; A Model of Motivation and Achievement in Taiwan_ Children’s Autonomy and Perceived Control in Learning
Learning Center
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

A Model of Motivation and Achievement in Taiwan_ Children’s Autonomy and Perceived Control in Learning


Educational psychologu articles

More Info
  • pg 1
									Journal of Educational Psychology 2003, Vol. 95, No. 1, 84 –96

Copyright 2003 by the American Psychological Association, Inc. 0022-0663/03/$12.00 DOI: 10.1037/0022-0663.95.1.84

Children’s Autonomy and Perceived Control in Learning: A Model of Motivation and Achievement in Taiwan
Hsiao d’Ailly
Renison College at University of Waterloo
A model of motivation and achievement was tested with data from 50 teachers and 806 Grade 4 – 6 students in Taiwan. Autonomy as a construct was shown to have ecological validity in Chinese children. The proposed model fit the data well, showing that maternal involvement and autonomy support, as well as teachers’ autonomy support, are important for children’s autonomy and perceived control. Without the mediation of perceived control, autonomy had a small negative effect on performance; controlling for perceived control, external motivation orientation was a positive predictor for Chinese children’s effort and performance. The teachers’ reported motivating style, as construed in Western research, does not correspond with Chinese children’s perceptions of their teachers nor does it have any relationship with their motivation measures.

Globalization has led to a need for increasing awareness and understanding of the differences and similarities across cultures. Although much research has been done in North America examining factors that influence students’ motivation in learning, not many parallel studies have been carried out in other countries. The present study, conducted in an Asian cultural context (Taiwan), examined factors that have been shown to be important in Western children’s motivation in learning. The variables tested include teachers’ beliefs and practices in autonomy support, children’s perceived level of parental involvement and perceived level of autonomy support from teachers and parents, children’s level of autonomy–motivation orientation in learning, children’s perceived control in school, children’s effort expenditure (diligence), and, finally, children’s academic performance. Because these variables have been shown to be important factors in children’s learning in North America (Ryan & Deci, 2000), parallel studies in a different cultural context would prove informative.

Autonomy, an Indicator of Children’s Motivation Orientation in Learning
The construct of autonomy, as described in self-determination theory (Deci & Ryan, 1985), is one of the major foci in the present
The present study was supported by Social Sciences and Humanities Research Council of Canada Grant 410-1999-0470. Travel to Taiwan to collect data was made possible by a grant from the Chinese Studies: Faculty Research Award Program for Canadians, which was sponsored by the Education Ministry in Taiwan. I thank Ching-shan Wu, the president of Taipei Municipal Teachers College, who provided generous support during my visit in Taiwan; my sister, Hsiu-min Hsiao, who was instrumental in assuring the success of my data collection; and my research assistant, Elin Moorlag, who helped enter all the data into the computer and provided many suggestions for the editing of the manuscript. Special thanks also go to the students and teachers from participating schools in Taipei. Without their patience and helpful support, this study could not have been completed. Correspondence concerning this article should be addressed to Hsiao d’Ailly, Department of Social Development Studies, Renison College, University of Waterloo, Waterloo, Ontario N2L 3G4, Canada. E-mail: hdailly@renison.uwaterloo.ca 84

study. According to self-determination theory, children’s level of autonomy in learning can be gauged from their motivation orientations. The present study examined children’s autonomy in learning, applying the notion that behaviors initiated from a more intrinsic type of motivation orientation are more self-determined, thus representing a higher level of autonomy. Self-determination theory uses the distinction between two broad types of motivation: intrinsic motivation and extrinsic motivation (Grolnick, Deci, & Ryan, 1997). Intrinsic motivation is defined as “an innate, rather than derivative, propensity to explore and master one’s internal and external worlds” (Ryan, Connell, & Grolnick, 1992, p. 169); when children are intrinsically motivated, the task is done “for its own sake or for the satisfaction inherent in the process of activity” (Ryan et al., 1992, p. 170). In contrast, extrinsic motivation is based on one’s needs to respond to socially prescribed demands, limits, and patterns of behavior. When extrinsically motivated, as described by Ryan et al. (1992), individuals behave to attain some external reward, to avoid some threat, to gain some recognition by another, or to conform to some extant value. Although most teachers strive to elicit students’ intrinsic interest in the classroom, the reality is that many learning activities at school do not in themselves elicit such interest. Much schoolwork requires self-regulation and volition on the part of students for successful learning to take place. The challenge teachers or parents face is how to facilitate the internalization of an external regulation (e.g., the obligation to do work) in such a way that allows children to be more competently self-determining. Many researchers have defined the internalization process within their theoretical frameworks (e.g., Meissner, 1981; Ryan et al., 1992; Schafer, 1968). Internalization processes, as Ryan et al. (1992) pointed out, are relevant to all behaviors and regulations whose occurrence initially depended on extrinsic incentives. As Grolnick et al. (1997) put it, internalization is a “natural developmental process in which children (as well as adolescents and adults) progressively integrate societal values and proscriptions into a coherent sense of self ” (p. 136). According to selfdetermination theory, as the process of internalization functions more fully and effectively, individuals become increasingly auton-



omous or self-determined, which brings the initially external regulations into coherence with one’s self. Ryan and Connell (1989) identified several styles of selfregulation in learning and conceptualized them as falling along a continuum of internalization: external regulation, introjected regulation, identification, and integration. Children who are externally regulated are said to do their schoolwork to stay out of trouble, to avoid punishment, or because of salient externally imposed constraints. Children who have an introjected regulatory style focus on doing their schoolwork for approval or to avoid negative selfrelated feelings, such as guilt or shame. Children who have an identified style of regulation experience actions as initiated out of his or her own values or choices and, thus, would perform schoolwork out of a personal sense of the importance of learning. The highest level of internalization is integration, in which the external regulations are integrated into a flexible, coherent hierarchical system of motives and values. The integrated level of internalization, according to Ryan et al. (1992), is developmentally advanced and thus difficult to observe in children. Ryan and Connell (1989) constructed the Self-Regulation Questionnaire: Academic (SRQ–A) to assess students’ motivation orientations in learning along the continuum from external to intrinsic, as specified in self-determination theory. The questionnaire consists of four subscales (External Motivation, Introjected Motivation, Identified Motivation, and Intrinsic Motivation), which are used to derive a Relative Autonomy Index (RAI), an indicator of the level of autonomy in children’s learning. This scale has been adopted in the present study to measure students’ autonomy. Various other instruments have been used in North American motivation research to test students’ motivation orientation in learning, such as Gottfried’s (1986) Children’s Academic Intrinsic Motivation Inventory and Harter’s (1981) Scale of Intrinsic Versus Extrinsic Motivation. As Ryan and Connell (1989) reported, mastery motivation, as measured by the Harter’s scale, was positively related to the identified and intrinsic motivation orientation in their study and negatively related to the external motivation orientation. The introjected type of motivation orientation, however, was unrelated to mastery motivation. To examine the construct validity of autonomy with the Chinese children in the present study, I included Harter’s Scale of Intrinsic Versus Extrinsic Motivation in my questionnaire.

tuting two types of beliefs: strategy beliefs (what it takes to do well) and capacity beliefs (whether the individual believes he or she has it; Connell, 1985; Skinner, 1991; Skinner, Chapman, & Baltes, 1988; Skinner & Connell, 1986). Researchers have identified combinations or profiles of beliefs that either promote or undermine children’s sense of perceived control. Students who hold beliefs that promote their sense of perceived control in academics are those who believe that effort is important for their success or failure in school and that they themselves can exert effort. Moreover, although these students do not consider ability, powerful others, or luck as important causes for academic success, they do believe that they themselves are smart, lucky, and have access to powerful others. In contrast, students who hold beliefs that undermine their sense of perceived control tend to attribute school success or failure to unknown causes and uncontrollable factors, such as luck and powerful others. They consider themselves unlucky and not having access to powerful others as well as having low ability and being incapable of exerting effort. Perceived control has been shown to be an important indicator of children’s motivation in learning; it can “account for more than 25% of the variance in teacher’s ratings of children’s engagement and disaffection in the classroom” (Patrick et al., 1993, p. 781). The construct of perceived control, as Harter (1981) explained, is conceptually and operationally distinct from motivation orientation. Motivation orientation refers to the reasons why a child performs an activity, whereas perceived control refers to the attributions children make concerning the outcomes of behavior and their beliefs in their own capacity. In Patrick et al.’s (1993) study, the researchers examined the influences perceived control and autonomy had on children’s self-reported behavior and emotion in the classroom. They found that despite the strong effects of perceived control on children’s motivation, autonomy was able to account for additional unique variance in children’s behavior and emotion in the classroom. Specifically, of the four types of motivation orientations, the identified motivation orientation was shown to have a positive and unique impact on students’ behavior after controlling for the influence of perceived control. As for children’s emotions, both intrinsic and identified motivation orientation contributed significantly to positive emotions. In contrast, the introjected motivation was related to anger and distress, and the external motivation predicted boredom.

