To date, homework research has been only loosely tied to theories of educational psychology and has relied mainly on time-on-task measures. The two studies (414 and 1,501 eighth graders) presented in this paper provide support for a domain-specific, multilevel model that includes an expanded effort measure, motivational predictors (expectancy and value components), learning environment variables, parental behavior variables, and stable personal characteristics such as cognitive ability and conscientiousness. Homework effort was found to be positively related to achievement. Only moderate intercorrelations were observed between the corresponding constructs of homework motivation and behavior in math and English as a foreign language. Conscientiousness and

No abstract

No abstract

Conscientiousness and domain-specific competence beliefs are known to be highly important predictors of academic effort and achievement. Given their basis in distinct research traditions, however, these constructs have rarely been examined simultaneously. Three studies with 571, 415, and 1,535 students, respectively, found a moderate association between conscientiousness and competence beliefs, but competence beliefs meaningfully predicted both conscientiousness and academic effort, irrespective of how academic effort was measured (student report or diary data). The associations of competence beliefs with academic effort were highly domain specific, whereas conscientiousness was predictive of academic effort across a wide range of academic subjects. Conscientiousness and competence beliefs were also associated with academic achievement. Cognitive ability, although associated with academic achievement, only loosely predicted academic effort.

Using a data set specifically tailored to homework research, with a sample of 1,275 students from 70 classes in Switzerland, the association between homework and achievement in French as a second language was tested at three levels (class level, between-student level, withinstudent level). The strength and direction of the homework-achievement association depended on the homework indicator chosen and differed to some degree across analytical levels. At the class level, achievement was higher in classes set frequent homework assignments and in classes where students reported low overall levels of negative emotions when doing homework. At the between-student level, high homework effort and low levels of negative homework emotions predicted favorable developments in French achievement, whereas high homework time predicted lower achievement. At the intraindividual level, high homework effort, high homework time, and low levels of negative homework emotions were statistically significantly associated with positive student evaluations of the specific assignment.Keywords: homework; frequency; time on task; effort; achievementThe Homework-Achievement Association 3 Chameleon Effects in Homework Research:The Homework-Achievement Association Depends on the Measures Used and the Level of Analysis Chosen Does homework enhance students' achievement levels? In their recent state-of-the-art meta-analysis, Cooper, Robinson, and Patall (2006) found mostly positive associations between homework time and achievement, leading them to state that "both within and across design types, there was generally consistent evidence for a positive influence of homework on achievement" (Cooper et al., 2006, p. 3). Although this conclusion adequately summarizes the existing body of research on the relationship between homework time and achievement, it may not fully reflect the complex relationship between homework and achievement (e.g., Cooper, Lindsay, Nye, & Greathouse, 1998; De Jong, Westerhof, & Creemers, 2000; Trautwein & Köller, 2003). In our view, definitive insights into the homework-achievement relationship are as yet precluded by the paucity of data collected specifically for purposes of homework research.This article has two central goals. The first is to delineate the methodological challenges that must be overcome in order to draw valid conclusions about the strength of the homeworkachievement association. We argue that this complex relationship can only properly be determined by (a) additionally including indicators of homework assignment and completion other than the time-on-homework variable, (b) distinguishing different levels of analysis (class level, between-student level, within-student level), and (c) using repeated-measurement designs. Our second goal is to document homework effects using a data set specifically tailored to homework research that was obtained from a large sample of eighth graders learning French as a second language. The Homework-Achievement AssociationThe majority of reviews of homework research ...