Intrinsic, Extrinsic, and Amotivational Styles as Predictors of Behavior: A Prospective Study

This research ascertained the role of intrinsic, extrinsic, and amotivational styles as predictors of behavioral persistence in a real-life setting. At the beginning of the academic year, 1,042 first-tertn junior-college students enrolled in a compulsory college course completed a scale assessing intrinsic motivation, four styles of extrinsic motivation (namely, external regulation, introjection, identification, and integration), and amotivation toward academic activities. At the end of the semester, individuals who had dropped out of the course and those who had persisted were identified. Results showed that individuals who persisted in the course had reported at the beginning of the semester being more intrinsically motivated, more identified and integrated, and less amotivated toward academic activities than students who dropped out of the course. Gender differences also emerged. These revealed that females This article was prepared while the first author was supported by research grants from the Social Sciences and Humanities Research Council of Canada, the Quebec Council on Social Research, Quebec Funds for Research Aid and Researchers' Training, and the University of Quebec at Montreal. The second author was supported by a grant from the Fonds FCAR. We would like to thank Patrick Colavecchio for conducting some of the statistical analyses reported in this article. Reprint requests as well as correspondence related to the scale used in this study should be addressed to Robert J. Vallerand, Research Laboratory on Social Behavior, University of Quebec at Montreal, C. P. 8888, Station "A." Montreal, Quebec, Canada H3C 3P8. Journal of Personality 60:3, September 1992. Copyright © 1992 by Duke University Press. CCC 0022-3506/92/Sl .50 600 Vallerand and Bissonnette were more intrinsically motivated, integrated, and identified and less externally regulated and amotivated than male students. Females also displayed higher levels of behavioral persistence than males. Results are discussed in light of Deci and Ryan's (1985a) self-determination theory, and suggestions for future research are proposed. The concept of motivation has been studied from several perspectives (e.g., Freud, 1923/1962; Hull, 1943; Skinner, 1953). One perspective which has proven useful over the past 20 years suggests that behavior can be seen as intrinsically or extrinsically motivated (e.g., de Charms, 1968; Deci, 1971, 1975). Intrinsically motivated behaviors are those that are engaged in for their own sake, in other words, for the pleasure and satisfaction derived from performing them (Deci, 1971). They are activities that people voluntarily perform in the absence of material rewards or constraints (see Deci & Ryan, 1985a, 1987). To read a book for the sheer pleasure of leaming something interesting is an example of intrinsic motivation. On the other hand, extrinsic motivation pertains to a wide variety of behaviors where the goals of action extend beyond those inherent in the activity itself. They are behaviors that are engaged in as a means to an end and not for their own sakes (Deci, 1975; Kruglanski, 1978). Originally, it was thought that extrinsic motivation referred to behaviors performed in the absence of self-determination and thus which could only be prompted by external contingencies. However, more recently, Deci, Ryan, and their colleagues (Deci & Ryan, 1985a, 1987; Ryan & Connell, 1989; Ryan, Connell, & Deci, 1985; Ryan, Connell, & Grolnick, in press) have proposed that different types of extrinsic motivation exist, some of which are self-determined and may be performed through self-regulation. According to these researchers, there are four types of extrinsic motivation which can be ordered along a self-determination continuum. From lower to higher levels of self-determination, they are {a) external, {b) introjected, (c) identified, and {d) integrated regulation. External regulation occurs when behavior is externally regulated (usually through rewards or constraints). For example, students may participate in activities because they feel urged to do so by the teacher. In this case, an activity that can or should be fun is performed in order to avoid negative consequences (e.g., criticisms from the teacher). The motivation is extrinsic because the reason for participation lies outside the activity itself. Furthermore, the behavior is not chosen or Intrinsic, Extrinsic, and Amotivational Styles 601 self-determined. Extemal regulation may also be fueled by a desire for rewards. For example, students may work hard at school in order to receive a prize promised by their parents. In this case the motivation is still extrinsic and nonself-determined, but the instigating factor is the desired reward rather than a constraint. Regardless of whether the goal of behavior is to obtain rewards or to avoid sanctions, the individual experiences an obligation to behave in a specific way, and feels controlled by the reward or by the constraint (Deci & Ryan, 1985a). With introjected regulation, individuals begin to intemalize the reasons for their actions. Thus, the source of control is inside the individual. However, while internal to the person, this form of internalization is not truly self-determined since it is limited to the internalization of external contingencies. Rewards