Children’s Academic Perceived Control
Patrick, Skinner, and Connell (1993) raised an important issue about the distinctiveness of autonomy as a construct in relation to perceived control, another much studied construct in education. Arguments have been made that autonomy and perceived control are conceptually and empirically unique in their effect on students’ learning. The present study has included the testing of both constructs so that their influence on students’ learning can be examined simultaneously. Perceived control has been shown to be a powerful factor on students’ behavior and emotion in North American literature. As Skinner, Zimmer-Gembeck, and Connell (1998) described, the construct is based on a multidimensional conceptualization that integrates major theories of control, including locus of control, causal attributions, learned helplessness, and self-efficacy. Children’s perceived control in academics is conceptualized as consti-

Adults’ Influences on Children’s Motivation in Learning
Research has shown that teachers and parents have a great impact on children’s perceived control and autonomy (e.g., Connell & Wellborn, 1991; Deci, Vallerand, Pelletier, & Ryan, 1991; Grolnick, Ryan, & Deci, 1991; Skinner, 1995; Skinner et al., 1998). Various instruments have been designed in North America to examine adults’ influence on children. For example, Deci, Schwartz, Sheinman, and Ryan (1981) designed the Problems in Schools (PS) Questionnaire to assess teachers’ motivating style. The questionnaire consists of scenarios depicting a problem situation involving a child. The rationale is that teachers who decide on the problem solution themselves and try to implement the solution by using sanction (highly controlling), invoking guilt, or emphasizing that it is for the child’s own good (moderately controlling) are more control oriented, whereas teachers who encourage the child to use social comparison information (moderately



autonomous), to consider the various elements of the problems, and to arrive at a solution for him- or herself (highly autonomous) are more autonomy oriented. Deci et al. (1981) found that teachers who were more autonomy oriented were rated as such by their students. Of more importance, the children taught by autonomyoriented teachers were shown to be more intrinsically motivated and had higher self-esteem than children taught by teachers who were more control oriented. In a more recent study, Reeve, Bolt, and Cai (1999) also reported that the pre- and in-service teachers in their study who were more autonomy supportive, as measured with the PS Questionnaire, were rated or self-reported as more supportive of students’ intrinsic motivation and internalization. Grolnick et al. (1991) also designed an instrument to access children’s perceptions of their parents’ autonomy support and involvement and to examine the effect of parental influence on children’s autonomy as well as on their control understanding. The researchers found that children’s perceived parental involvement and autonomy support were positively associated with their control understanding as well as their perceptions of autonomy.

1995; Huntsinger, Jose, & Larson, 1998; Huntsinger, Jose, Larson, Krieg, & Shaligram, 2000; Steinberg, Elman, & Mounts, 1989; Stigler, Lee, Lucker, & Stevenson, 1982; Stigler, Lee, & Stevenson, 1987; Sun, 1998). Thus, children from Asian cultures are socialized differently and, therefore, may interpret and perceive events in a different way than their Western counterparts (e.g., Asakawa & Csikszentmihalyi, 1998, 2000; Eaton & Dembo, 1997; Yu & Yang, 1994). In short, autonomy and perceived control are postulated, and have been shown, to be important factors in children’s learning in North American research. However, with studies showing sharp contrasts in the motivational processes between children from different cultural backgrounds, such as the one reported by Iyengar and Lepper (1999), it is necessary to test the ecological validity of these motivation constructs and theories.

The Present Study
In the present study, various instruments developed and tested by researchers in motivation research in North America were translated into Chinese and were administered to teachers and students in Taiwan. The variables that are of interest include parents’ and teachers’ orientations toward control versus autonomy with children and parental involvement in children’s learning. Children’s motivation orientations and academic perceived control were tested. Moreover, children’s level of effort expenditure (diligence) and academic performance were assessed. The model of motivation and achievement proposed in the present study presumes that parents and teachers play an important role in children’s motivation orientation and perceived control in school, which can have a direct impact on children’s behavior and academic performance. As seen in Figure 1, the model has the following assumptions: (a) Students’ perceived autonomy support from their mothers, fathers, and teachers as well as their perceived parental involvement are correlated; (b) parental autonomy support and involvement along with teacher’s autonomy support have an impact on students’ self-regulatory style in learning (autonomy), as indicated by their RAI scores, and on their academic perceived control; (c) students’ autonomy has an effect on their perceived control; (d) students’ autonomy and perceived control both have a direct impact on their effort expenditure in school as well as their academic achievement; and (e) students’ effort expenditure affects their academic achievement. Placing autonomy and perceived control in one model allows for an examination of the unique and combined contribution these two distinct constructs have on students’ academic behavior and achievement. Children’s perceived autonomy and perceived control are not considered independent in the present study. As shown in Figure 1, the proposed model treats children’s autonomy (RAI score) as the antecedent to their level of perceived control. The reasoning is that autonomy as a construct taps into the conceptual link between intention and behaviors, that is, why a child performs an academic activity. In contrast, perceived control assesses the conceptual link between behaviors and consequences, that is, how the child attributes his or her successes or failures and how the child assesses his or her own capacity. It is postulated that with a higher sense of autonomy, children tend to have a higher level of perceived control. Thus, the model provides a test of the unique effect of autonomy (motivation orientation) on students’ effort and

Cultural Factors in Motivation in Learning
The idea of autonomy is deeply rooted in Western culture and is considered one of the essential human psychological needs (e.g., Ryan & Deci, 2000). However, researchers are now becoming more and more aware of the limitation in applying or generalizing to different cultural contexts the concepts or theories that are originated and construed mainly on the basis of Western perspectives and beliefs (Dornbusch, Ritter, Leiderman, Roberts, & Fraleigh, 1987; Iyengar, Lepper, & Ross, 1999; Steinberg, Dornbusch, & Brown, 1992). One example of strong evidence implicating cultural influences in children’s motivation to learn is shown by Iyengar and Lepper (1999). In Iyengar and Lepper’s study, children were either given an opportunity to make some choices in their own learning or were told that the choices had been made for them by trusted authority figures (e.g., mothers) or peers (in-group members). The children’s intrinsic interest in learning was measured by observing their voluntary engagement on the learning task while left alone. A striking difference was found between Anglo American children and Asian American children: Anglo American children showed less intrinsic interest in learning when choices were made for them by others than when they made their own choices; in contrast, Asian American children were most intrinsically motivated when choices were made for them by trusted authority figures or peers. Indeed, researchers interested in cross-cultural comparisons in children’s development and learning have found interesting differences in children’s behaviors as well as academic performance (e.g., Stevenson, Stigler, Lee, & Lucker, 1985; Sue & Okazaki, 1990). Various explanations have been offered for the observed discrepancies, especially those found in children from Asian cultures (e.g., d’Ailly, 1992; Kim & Chun, 1994; Sue & Okazaki, 1990). Research has shown that people from Asian cultures hold different beliefs and values than those from Western cultures (e.g., Hofstede, 1980; Markus & Kitayama, 1991; Triandis, 1994). The different beliefs and values bring about different perceptions and attribution patterns in various social contexts (see review by Bond & Smith, 1996), different child-rearing practices in the family, and different teaching practices in the school (e.g., Chen & Stevenson,



Figure 1. A proposed model of children’s motivation and achievement. e_1– e_4 associated with the variable it points to; RAI Relative Autonomy Index.

error term (error variance)

academic performance, as well as its indirect effect on these outcome variables mediated through perceived control. In summary, the major tasks in this research are first to examine the psychometric properties of the North American motivation instruments with a Chinese population and second to test the interrelationships of these measured motivational constructs. Moreover, a model of motivation and achievement is tested to examine adults’ influences on children’s motivation orientation and perceived control in school and to assess the effect of these motivational factors on children’s academic outcome.

and 6 students completed the questionnaire on their own after a brief explanation from the researcher, whereas Grade 4 students completed the questionnaire with the researcher reading each question aloud. The Problems in the Schools Questionnaire (Deci et al., 1981) was given to the teachers to complete on their own. Teachers were also asked to rate the participating students on their diligence (how hard the student works). In addition, students’ average academic scores for the current school term were obtained after the term was over at the end of July.

