Effects of Scale Differences on the Generality of Academic Self-Efficacy Judgments.

Two of the most widely used academic self-efficacy assessment techniques, problem-referenced measurement and the Motivated Strategies for Learning Questionnaire (MSLQ) were compared. Participants were 383 high school students from 4 Los Angeles (California) schools. Multi-trait multi-method analyses revealed that the two techniques were not measuring exactly the same thing. In particular, students' responses became more uniform in each school subject as the assessment procedures referred to more global events than specific problems. The two techniques also differed in generality. The relationship between students' verbal and mathematics self-efficacy perceptions was noticeably stronger with the problem-referenced technique than with the MSLQ. (Contains one table, three figures, and six references.) (Author/SLD) ******************************************************************************** Reproductions supplied by EDRS are the best that can be made from the original document. ******************************************************************************** Effects of Scale Differences on the Generality of Academic Self-Efficacy Judgments Mimi Bong Ewha Womans University Seoul, Korea U.S. DEPARTMENT OF EDUCATION Office of Educational Research and Improvement EDUcATIONAL RESOURCES INFORMATION CENTER (ERIC) This document has been reproduced as received from the person or organization originating it. 0 Minor changes have been made to improve reproduction quality. Points of view or opinions stated in this document do not necessarily represent official OERI position or policy. 1 PERMISSION TO REPRODUCE AND DISSEMINATE THIS MATERIAL HAS BEEN GRANTED BY m twfvl oin TO THE EDUCATIONAL RESOURCES INFORMATION CENTER (ERIC) Abstract Two of the most widely used academic self-efficacy assessment techniques, problemreferenced measurement and the Motivated Strategies for Learning Questionnaire (MSLQ), were compared. Multi-trait multi-method analyses revealed that the two V" techniques were not measuring exactly the same thing. In particular, students' tt Ci) responses became more uniform in each school subject as the assessment procedures CO referred to more global events than specific problems. The two techniques also Cs1 differed in the generality. In particular, the relationship between students' verbal and 2 math efficacy perceptions were noticeably stronger with problem-referenced assessment technique than with the MSLQ.Two of the most widely used academic self-efficacy assessment techniques, problemreferenced measurement and the Motivated Strategies for Learning Questionnaire (MSLQ), were compared. Multi-trait multi-method analyses revealed that the two V" techniques were not measuring exactly the same thing. In particular, students' tt Ci) responses became more uniform in each school subject as the assessment procedures CO referred to more global events than specific problems. The two techniques also Cs1 differed in the generality. In particular, the relationship between students' verbal and 2 math efficacy perceptions were noticeably stronger with problem-referenced assessment technique than with the MSLQ. Poster session presented at the annual meeting of the American Educational Research Association, San Diego, CA, April 1998. Academic Self-Efficacy Scale Differences Academic self-efficacy is a predictive construct that corresponds directly to the outcome of interest (Zimmerman, 1996). Self-efficacy researchers thus assess students' efficacy beliefs toward a set of specific and particularized tasks and relate them to diverse indexes of achievement behaviors on the very tasks. It is noteworthy that most academic self-efficacy investigations so far concerned the strength of students' efficacy perception and its impact on subsequent performance. Although absolutely important in its own right, assessing strength of efficacy beliefs toward a limited range of very specific tasks does not permit researchers to make inference on the relation between various task-specific efficacy beliefs. In fact, Bandura (1986) claimed that persons' efficacy beliefs can differ along the dimensions of strength, level, and generality. A recent investigation on the generality of academic self-efficacy judgments revealed that students' efficacy perceptions indeed generalize beyond the boundary of a single, specific task. Using confirmatory factor analyses, Bong (1997) reported that high school students' judgments of their own academic competence were more or less equivalent within the boundary of each school subject. Evidence also showed that students' efficacy beliefs might generalize even further to provide bases for two higher-order factors, verbal and math academic self-efficacy. Such results can provide valuable information for both researchers and practitioners in education, because they shed light on the possibility that efforts invested in educational intervention programs, especially those designed for bolstering students' academic confidence in specific tasks, may bring added benefits to participants. Currently, there are two types of scales or measurement techniques most frequently utilized in academic self-efficacy research. One is to ask students to rate their confidence for successfully solving a set of specific problems presented. The other is to ask students to report their confidence for successful performance by presenting simply verbal descriptions of the tasks or domains of interest (see Pajares, 1996). Such difference in the measurement technique and specificity has been found to wield notable impact on efficacy beliefs' potency in explaining students' academic performance. The impact of scale differences on the generality of efficacy beliefs has yet to be investigated empirically. The purpose of the present investigation, therefore, is to see whether academic self-efficacy beliefs in the same domains assessed by different techniques demonstrate comparable degrees of generality and, more important, represent the same construct. Results from the present investigation are also expected to help tease out the sources of difference between academic self-concept and self-efficacy research. Differences in theoretical stipulation set aside, there have been conspicuous differences in the assessment procedure typically associated with the two bodies of research. Unlike academic self-efficacy research that often resorts to obtaining students' confidence ratings by presenting specific tasks or problems, academic selfconcept research has relied on more general, survey-type measurement procedures. Several researchers noted that such difference might have been at least partially responsible for the observed differences in the two constructs' relations to achievement as well as relations among domain-specific self-perceptions, particularly between the verbal and math selves (e.g., Marsh, Walker, & Debus, 1991). Method and Procedures Participants were three-hundred and eighty-three high school students (49% male, 51% female) from four Los Angeles-based high schools. The sample came from a larger study that involved 588 students (see Bong, 1997). Students (a) reported their