or constraints are now imposed by the individual and not by others. The individual is intemally controlling (Ryan, 1982). Thus, a student might say, "I study the night before exams because I feel guilty when I don't." Beliefs and controls are now internalized, although these are not self-determined and are experienced as pressure and tension toward specific aims. In contrast, identified regulation occurs when a behavior is valued by the individual and is perceived as being chosen by oneself. Behavior is internally regulated but in a self-determined way. An example would be students who choose to do extra work in math because they believe that this will eventually improve their ability in that particular subject. The motivation is extrinsic because the activity is not performed for itself but as a means to an end (to improve their ability in math). However, the behavior is nevertheless self-determined: Rather than being bribed into doing extra work in math, the students have chosen to do it because they feel that it would be beneficial for them. Students then experience a sense of direction and purpose, instead of obligation and pressure, in performing the behavior. The last type of extrinsic motivation is integrated regulation. At this level the person does the behavior willingly and the self-regulation is consistent with the individual's self-concept. The focus is on how the chosen extrinsically motivated behavior fits in with the rest of the person's life activities and valued goals. To the extent that there is harmony between the behavior and the individual's other facets of his or her self, there is integration. For instance, a student might say "I've decided to study for this exam and forego other interesting activities because doing well in school is important for me as a person." When there is conflict, however, the behavior is not integrated. It should be noted that it is 602 Vallerand and Bissonnette at this stage of integration that the individual experiences the greatest level of self-determination for extrinsically motivated behaviors. Apart from intrinsic and extrinsic motivation, Deci and Ryan (1985a) claim that a third construct, amotivation, must be considered to fully understand human behavior. Individuals are amotivated when they perceive a lack of contingency between their behavior and outcomes. There is an experience of incompetence and lack of control. Amotivated behaviors are neither intrinsically nor extrinsically motivated: They are nonmotivated. There are no rewards (intrinsic or extrinsic) and participation in the activity will eventually cease. Amotivated behaviors are the least self-determined because there is no sense of purpose and no expectation of reward or of the possibility of changing the course of events. Amotivation can be seen in many ways as similar to leamed helplessness (Abramson, Seligman, & Teasdale, 1978) since the individual will experience feelings of incompetence and expectancies of uncontroUability. The reader is referred to Deci and Ryan (1985a) for a more elaborate discussion of these different types of motivation. Research in this area has typically focused on intrinsic motivation and more specifically on its determinants. Results of such research generally support cognitive evaluation theory (Deci & Ryan, 1980, 1985a), in that changes in intrinsic motivation can take place either through changes in feelings of self-determination or through changes in feelings of competence. Increases or decreases in either of these processes lead to corresponding changes in intrinsic motivation (see Deci & Ryan, 1980, 1985a, for reviews). Because the various types of extrinsic motivation and amotivation have only recently been postulated, research on their determinants has been scarce (see Deci & Ryan, 1985a, 1987, 1991, for a review of available research). More recently, research has focused on the consequences and correlates of intrinsic motivation. Such research reveals that situational events known to facilitate intrinsic motivation produce greater interest (Harackiewicz, 1979; Ryan, Mims, & Koestner, 1983), more creativity (Amabile, 1979, 1982, 1983; Amabile, Hennessey, & Grossman, 1986; Koestner, Ryan, Bernieri, & Holt, 1984; Kruglanski, Friedman, & Zeevi, 1971), more cognitive flexibility (McGraw & McCullers, 1979), better conceptual leaming (Benware & Deci, 1984; Grolnick & Ryan, 1987), and a more positive emotional tone (Garbarino, 1975) than events known to be controlling. The correlates and consequences of the various forms of extrinsic motivation have been less documented but appear to be less positive both in terms of affect and performance as the Intrinsic, Extrinsic, and Amotivational Styles 603 type of extrinsic motivation is less self-determined (see Deci & Ryan, 1985a, for a review). Finally, amotivation is generally associated with impaired cognitive performance, negative affect, and, at times, low self-esteem (see Abramson et al., 1978; Peterson & Seligman, 1984, for reviews). The studies reviewed above have examined situational factors that facilitate intrinsic and extrinsic motivation as well as their consequences. More recently, researchers have started to ascertain the relation between motivational styles and various outcomes. Motivational styles refer to rather stable motivational orientations of the individual to behave in predisposed ways. Thus, to the same