Teachers’ motivating style. The PS Questionnaire (Deci et al., 1981) was designed to measure adults’ orientation toward control versus autonomy with children. The questionnaire is composed of eight short vignettes describing typical kinds of problems that occur in schools. Following each vignette are four possible ways of dealing with the problem situation. The four items following each vignette represent four points along a continuum from highly controlling to highly autonomous. A 7-point Likert scale was used for teachers to indicate how appropriate they thought the solutions to the problems were. The 32-item paper-and-pencil measure has four subscales that are combined to provide an overall orientation: Highly Controlling (HC), Moderately Controlling (MC), Moderately Autonomous (MA), and Highly Autonomous (HA). A composite score was calculated by weighting the HC score with –2, the MC score with –1, the MA score with 1, and the HA score with 2 and then summing the weighted values. Thus, the total scale score could range from –18 to 18, with a higher score indicating a more autonomy-orientated motivating style. The scale has been shown to be internally consistent, with alpha ranging from .63 to .76 for the subscales, and temporally stable, with a 2-month test–retest reliability of .70 for the total scale (Deci et al., 1981). Children’s perceptions of parents and teachers. The Children’s Perceptions of Parents Scale, developed by Grolnick et al. (1991), has 21 items and was used in the present study to assess children’s perceptions of their mothers and fathers on two dimensions: involvement and autonomy versus control orientation. Each item gives two opposite statements describing the mothers or fathers as either involved or not involved, or as either autonomy supportive or controlling. The child chooses one of the two descriptions that best describes his or her parents and then decides whether the statement is “sort of true” or “really true” of his or her parent. The scale has been shown to be internally consistent, with Cronbach’s alpha ranging from .55 to .70 (Grolnick et al., 1991). Using the same format and similar item content as in the Children’s Perceptions of Parents Scale, a 6-item scale was developed for the use in the present study to assess children’s perceived autonomy support from their homeroom teachers.

Method Subjects
Fifty classes (16 fourth-grade classes, 17 fifth-grade classes, and 17 sixth-grade classes), consisting of 50 homeroom teachers and their 806 students from two different elementary schools in Taipei, Taiwan, participated in this study. Among the teacher participants, there were 45 women and 5 men. The teachers’ teaching experiences ranged from 1 to 34 years, with a median of 10 years. The students from the 50 classes were randomly selected, with approximately half of each class sampled. The sample consisted of 252 fourth-grade students (132 boys and 120 girls), 276 fifth-grade students (159 boys and 117 girls), and 278 sixth-grade students (127 boys and 151 girls).

The data collection procedure took place toward the end of the school year (the end of May and the beginning of June) in Taiwan. The survey adapted many items from existing scales that were in English. Two bilingual research assistants each independently translated all of the scale items used in this study from English to Chinese. I compared and examined the two versions of translation and used them to construct a final Chinese version. The questionnaire included items from the Children’s Perceptions of Parents Scale (Grolnick et al., 1991), the Student Perceptions of Control Questionnaire: Academic Domain (Short Form; SPOCQ–SF; Wellborn, Connell, & Skinner, 1989), the Self-Regulation Questionnaire: Academic (Ryan & Connell, 1989), and the Scale of Intrinsic Versus Extrinsic Orientation in the Classroom (Harter, 1981). Some new items were also constructed for this study to measure children’s perceived autonomy support from their homeroom teachers. The questionnaire for students was administered in groups and completed in one 30 – 40-min session. Grade 5


D’AILLY Furthermore, children’s academic records were obtained at the end of the school term, and the final average scores on various academic subjects were taken as an indication of their academic performance.

Children’s intrinsic versus extrinsic orientation. Harter’s (1981) 30item Scale of Intrinsic Versus Extrinsic Orientation in the Classroom was designed to assess five dimensions of classroom learning that have been theorized as having both an intrinsic and extrinsic motivational pole: preference for challenge versus preference for easy work assigned (Challenge subscale), curiosity–interest versus pleasing the teacher– getting grades (Curiosity subscale), independent mastery versus dependence on the teacher (Mastery subscale), independent judgment versus reliance on teacher’s judgment (Independent Judgment subscale), and internal criteria versus external criteria (Internal Criteria subscale). Harter reported good internal consistency for each subscale: Challenge, .78 –.84; Curiosity, .70 –.78; Mastery, .68 –.82; Independent Judgment, .72–.81; and Internal Criteria, .75–.83. The Challenge, Curiosity, and Mastery subscales, as Harter (1981) indicated, tap issues involving what the child wants to do, likes to do, and prefers and thus can be seen as a measurement of mastery motivation. The Independent Judgment and Internal Criteria subscales seem to tap more cognitive informational structures and can be seen as a measure of autonomous judgment. Thus, a composite score for mastery motivation was derived from the Challenge, Curiosity, and Mastery subscales. The Independent Judgment and the Internal Criteria subscales were combined to calculate a score for autonomous judgment. Children’s self-regulatory styles. Ryan and Connell’s (1989) SRQ–A contains four sections asking students to describe why they do their home work, why they work on their classwork, why they answer hard questions in class and why they try to do well in school. These reasons were differentiated along a continuum of autonomy according to selfdetermination theory: External Motivation (e.g., Because that is the rule), Introjected Motivation (e.g., Because I want the teacher to think I’m a good student), Identified Motivation (e.g., Because it’s important to me), and Intrinsic Motivation (e.g., Because it’s fun). Ryan et al. reported a moderate to high level of internal consistency for each subscale, with Cronbach’s alpha ranging from .62 to .82. The four different types of reasons for achievement-related behaviors, as measured by SRQ–A, were shown to conform to a simplex-like (ordered correlation) structure in four different samples in their study. As the items contained in the four sections of the questionnaire are quite similar, only the first two sections, which contain a total of 16 items, were used in the present study to evaluate children’s self-regulation styles. Instead of a 4-point scale like the one used in Ryan and Connell’s (1989) original scale, a 5-point Likert scale was adopted in this study for students to indicate their degree of agreement or disagreement. A summative RAI score was derived, as suggested by Ryan et al., with different weighting applied to the four subscales: –2 for External Motivation, –1 for Introjected Motivation, 1 for Identified Motivation, and 2 for Intrinsic Motivation scores. Thus, the RAI scores could range from 12 to –12 in the present study, which was taken as an indicator of children’s level of autonomy in learning. Children’s academic perceived control. The SPOCQ–SF (Wellborn et al., 1989) was adopted in this study to measure students’ control-related beliefs, including control beliefs, strategy beliefs, and capacity beliefs. Instead of using a 4-point scale, as suggested in the SPOCQ–SF, I used a 5-point scale in my survey to include a neutral point on the rating scale. The scoring procedure took this variation into account in calculating the composite scores. As suggested by Wellborn et al., two subscores were derived from this scale: promote and undermine. A maximum control score was then calculated for each student by taking the difference between the promote score and the undermine score. The maximum control score was taken as an indicator of children’s level of perceived control. Children’s effort expenditure and academic performance. The teachers in this study were asked to rate how diligent the students were. The rating was done on a 5-point Likert scale, ranging from 1 ( poor), 2 (below average), 3 (average), 4 (above average), and 5 (excellent). The ratings were taken as an indicator of children’s effort expenditure in school.

