Age-Related Differences in Achievement Goal Differentiation Mimi Bong Korea University Validity of the 2  2 achievement goal framework for school-aged children and adolescents was examined, using self-report responses from 1,196 Korean elementary and middle school students. Confirmatory factor analysis models hypothesizing 4 distinct achievement goal factors demonstrated the best fit in all age groups. Nevertheless, achievement goals of these young students were strongly correlated with each other, regardless of the goal definition or valence. The correlation became increasingly weaker with the increasing age of the respondents. Students in Grades 1– 4 endorsed a mastery-approach goal most strongly, but those in Grades 5–9 endorsed a performance-approach goal. Performance-avoidance and mastery-avoidance goals received significantly lower average ratings than did the 2 approach goals in all age groups. Whereas both mastery-approach and performance-approach goals correlated positively with self-efficacy, strategy use, and performance in math, only the performance-approach goal correlated positively with anxiety. Anxiety also correlated positively with the 2 avoidance goals. A performance-avoidance goal further demon- strated positive correlation with help-seeking avoidance, whereas a mastery-avoidance goal did so with strategy use. Keywords: achievement goals, age differences, adolescence, mastery avoidance Students demonstrate achievement behaviors for many different reasons. For some, it is the belief that acquiring new knowledge and mastering new skills will improve their competence that leads them to invest genuine effort in learning. Others study hard with the goal of outperforming their peers, because they believe doing so is the surest way to verify their superior ability. For yet others, the primary purpose of engaging in schoolwork is to neither improve their competence nor document their superiority but, rather, to hide their inadequacy from their teachers and peers. Achievement goals refer to these underlying reasons and purposes that explain why individuals demonstrate achievement-related be- haviors in specific settings the way they do (Ames, 1992). Research on achievement goals has risen as one of the most active areas in classroom motivation research during the past 15 years (Pintrich, 2003). Early contributors in this area conceptual- ized students’ achievement goals within a dichotomous and uni- dimensional framework. Students’ orientations toward learning and understanding, developing new skills, and mastering challeng- ing tasks for the purpose of improving their competence were variously termed as learning goals (Dweck & Leggett, 1988), mastery goals (Ames, 1992), or task-involvement (Nicholls, 1984). These goals were thought to represent an adaptive end of the motivational continuum. In contrast, students’ desires to outper- form their peers and publicly validate their intellectual superiority, called performance goals (Ames, 1992; Dweck & Leggett, 1988) or ego-involvement (Nicholls, 1984), were viewed to represent a maladaptive end. Whereas the adaptive nature of a mastery goal was arguably well documented within this framework, the presumed maladap- tive nature of a performance goal was not. Across studies, stu- dents’ performance goals demonstrated negative, nonsignificant, or even positive relationships with beneficial student outcomes, such as grades. This led several researchers to call for a distinction between the approach and the avoidance properties of a perfor- mance goal (Elliot & Harackiewicz, 1996; Middleton & Midgley, 1997; Skaalvik, 1997; Urdan, 2004). Subsequent studies provided empirical support for the proposed separation. A performance- approach goal demonstrates non-negative and, more often, positive associations with students’ self-efficacy and academic perfor- mance. A performance-avoidance goal, on the contrary, displays negative associations with those adaptive outcomes but, instead, demonstrates positive associations with the maladaptive outcomes such as anxiety and use of self-defeating strategies (Elliot & Church, 1997; Middleton & Midgley, 1997; Skaalvik, 1997). This trichotomous framework is the most widely accepted view in contemporary achievement goal literature. 2  2 Achievement Goal Framework More recently, researchers such as Elliot (1999) and Pintrich (2000) argued that the reasons why students would engage in particular academic pursuits could be better understood by simul- taneously considering both their general purposes of engagement (i.e., goal valence) and the criteria they use to judge their own Mimi Bong, Department of Education and Brain and Motivation Re- search Institute, Korea University, Seoul, Korea. This research was inspired by a discussion with the late Paul R. Pintrich of the University of Michigan. I gratefully acknowledge his valuable insights, openness to new ideas, and willingness to collaborate with others with conflicting views. This work was supported by the 2007 Korea University Research Grant. Correspondence concerning this article should be addressed to Mimi Bong, Department of Education, Korea University, Anam-dong, Seongbuk-gu, Seoul 136-701, Korea. E-mail: mimibong@korea.ac.kr Journal of Educational Psychology © 2009 American Psychological Association 2009, Vol. 101, No. 4, 879 – 896 0022-0663/09/$12.00 DOI: 10.1037/a0015945 879