extent that situational variables can lead individuals to engage in a task with a specific orientation, motivational styles predispose individuals generally to engage in activities with a given orientation (e.g., intrinsic orientation). While some researchers have attempted to measure broad causality orientations across domains (e.g., Deci & Ryan, 1985b), most researchers have limited their assessment of motivational styles to specific domains, with education being the most researched domain (e.g., Harter, 1981; Ryan & Connell, 1989). Three studies (Grolnick & Ryan, 1987; Harter & Connell, 1984; Vallerand, Blais, Briere, & Pelletier, 1989) have recently explored the relationship between motivational styles and educational outcomes. In the first study, Harter and Connell (1984) showed that mastery motivation toward school was related to students' academic achievement. Unfortunately, Harter and Connell used Harter's (1981) Intrinsic/Extrinsic Motivation Scale, which pits intrinsic against extrinsic motivation. Therefore, the scale does not allow an independent assessment of these two constructs. In addition, the scale does not assess amotivation. Therefore, the role of extrinsic motivation and amotivation in educational outcomes remains unclear. In the second study, Grolnick and Ryan (1987) measured children's external, introjected, and identified regulation and intrinsic motivation toward school through the Self-Regulation Questionnaire (see Grolnick & Ryan, 1987; Ryan & Connell, 1989). Using a self-determination index (the Relative Autonomy Index) derived from the questionnaire, the authors were able to show that higher forms of self-determination were related to better conceptual learning. However, the authors did not assess the respective role of each construct in leaming. Finally, in the third study, Vallerand et al. (1989) used the Academic Motivation Scale in order to assess the concepts of amotivation, external, introjected, and 604 Vallerand and Bissonnette identified regulation, and intrinsic motivation toward school with college students. In addition, various other educational measures dealing with perceptions of competence, positive emotions, concentration, and time spent on academic tasks were also assessed. Results showed that intrinsic motivation was consistently positively associated with educational outcomes. Identified regulation was also positively related to outcomes, although not as strongly as intrinsic motivation. External regulation and introjection were either slightly negatively related or not related to outcomes. Finally, amotivation was strongly negatively correlated with educational outcomes. The above findings, and especially those of the last two studies, are encouraging because they show that intrinsic, extrinsic (especially identified regulation), and amotivational styles can be related to important outcomes. However, because outcomes in these studies were measured concomitant to the motivational style measures, it is difficult to determine the role of motivational styles in producing these outcomes. Indeed, causality cannot be inferred since both motivational styles and outcomes could be caused by a third spurious variable. Therefore, it would be important to complement these results with a prospective study, in which motivational styles are assessed at Point A and outcomes are assessed much later at Point B. While such a prospective study does not use a true experiment design per se, and thus, one should avoid speaking in terms of causality, this study would nevertheless provide a much-needed test of the predictive (rather than concomitant) effects of motivational styles on outcomes. In light of the above, the purpose of this study was to extend past research by assessing the role of intrinsic, extrinsic, and amotivational styles as predictors of future behavior using a prospective design. Specifically, the predictive effects of intrinsic, extrinsic, and amotivational styles toward academic activities on persistence in a junior-college course were assessed. We selected behavioral persistence as a dependent variable because it represents a real-life direct analog to the free-choice measure of intrinsic motivation used in laboratory research (see Deci & Ryan, 1985a). In addition to informing us on the predictive role of motivational styles in real-life outcomes, the use of behavioral persistence could also allow comparison between the findings of this study and that of previous laboratory experiments. This could help establish a parallel between the role of individual differences and that of the situation in behavior. During the second week of class, first-term French-Canadian juniorIntrinsic, Extrinsic, and Amotivational Styles 605 college students who were taking a compulsory French course completed a questionnaire assessing intrinsic, extrinsic, and amotivational styles toward academic activities. Four months later, at the end of the semester, persistent and drop-out behavior was assessed and related to the motivational styles. It was hypothesized that individuals who persisted in the course would be found to have displayed more selfdetermined motivational styles (i.e., students less amotivated, externally regulated, and introjected, but more identified, integrated, and intrinsically motivated toward academic activities) at the beginning of the term than students who dropped out of the course.