The results are divided into three parts. The first part describes the psychometric properties of the scales used in the present study. The second part examines the interrelationship between similar motivation measurements and tests the validity of these measures. Finally, the testing of the proposed structural equation model is presented. As in any survey, there were some missing data points in the data set. Composite scores were calculated using available item information. For other statistics, the missing values were dealt with by excluding cases pairwise. The AMOS program (Arbuckle, 1999) was used to conduct the regression analyses and the model testing, which applied the maximum-likelihood estimation method and included all the cases in the analyses.

The Psychometric Properties of the Scales
Teachers’ motivating style. As shown in Table 1, the internal consistency of the four subscales in this measure ranged from .65 to .73, which was comparable with the internal consistency statistics reported in Deci et al.’s (1981) study (.63 to .76). One of the HC subscale items was dropped from the analysis because of an inadequate translation. By doing so, the internal consistency of the HC scale improved from .61 to .65. The interrelations among the subscales showed a simplex pattern in which the correlations close to the diagonal were the highest. The further away from the diagonal, the smaller the correlation was. This simplex pattern confirmed the theoretically specified continuum of autonomy amongst the four scales. The 50 Chinese teachers who participated in the study had motivating style composite scores ranging from 1.04 to 10.33 (M 6.04, SD 2.48). The variability of this measurement shown in the present study was comparable to that reported by Deci et al. (1981); of the 68 teachers they tested, virtually all the scores were between 2.13 and 12.13, with one extreme score of –10.13 (SD 3.11). Children’s perception of parents and teachers. The four subscales of parental autonomy support and involvement were shown

Table 1 The Internal Consistency and Intercorrelations of the Four Subscales in the Problems in School Questionnaire Measuring Teachers’ Motivating Style
Subscale 1. 2. 3. 4. HC MC MA HA 1 .65 .47** .25 .20 2 .73 .63** .14 3 4 M 3.30 4.83 5.15 6.17 SD 0.85 0.82 0.69 0.53

.70 .34*


Note. The diagonal shows the internal consistency (Cronbach’s alpha) for each scale. HC Highly Controlling subscale; MC Moderately Controlling subscale; MA Moderately Autonomous subscale; HA Highly Autonomous subscale. * p .05, two-tailed. ** p .01, two-tailed.



to have moderate internal consistency in the present study, with Cronbach’s alpha equal to .66, .71, .70, and .62 for the four subscales, Maternal Involvement, Maternal Autonomy Support, Paternal Involvement, and Paternal Autonomy Support, respectively. These results were quite similar to the ones ( .55 to .70) reported in Grolnick et al.’s (1991) study. The scale on teachers’ autonomy support also showed a reasonable internal consistency ( .70). As shown in Table 2, children’s perceived paternal involvement appeared to be the lowest for the Grade 6 children. A multivariate analysis of variance on parental involvement, with parent as the within factor and grade as the between factor, showed a significant parent effect, F(1, 768) 207.34, p .01, MSE 0.34, and a significant grade effect, F(2, 768) 3.56, p .05, MSE 0.65. The Grade Parent interaction effect was also statistically significant, F(2, 768) 3.74, p .05, MSE 0.34. In general, the mothers were perceived as more involved than the fathers. The level of perceived paternal involvement appeared to decline along with children’s grade level, whereas the perceived maternal involvement was relatively stable across the three age groups. A similar analysis was conducted to test children’s perceived parental autonomy support. The results showed no significant

parent effect, F(1, 768) 2.09, p .05, MSE 0.32, or grade effect, F(2, 768) 2.36, p .05, MSE 0.68. The interaction effect between the two factors was not significant either, F(2, 768) 0.54, p .05, MSE 0.32. Thus, although the Chinese children rated their fathers as less involved in their schoolwork in general, their perceived paternal autonomy support was not significantly different from their perceived maternal autonomy support. In their study with American Grade 3–Grade 6 children, Grolnick et al. (1991) also found that older children perceived their parents as less involved than younger children, with the fathers decreasing in their involvement across grade level more than the mothers. The mothers were rated as significantly more involved as well as more autonomy supportive than the fathers in their study. Children’s intrinsic versus extrinsic orientation. Harter’s (1981) 30-item Scale of Intrinsic Versus Extrinsic Orientation in the Classroom examines five dimensions of classroom learning. To ease the interpretation of the results, I unified the scoring so that a high score on the scale represented a high intrinsic motivation. The results showed that the scale overall had good internal consistency, with an alpha range of .60 to .83 for each of the

Table 2 Means and Standard Deviations for the Measures Used in the Present Study
Grade 4 Variable M SD Grade 5 M SD Grade 6 M SD M Total SD

Children’s Perceptions of Parents and Teachers Scale Maternal Involvement Paternal Involvement Maternal Autonomy Support Paternal Autonomy Support Teachers’ Autonomy Support 2.90 2.59 2.76 2.64 2.61 0.71 0.76 0.75 0.66 0.63 2.91 2.47 2.73 2.54 2.39 0.68 0.72 0.78 0.68 0.65 2.87 2.35 2.68 2.51 2.48 0.65 0.71 0.71 0.69 0.68 2.89 2.46 2.72 2.56 2.49 0.67 0.73 0.75 0.68 0.66

Scale of Intrinsic Versus Extrinsic Orientation in the Classroom Challenge Curiosity Mastery Independent Judgment Internal Criteria Mastery motivationa Autonomous judgmentb 2.72 3.16 3.10 2.74 2.79 3.00 2.77 0.75 0.55 0.55 0.58 0.65 0.51 0.49 2.64 3.01 2.99 2.90 2.91 2.88 2.91 0.80 0.62 0.55 0.51 0.59 0.54 0.42 2.57 2.93 2.92 2.94 2.91 2.80 2.93 0.66 0.55 0.50 0.54 0.62 0.47 0.44 2.64 3.03 3.00 2.86 2.87 2.89 2.87 0.74 0.58 0.54 0.55 0.62 0.51 0.46

The Adapted Self-Regulation Questionnaire: Academic External Motivation Introjected Motivation Identified Motivation Intrinsic Motivation Relative Autonomy Indexc 3.10 3.06 4.12 3.41 1.68 1.02 1.06 0.84 1.13 4.18 3.11 2.95 3.95 3.08 0.95 1.10 1.04 0.87 1.15 4.47 3.15 2.85 3.73 2.81 0.21 1.05 1.02 0.95 1.09 3.99 3.12 2.95 3.93 3.09 0.92 1.06 1.04 0.90 1.15 4.25

Student Perceptions of Control Questionnaire: Academic Domain (Short Form) Promote Undermine Maximum controld

72.37 46.75 26.07

17.37 17.48 30.34

70.33 48.41 21.00

16.83 17.65 30.31

68.16 48.27 20.33

16.42 16.86 28.88

70.19 47.86 22.31

16.92 17.21 29.89

The composite scores from the Challenge, Curiosity, and Mastery subscales. b The composite scores from the Independent Judgment and Internal Criteria subscales. c The composite scores from the Intrinsic Motivation ( 2), Identified Motivation ( 1), Introjected Motivation ( 1), and External Motivation ( 2) subscales. d The indicators for children’s perceived control as calculated by taking the difference between the promote and undermine scores for each individual student.



