, 34, 153-169. Warshaw, P. R. (1980). A new model for predicting behavioral intentions: An alternative to Fishbein.Journal ofMarketing Research, 17, 153-172.
Ybarra, O., & Trafimow, D. (1998). How priming the private self or collective self affects the relative weights of attitudes and subjective norms. Personality and Social Psychology Bulletin, 24, 362-370.
Hee Sun Park (MA. 1998, University of Hawaii) is a doctoral student in the Department of Communication at the University of California at Santa Barbara. The data for the current study were part of the author’s master’s thesis (Dr. Timothy Levine, Advisor). The author would like to thank Dr. William F Sharkey, Dr. Kelly Aune, and Dr. Min-Sun Kim for their thoughtful comments as thesis committee members; Maria Lapinsky for data collection in Michigan; Dong-Wook Lee for data collection in Korea; Yoon-Kyoung Cho for her help with the translation; and Dr. David Seibold, Dr. Kathy Kellermann, Dr. Paul Mongeau, and CS reviewers for their helpful comments on earlier versions of the paper. Direct correspondence to Hee Sun Park, Department of Communication, University of California at Santa Barbara, Santa Barbara, CA 93106-4020. Address e-mail to heesun@gte. net.
Copyright Central States Speech Association Summer 2000
Provided by ProQuest Information and Learning Company. All rights Reserved