subscales (see Table 3). The pattern of correlations among the subscales was very similar to the ones reported by Harter (1981) for New York and California samples. In general, there was a significant correlation between the three subscales measuring mastery motivation: Challenge, Curiosity, and Mastery. The two subscales assessing autonomous judgment, Internal Criteria and Independent Judgment, were found to be positively related as well. However, the scores from the Internal Criteria subscale appeared to have a higher correlation with the scores from the three subscales for mastery motivation than with those of the Independent Judgment subscale. As shown in Table 2, along with the grade level, there was a systematic downward shift in Chinese children’s preference for challenge, curiosity, and mastery. In contrast, older children’s independent judgment and internal criteria appeared to be more intrinsic than their younger counterparts. The significance testing showed that the observed grade (age) effect was statistically significant for Chinese students’ mastery motivation, F(2, 785) 8.88, p .01, MSE 0.26, as well as for their autonomous judgment, F(2, 785) 9.18, p .01, MSE 0.20. This was very similar to Harter’s (1981) findings on the American samples: Older children showed a lower level of mastery motivation than the younger ones, although their level of autonomous judgment appeared to be higher than their younger counterparts. Children’s self-regulatory style. Ryan and Connell’s (1989) SRQ–A consists of four subscales: External Motivation, Introjected Motivation, Identified Motivation, and Intrinsic Motivation. A reliability check on each subscale showed that the Intrinsic Motivation subscale ( .89) and the Identified Motivation subscale ( .82) were internally consistent. However, the External Motivation subscale ( .48) and the Introjected Motivation subscale ( .58) showed poor internal consistency. An examination of the items in the Introjected and the External Motivation subscales revealed that if the items “Because I will feel bad about myself if I don’t do it” and “Because that’s what I’m supposed to do” were excluded from the Introjected and External Motivation subscales respectively, then the internal consistency would improve substantially, with .63 for the Introjected Motivation subscale and .66 for the External Motivation subscale. One possible reason for this discrepancy could be due to the subtle differences in the translation. The item “Because that’s what I’m supposed to do” was translated as “Because that’s what I

‘should’ do” because there was no natural way to express the passive mode in Chinese in this context. The item “Because I will ‘feel bad about myself’ if I don’t do it” was translated as “Because I will be ‘unhappy with myself ’ if I don’t do it.” The small variation in the wording may have caused the difference in how children responded to these items. The results of the present study showed that these two items correlated with items in the Identified and Intrinsic Motivation subscales more than the Introjected or External Motivation subscales. Thus, when the Chinese children agreed with these two items, their motivation orientation was more intrinsic and identified than external or introjected. A composite score was calculated for each of the four subscales in the final analysis, excluding the two items mentioned above. As shown in Table 4, the four different types of reasons for achievement-related behaviors, as measured by the Chinese version of adapted SRQ–A, did conform to a simplex-like (ordered correlation) structure in this Chinese sample. As shown in Table 2, Chinese children’s RAI scores showed a downward shift along with their grade levels, F(2, 798) 8.04, p .01, MSE 17.78. This was similar to Harter’s (1981) findings on mastery motivation in their California sample of Grade 3–9 students, where there was a systematic shift from the intrinsic to the extrinsic pole in students’ motivation orientation across the grade levels. Children’s academic perceived control. The SPOCQ–SF used in the present study only had a positive and negative item associated with each subscale. Thus, no reliability check was performed for this questionnaire. Table 2 shows the descriptive statistics for the scores derived from this measurement. As expected, there was a significant negative correlation between the promote and undermine scores (r –.51, p .01). Although it seemed that Grade 4 children had a somewhat higher average maximum control score than their older counterparts, the analysis of variance showed that the grade effect was not statistically significant, F(2, 658) 2.33, p .05, MSE 889.66.

Examining the Interrelations Among the Constructs
Does teachers’ motivating style correspond to children’s perceived autonomy support from the teacher? Teachers’ autonomy support was measured in two ways in the present study: through the PS Questionnaire (Deci et al., 1981) administered to the teachers and through the Children’s Perception of Teachers Scale administered to the children. Because the measurements were taken from both the teachers and the children in the present study, a multilevel modeling approach (Bryk & Raudenbush, 1992) was used to analyze the data using the PROC MIXED procedure in the SAS program (Singer, 1998). The classes the children belonged to were used as the cluster variable. The results showed that although there was a significant class variation in children’s perceived teachers’ autonomy support (Z 4.41, p .01), teachers’ reported motivating style could not account for a significant amount of this variance, t(48) 0.21, p .05. Indeed, with the class as a unit of analysis (N 50), the results from the present study indicated that the teachers’ motivating style did not show any significant relationship with any of the children’s measures, including children’s perceived teacher autonomy support (r .04, p .05), children’s RAI (r .16, p .05), children’s mastery motivation (r .15,

Table 3 The Internal Consistency and Intercorrelations of the Five Subscales From the Scale of Intrinsic Versus Extrinsic Orientation in the Classroom
Subscale 1. 2. 3. 4. 5. Challenge Curiosity Mastery Independent Judgment Internal Criteria 1 .83 .54** .56** .08* .32** 2 .63 .43** .03 .25** 3 4 5

.61 .21** .34**

.60 .22**


Note. The diagonal shows the internal consistency (Cronbach’s alpha) for each scale. * p .05, two-tailed. ** p .01, two-tailed.



Table 4 The Intercorrelations Among the Four Subscales of the Chinese Version of Adapted SelfRegulation Questionnaire: Academic (N 801) and Their Correlations With Other Variables
Subscale or variable 1. External Motivation 2. Introjected Motivation 3. Identified Motivation 4. Intrinsic Motivation Mastery motivation Autonomous judgment Perceived control Diligence Academic performance 1 .66 .52** .13** .17** .30** .08* .22** .05 .01 2 .63 .15** .09* .08* .08* .02 .02 .03 3 4 RAI

.82 .63** .49** .01 .39** .21** .15**

.89 .51** .01 .35** .14** .08*

.54** .05 .39** .14** .10** Relative

Note. The diagonal shows the internal consistency (Cronbach’s alpha) for each subscale. RAI Autonomy Index. * p .05, two-tailed. ** p .01, two-tailed.

p .05), and children’s autonomous judgment (r –.02, p .05) as well as children’s perceived control (r .07, p .05). As described earlier, Reeve et al. (1999) recently reported questionable validity of the MA scale in their study and suggested that the MA scores be dropped from the calculation of the final composite score for the teachers’ motivating style measure. With this adjustment to the teachers’ motivating style score, however, the results in the present study remained unchanged. The modified Chinese teachers’ motivating style scores also did not have any significant correlations with any of the students’ measures. Because of the lack of concurrent and predictive validity, no further analysis was done on this measure from the teachers’ questionnaire. Does autonomy relate to mastery motivation and autonomous judgment? Both Ryan and Connell’s (1989) measure on children’s self-regulatory styles and Harter’s (1981) measure on children’s motivation orientation tapped into the reasons why a child performed an academic activity. Harter’s measurement used a polar scale of intrinsic versus extrinsic motivation, whereas Ryan et al. organized motives into four graded categories (external, introjected, identified, and intrinsic). As shown in Table 4, mastery motivation as measured by Harter’s scale was positively related to children’s identified (r .49) and intrinsic motivation (r .51) orientations and negatively related to the external (r –.30) and introjected (r –.08) orientations. Although autonomous judgment had a small negative correlation with the external and introjected self-regulatory styles, it did not seem to relate to the identified and intrinsic scores.

Indeed, Chinese children who showed a higher RAI score in Ryan et al.’s scale also tended to have a higher mastery motivation score on Harter’s scale (r .54). These findings were very similar to the ones reported in Ryan et al.’s study with American children. Adults’ influences on children’s motivation orientation. The influence of children’s perceived parental and teacher autonomy support and parental involvement was examined by regressing these parental and teacher factors onto each of the four types of motivation orientations (see Table 5). As expected, when children perceived a higher level of autonomy support from teachers and mothers, they tended to have a higher level of intrinsic and identified motivation orientation, as well as a significantly lower level of introjected and external motivation orientation. The maternal involvement also showed a significant positive contribution to children’s intrinsic and identified motivation. However, the paternal autonomy support and the paternal involvement did not add significant additional predictive power to any of the children’s motivation orientations after the mothers’ and the teachers’ influences were accounted for. Perceived control and autonomy. Using the multiple-group analyses with the AMOS 4 program, I found that the three age groups as well as the two gender groups did not differ significantly in their regression weights in any of the following analyses. When the regression weights of all the paths were set to be equivalent between groups, the chi-square statistics were not significant. Thus, the results reported below were analyzed with the combined total Chinese sample.

Table 5 Adults’ Influences on Children’s Motivation Orientations
Intrinsica Predictor Maternal autonomy support Paternal autonomy support Maternal involvement Paternal involvement Teachers’ autonomy support r2 *p

Identifiedb .10* .08 .19** .02 .20**

Introjectedc .11** 0 .08 .03 .02

Externala .20** .03 .05 .06 .19**

.12* .02 .16** 0 .20** .02.

.10. b r 2 .05. ** p

.09. .01.





I first examined the simple correlations between the four motivation orientation scores and the two outcome measures, students’ diligence and academic performance. The results showed that only the identified and intrinsic types of motivation were significantly correlated with students’ diligence (r .21 and r .14, respectively; both ps .01) and academic achievement (r .16 and r .09, respectively; both ps .01). When all four motivation orientation scores were regressed onto children’s effort expenditure and achievement, identified motivation became the only significant predictor (see Table 6, Model 1). Because perceived control has been shown to be very successful in predicting children’s learning behavior, one interesting question was whether children’s motivation orientation, the reason why they engaged in learning activity, had any additional predictive power beyond the perceived control construct. To address this question, I regressed the four motivation subscales onto children’s effort expenditure and academic performance in school, along with perceived control (see Table 6, Model 2). The results showed that children’s perceived control was indeed the strongest predictor for children’s effort as well as their academic performance; by itself, the variable accounted for 17% of the variance in effort and 18% in performance. With the influence of perceived control partialled out, both the identified ( .10, p .05) and the external ( .08, p .05) motivation orientation added a significant contribution to explaining the variance in Chinese students’ effort expenditure. As for the academic performance, the external ( .13, p .01) and the introjected ( –.09, p .05) type of motivation also added a statistically significant contribution to the prediction. It is interesting to note that perceived control aside, the influence the external motivation orientation had on Chinese students’ diligence and academic performance was positive. The additional contribution of the motivation orientation factors was quite small though, with only 1% additional variance accounted for in both instances.

model testing, noting that although all the paths in the proposed model were included in the analysis, only the significant paths are shown. As shown in Figure 2, children’s perceptions of teachers’ autonomy support ( .25, p .01) and maternal autonomy support ( .20, p .01) as well as maternal involvement ( .07, p .05) all had a significant positive impact on children’s autonomy (r 2 .14). The most important adult influence on children’s perceived control was from the mothers, with the maternal autonomy support ( .10, p .05) and involvement ( .12, p .01) both showing significant impact. Surprisingly, neither the paternal autonomy support nor the paternal involvement added significant predictive power to children’s perceived control or autonomy. With perceived control and diligence partialled out, the direct impact of students’ autonomy on their academic achievement appeared to be negative. This negative effect, albeit small, was statistically significant ( –.06, p .05). Although children’s autonomy did not add a unique contribution to their effort, it did have a significant impact on children’s perceived control ( .33, p .01); along with the maternal involvement and autonomy support, autonomy accounted for 19% of the variance in students’ level of perceived control. With the direct and indirect effects combined, the total effect of student’s autonomy on academic achievement in this model turned out to be a positive one (total effect .08). The construct, perceived control, appeared to have a significant influence on children’s effort expenditure in school ( .41, p .01, r 2 .17), which, in turn, strongly predicted children’s academic achievement ( .63, p .01). Perceived control also had a direct impact on academic achievement ( .18, p .01). All in all, perceived control showed a strong impact on students’ academic achievement in this model, with a total effect of .45.

Discussion A Model for Children’s Motivation and Achievement
The proposed model of motivation and achievement was tested using the AMOS 4 program. Table 7 shows the sample correlations as well as the implied correlations among the variables involved in the model. The results showed that the proposed model fit the data well, 2(10, N 806) 14.46, p .05, relative fit index .996, root-mean-square error .02. Figure 2 depicts the results of the The present study was designed to examine children’s autonomy and perceived control in learning in an Asian culture (Taiwan) and to test the ecological validity of motivational constructs as defined and tested in North American motivation research. In particular, the present study focused on the constructs of autonomy and perceived control and examined their antecedents and their effects in a model of motivation and achievement for children in Taiwan.

Table 6 Motivation Orientations and Perceived Control as Predictors for Children’s Diligence and Academic Performance
Diligence Model 1 Predictor Intrinsic motivation Identified motivation Introjected motivation External motivation Perceived control (for Model 2 only) r2 *p
a a

Academic performance

Model 2

Model 1c .01 .18** .08 .05

Model 2d .08 .05 .09* .13** .45**

.01 .21** .06 .01

.05 .10* .06 .08* .41**

.05. b r 2 .05. ** p

.18. .01.








Table 7 Sample Correlations and Implied Correlations for the Variables in the Proposed Model of Motivation and Achievement
Variable 1. 2. 3. 4. 5. 6. 7. 8. 9. TAC MAC PAC MI PI RAI Con Dilig Acad 1 .18 .12 .01 .05 .28 .12 .08 .07 2 .18 .38 .32 .14 .27 .24 .06 .06 3 .12 .38 .06 .17 .10 .11 .03 .04 4 .01 .32 .06 .32 .14 .22 .08 .07 5 .05 .14 .17 .32 .09 .13 .06 .08 6 .28 .26 .10 .14 .09 .39 .14 .10 7 .12 .24 .10 .22 .13 .38 .42 .42 8 .05 .10 .04 .09 .05 .14 .42 .70 9 .03 .09 .04 .09 .05 .10 .42 .70

Note. Numbers below the diagonal are correlations from the sample. Correlations at or above .10 are significant at the .01 level; correlations at or above .08 are significant at the .05 level. Numbers above the diagonal are implied correlations in the final model shown in Figure 2. TAC perceived teachers’ autonomy support; MAC perceived maternal autonomy support; PAC perceived paternal autonomy support; MI perceived maternal involvement; PI perceived paternal involvement; RAI autonomy level as indicated by the Relative Autonomy Index; Con perceived control in academics; Dilig diligence level as rated by teachers; Acad average scores from the school-term academic record.

The instruments used to measure the motivational constructs in the present study were translated and kept as similar as possible to the ones used in previous research. This allows for a comparison of the results obtained from the present study with those of the previous research literature. The findings indicate that the North American instruments tested in this study show similar psychometric properties in the Chinese population. Similar patterns are observed in the intercorrelations between the conceptually similar constructs in children’s motivation measures. Specifically, the findings show that children’s self-regulatory styles (motivation orientations), as defined by Ryan and Connell (1989), can be reliably measured in Chinese children as well. The four different types of reasons for engaging in academic behaviors (external, introjected, identified, and intrinsic) are observed in

Chinese children and can be differentiated along a continuum of autonomy. The correlations among the scales conform to a predicted simplex-like (ordered correlation) structure. Moreover, the RAI scores derived from this measurement correlate well with children’s mastery motivation, as measured by Harter’s (1981) scale. The developmental trend observed in children’s autonomy, mastery motivation, and autonomous judgment in North American children is also replicated in the present cross-sectional sample of Chinese Grade 4 – 6 children. I found that like their American counterparts, Chinese children’s mastery motivation and autonomy in learning decline along with their grade level. As Anderman and Maehr (1994) pointed out, the age-related decline in intrinsic motivation in learning during the middle grades has been well documented in the North American literature. Findings from the

Figure 2. Path estimates for the proposed model of Chinese children’s motivation and achievement. Only the significant paths in the model are shown. The value for the model fit is as follows: 2(10, N 806) 14.46, p .15, relative fit index .996, root-mean-square error .024 (90% confidence interval .000 –.048). e_1– e_4 error term (error variance) associated with the variable it points to; RAI Relative Autonomy Index. * p .05. ** p .01.



present study show that this age-related decline in intrinsic motivation in learning is a common phenomenon in the Chinese population as well. Thus, the results from the present study showed that the construct of autonomy, as defined in the self-determination theory, does have ecological validity in the Chinese population. I found that Chinese children who study because it is important to them (identified) or because it is interesting and/or fun to do the work (intrinsic) tend to, as Harter (1981) described, have a preference for challenge, work to satisfy their own interest and curiosity, and prefer to do their own work and figure out problems on their own. In contrast, children who study because it is the rule or because they want the teacher to think they are a good student tend to like the easier assignments and school subjects, do their school work to obtain good grades and to satisfy adults, and rely more on the teacher for help and guidance, particularly when it comes to figuring out problems and assignments. The theoretical model of children’s motivation and achievement proposed in the present study also fits the data collected in Taiwan. I found that students’ autonomy is mainly affected by teacher’s autonomy support and maternal autonomy support and involvement. Although there are no significant differences between children’s perceived parental autonomy support, fathers in Taiwan are rated as less involved in children’s school than the mothers. In the final analysis, paternal involvement and autonomy support did not make a unique contribution to Chinese children’s RAI after mothers’ and teachers’ influences were accounted for. Contrary to my findings, Grolnick et al. (1991) reported that paternal involvement and autonomy support both have a significant and unique impact on children’s autonomy in their North American study. Traditionally, mothers play a much more significant role than fathers in child rearing; this is probably true for both the Western and the Asian cultures. My findings, as well as those from Grolnick et al.’s (1991) study, show that children from both cultures perceive their fathers as significantly less involved than their mothers. However, it is possible that this differential parental involvement is more pronounced in Chinese families. It is also possible that the way the fathers are involved with their children and the dynamic in the parental involvement in the family is quite different in Chinese families. As Mirande (1991) pointed out, the role of fathers in different cultures has not been well researched. Further studies on the paternal role in children’s development are needed. Another interesting finding in the testing of the present model is that students’ level of autonomy does not appear to have a direct impact on their effort expenditure in school. Indeed, the only significant direct influence on students’ effort is their level of perceived control. This is not surprising given that earlier analyses in this study indicate that, with the influence of perceived control partialled out, both the external and the identified type of motivation can have a positive effect on children’s effort in Taiwan. Furthermore, when perceived control is held constant, the external type of motivation orientation is shown to have a positive effect on Chinese children’s academic performance. Thus, for children in Taiwan, interest and fun (intrinsic) or guilt and shame (introjected) may not be as strong a motivator for hard work as rules (external) and values (identified). The findings in the present study indicate that some of the high performance observed in Chinese children

can be attributed to their compliance to societal values and external pressure. The construct of autonomy (RAI) is based on the theory that intrinsic motivation is optimal for individual’s self-determination and that external motivation is the least self-determined type of motivation orientation (Ryan & Deci, 2000). Findings from the present study show that although autonomy has a positive impact on how the Chinese children see themselves (capacity belief ) and how they attribute the successes and failures (strategy belief) in their academic life, according to the present model, it does not have a direct influence on how hard a child works in school. In fact, the results show that holding perceived control and effort expenditure constant, Chinese children who have a high RAI score tend to do slightly less well in school. Thus, although the ideal scenario is that children are motivated by intrinsic interest in their learning, it is also logical to infer that students with a higher sense of autonomy, who study mainly for fun and interest and do not yield much to external pressures, are more likely to decide not to study when they cannot find fun and interest in their learning. This small direct negative effect of autonomy on Chinese children’s academic performance would have been suppressed were perceived control not included in the model because the simple correlation between autonomy and academic performance is positive. As Bollen (1989) pointed out, “the understanding of the process is far more complete when the mediating factors are explicitly incorporated” (p. 47) in a model. Thus, according to the model specified in the present study, a high level of autonomy is only facilitative for children’s academic performance when mediated by a high level of perceived control. Further studies are required to test whether this is a unique phenomena in Taiwan or whether this finding can be generalized to the North American school settings as well. In accordance with the North American literature, children’s perceived control in Taiwan is also a strong predictor of academic behaviors and achievement. In this study, children’s perceived control accounted for 17% of the variance in their effort expenditure in school. Half of the variance (51%) in children’s grades in school can be explained in the proposed model of motivation and achievement. One big discrepancy observed between these findings and the North American literature is the lack of construct validity in the measure of teachers’ motivating style. The PS Questionnaire was devised by Deci and his colleagues (1981) to gauge teachers’ style on the basis of the assumption that there is a traitlike characteristic in individuals to be more controlling or more autonomously supportive in their dealing with others. In their study with American children, Deci et al. (1981) found that children of the autonomyorientated teachers are more intrinsically motivated and have higher self-esteem than children of the teachers who are more control orientated. These results, however, are not replicated with the Chinese children in this study. Although measurements taken from the teachers in Taiwan show similar psychometric properties to those reported in North American literature, there is no correspondence between teachers’ reported autonomy versus control motivating styles and how children perceive these teachers as autonomously supportive or controlling. In fact, teachers’ motivating style as assessed in this instrument seemed to have little relationship with any of the measures of their students’ motivation.



Why is there such a lack of concurrent and predictive validity of this measure of teachers’ motivating style in the Chinese? The results from the present study clearly show that Chinese children’s perceived level of autonomy support from their teachers does have a significant effect on their motivation orientations, as shown in Figure 2. Thus, the major difference between American children and the children from Taiwan may lie in how they interpret or perceive adults’ behaviors or adults’ styles as defined in Western research. As Deci and Ryan (1980, 1985, 1987) pointed out, individuals may experience inputs relevant to the initiation or regulation of behavior differently, prompting different motivational processes. Thus, the style of teachers may not be the key indicator in its influence on students. It is how the input, or the style a teacher exhibits, is experienced or interpreted by the child that would determine the impact it has on the child’s motivation. The foregoing explanation may also apply to Iyengar and Lepper’s (1999) findings that Asian children differ from American children in their responses to the researchers’ manipulation of autonomy support. Actions or instructions from an authority or related others that can be perceived as controlling and thus decreasing a sense of autonomy in North American culture may, on the contrary, be perceived as caring or helping and experienced as an informational input in Asian cultures. As Chao (1994) pointed out, parental styles as defined by Western research often equate strictness with manifestations of parental hostility, aggression, mistrust, and dominance. This negative connotation implied in parental control may not apply in Asian contexts in which parental beliefs and parent– child relationships have different cultural roots and thus varied implications. Further cross-cultural studies on these issues will shed light on how children from different cultures may be motivated differently and learn differently under different learning contexts.

The present study is an important step in testing the ecological validity of some North American theoretical constructs in children’s motivation in an Asian culture. In general, in the Chinese population I was able to replicate the previous North American findings, especially the psychometric properties of the measurements. These findings suggest future cross-cultural studies in that parallel instruments can be used in both cultures and comparisons can be made across cultures.

Anderman, E. M., & Maehr, M. L. (1994). Motivation and schooling in the middle grades. Review of Educational Research, 64, 287–309. Arbuckle, J. L. (1999). Amos (Version 4.01) [Computer software]. Chicago: Small Waters Corporation. Asakawa, K., & Csikszentmihalyi, M. (1998). The quality of experience of Asian American adolescents in academic activities: An exploration of educational achievement. Journal of Research on Adolescence, 8, 241– 262. Asakawa, K., & Csikszentmihalyi, M. (2000). Feelings of connectedness and internalization of values in Asian American adolescents. Journal of Youth and Adolescence, 29, 121–145. Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley.

Bond, M. H., & Smith, P. B. (1996). Cross-cultural social and organizational psychology. Annual Review in Psychology, 47, 205–235. Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Applications and data analysis methods. Newbury Park, CA: Sage. Chao, R. K. (1994). Beyond parental control and authoritarian parenting style: Understanding Chinese parenting through the cultural notion of training. Child Development, 65, 1111–1119. Chen, C., & Stevenson, H. W. (1995). Motivation and mathematics achievement: A comparative study of Asian-American, CaucasianAmerican, and East-Asian high school students. Child Development, 66, 1215–1234. Connell, J. P. (1985). A new multidimensional measure of children’s perceptions of control. Child Development, 56, 1018 –1041. Connell, J. P., & Wellborn, J. G. (1991). Competence, autonomy, and relatedness: A motivational analysis of self-system processes. In M. R. Gunner & L. A. Sroufe (Eds.), Self processes and development. The Minnesota symposia on child psychology (Vol. 23, pp. 43–77). Hillsdale, NJ: Erlbaum. d’Ailly, H. (1992). Asian mathematics superiority: A search for explanations. Educational Psychologist, 27, 243–261. Deci, E. L., & Ryan, R. M. (1980). The empirical exploration of intrinsic motivational processes. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 13, pp. 39 – 80). New York: Academic Press. Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and selfdetermination in human behavior. New York: Plenum. Deci, E. L., & Ryan, R. M. (1987). The support of autonomy and the control of behavior. Journal of Personality and Social Psychology, 53, 1024 –1037. Deci, E. L., Schwartz, A. J., Sheinman, L., & Ryan, R. M. (1981). An instrument to assess adults’ orientations toward control versus autonomy with children: Reflections on intrinsic motivation and perceived competence. Journal of Educational Psychology, 73, 642– 650. Deci, E. L., Vallerand, R. J., Pelletier, L. G., & Ryan, R. M. (1991). Motivation and education: The self-determination perspective. Educational Psychologist, 26, 325–346. Dornbusch, S. M., Ritter, P., Leiderman, P. H., Roberts, D. F., & Fraleigh, M. J. (1987). The relation of parenting style to adolescent school performance. Child Development, 58, 1244 –1257. Eaton, M. J., & Dembo, M. H. (1997). Differences in the motivational beliefs of Asian American and non-Asian students. Journal of Educational Psychology, 89, 433– 440. Gottfried, A. E. (1986). Children’s Academic Intrinsic Motivation Inventory. Odessa, FL: Psychological Assessment Resources. Grolnick, W. S., Deci, E. L., & Ryan, R. M. (1997). Internalization within the family: The self-determination theory perspective. In J. E. Grusec & L. Kuczynski (Eds.), Parenting and children’s internalization of values: A handbook of contemporary theory (pp. 135–161). New York: Wiley. Grolnick, W. S., Ryan, R. M., & Deci, E. L. (1991). Inner resources for school achievement: Motivational mediators of children’s perceptions of their parents. Journal of Educational Psychology, 83, 508 –517. Harter, S. (1981). A new self-report scale of intrinsic versus extrinsic orientation in the classroom: Motivational and informational components. Developmental Psychology, 17, 300 –312. Hofstede, G. (1980). Culture’s consequences: International differences in work-related values. Beverly Hills, CA: Sage. Huntsinger, C. S., Jose, P. E., & Larson, S. L. (1998). Do parent practices to encourage academic competence influence the social adjustment of young European American and Chinese American children? Developmental Psychology, 34, 747–756. Huntsinger, C. S., Jose, P. E., Larson, S. L., Krieg, D. B., & Shaligram, C. (2000). Mathematics, vocabulary, and reading development in Chinese American and European American children over the primary school years. Journal of Educational Psychology, 92, 745–760. Iyengar, S., & Lepper, M. R. (1999). Rethinking the value of choice: A


D’AILLY ends, and agency beliefs: A new conceptualization and its measurement during childhood. Journal of Personality and Social Psychology, 54, 117–133. Skinner, E. A., & Connell, J. P. (1986). Control understanding: Suggestions for a developmental framework. In M. M. Baltes & P. B. Baltes (Eds.), The psychology of control and aging (pp. 35– 69). Hillsdale, NJ: Erlbaum. Skinner, E. A., Zimmer-Gembeck, M. J., & Connell, J. P. (1998). Individual differences and the development of perceived control. Monographs of the Society for Research in Child Development, 63(2–3), v–220. Steinberg, L., Dornbusch, S. M., & Brown, B. B. (1992). Ethnic differences in adolescent achievement. American Psychologist, 47, 723–729. Steinberg, L., Elman, J. D., & Mounts, N. S. (1989). Authoritative parenting, psychosocial maturity, and academic success among adolescents. Child Development, 60, 1424 –1436. Stevenson, H. W., Stigler, J. W., Lee, S., & Lucker, G. W. (1985). Cognitive performance and academic achievement of Japanese, Chinese, and American children. Child Development, 56, 718 –734. Stigler, J. W., Lee, S., Lucker, G. W., & Stevenson, H. W. (1982). Curriculum and achievement in mathematics: A study of elementary school children in Japan, Taiwan, and the United States. Journal of Educational Psychology, 74, 315–322. Stigler, J. W., Lee, S., & Stevenson, H. W. (1987). Mathematics classrooms in Japan, Taiwan, and the United States. Child Development, 58, 1272–1285. Sue, S., & Okazaki, S. (1990). Asian-American educational achievements: A phenomenon in search of an explanation. American Psychologist, 45, 913–920. Sun, Y. (1998). The academic success of East-Asian-American students– An investment model. Social Science Research, 27, 432– 456. Triandis, H. C. (1994). Theoretical and methodological approaches to the study of collectivism and individualism. In U. Kim, H. C. Triandis, C. Kagitcibasi, S. Choi, & G. Yoon (Eds.), Individualism and collectivism: Theory, method, and applications (pp. 41–51). London: Sage. Wellborn, J. G., Connell, J. P., & Skinner, E. A., (1989). Student Perceptions of Control Questionnaire: A new measure of perceived control in children (academic domain). Rochester, NY: University of Rochester, Graduate School of Education and Human Development. Yu, A. B., & Yang, K. S. (1994). The nature of achievement motivation in collectivist societies. In U. Kim, H. C. Triandis, C. Kagitcibasi, S. C. Choi, & G. Yoon (Eds.), Individualism and collectivism: Theory, method and applications. Cross-cultural research and methodology series (Vol. 18, pp. 239 –250). Thousand Oaks, CA: Sage.

cultural perspective on intrinsic motivation. Journal of Personality and Social Psychology, 76, 349 –366. Iyengar, S. S., Lepper, M. R., & Ross, L. (1999). Independence from whom? Interdependence with whom? Cultural perspectives on ingroups versus outgroups. In D. A. Prentice & D. T. Miller (Eds.), Cultural divides: Understanding and overcoming group conflict (pp. 273–301). New York: Russell Sage Foundation. Kim, U., & Chun, M. B. J. (1994). Educational “success” of Asian Americans: An indigenous perspective. Journal of Applied Developmental Psychology, 15, 329 –343. Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224 – 253. Meissner, W. W. (1981). Internalization in psychoanalysis. New York: International Universities Press. Mirande, A. (1991). Ethnicity and fatherhood. In F. W. Bozett & S. M. Hanson (Eds.), Fatherhood and families in cultural context (pp. 53– 82). New York: Springer. Patrick, B. C., Skinner, E. A., & Connell, J. P. (1993). What motivates children’s behavior and emotion? Joint effects of perceived control and autonomy in the academic domain. Journal of Personality and Social Psychology, 65, 781–791. Reeve, J., Bolt, E., & Cai, Y. (1999). Autonomy-supportive teachers: How they teach and motivate students. Journal of Educational Psychology, 91, 537–548. Ryan, R. M., & Connell, J. P. (1989). Perceived locus of causality and internalization: Examining reasons for acting in two domains. Journal of Personality and Social Psychology, 57, 749 –761. Ryan, R. M., Connell, J. P., & Grolnick, W. S. (1992). When achievement is not intrinsically motivated: A theory of internalization and selfregulation in school. In A. K. Boggiano & T. S. Pittman (Eds.), Achievement and motivation (pp. 167–188). New York: Cambridge University Press. Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68 –78. Schafer, R. (1968). Aspects of internalization. New York: International Universities Press. Singer, J. (1998). Using SAS PROC MIXED to fit multilevel models, hierarchical models, and individual growth models. Journal of Educational and Behavioral Statistics, 24, 323–355. Skinner, E. A. (1991). Development and perceived control: A dynamic model of action in context. In M. R. Gunner & L. A. Sroufe (Eds.), Self processes and development. The Minnesota symposia on child psychology (Vol. 23, pp. 167–216). Hillsdale, NJ: Erlbaum. Skinner, E. A. (1995). Perceived control, motivation, and coping. Thousand Oaks, CA: Sage. Skinner, E. A., Chapman, M., & Baltes, P. B. (1988). Control, means–

Received May 15, 2001 Revision received January 24, 2002 Accepted January 25, 2002

To top