Self-beliefs and self-perceptions of competence have been widely investigated and debated by researchers for decades in order to explain their role in educational settings. Children and adolescents show different levels of achievement, social and cognitive engagement in learning and they show different levels of motivation. One important set of determinants of these differences are levels of self-perceptions (Bandura, 1997; Bong & Skaalvik, 2003). Self-concept and self-efficacy are two main constructs within the field of personality and social psychology as well as in educational research. An overall aim of the present study is to investigate the constructs of self-concept and self-efficacy in order to clarify in which ways they are similar and different and to investigate how these constructs contribute to students well-being in school.

Self-concept is defined as a composite view of oneself, as “the totality of the individual´s thoughts and feelings having reference to himself as an object” (Rosenberg, 1979, p.7). It is formed by interactions with the environment such as reinforcements and significant others and influenced by different frames of reference. For example, social comparisons are a major source for developing academic self-concept (Marsh, 1987). Research within the self-concept field has argued for a global construct, a general self-concept definition. However, this view of self-concept as a global construct may be difficult to define conceptually and thus hard to operationalize. More recently, self-concept studies have investigated domain-specific self-concept and found that perceived competence is related to students´ learning and there is evidence of strong relations between self-concept and school achievement (Bong & Skaalvik, 2003).

Self-efficacy is a construct aiming at capturing a person´s emotions, thoughts and actions, as self-concept is, and is formed through previous experiences. Efficacy beliefs are built on previous experiences of what has been accomplished on a certain task or in a certain situation. Self-efficacy is a measure of students´ perception of competence on a specific task, or category of tasks, such as certain mathematic problems. Students relate their perception of competence to internal aspects, and particularly to their own knowledge and skills in a certain domain (Bandura et al., 1996; Marsh, 1986). Findings suggest that self-efficacy has an impact on students´ general adaption and development, affecting for example aspirations and motivation for learning (Bandura, 1997). Reinforcements from significant others also seem to influence a person´s judgement of self-efficacy. However, self-efficacy is about individual´s expectations and convictions of what can be accomplished in a specific context or situation.     

There are similarities between self-concept and self-efficacy, for example the influence from significant others and social comparisons. However, there also are differences, and proponents of the self-efficacy construct often point at the better predictive power of assessments of self-efficacy.

Furthermore, research has indicated the need for empirically investigating and testing the reciprocal effects model (REM), a model suggesting that prior academic achievement influences subsequent self-concept and prior self-concept influences subsequent achievement (Grygiel, Modzelewski & Pisarek, 2016; Marsh & O´Mara, 2008). It seems reasonable that a reciprocal relation is evident for other self-beliefs such as self-efficacy and self-esteem. Grygiel et al. (2016) found a reciprocal relation between general academic self-concept and achievement but the influence of prior achievement on later self-concept was stronger. However, the research results are inconclusive and the lack of consistency in research indicates a need for more empirical research.

Purpose

The overall purpose of the study is to investigate the relative importance and predictive validity of self-concept and self-efficacy for educational achievement. A specific purpose is to use two different methodological approaches, latent variable models and factor score models for creating the constructs and analyzing the data. A specific question is whether these different methodological approaches give similar results.

Method Data was retrieved from the Evaluation Through Follow-up (ETF) longitudinal project containing register and questionnaire data on a large national representative sample of Swedish students born 1982, 1992, 1998 and 2004 (N = 28975). The sampling was conducted by Statistics Sweden and was a two-step stratified sampling procedure where municipalities were selected in a first step, and classes were selected in a second step. The data used are a student questionnaire from 6th Grade (age 12-13), and result from cognitive tests, information on gender and socioeconomic status (SES). The student questionnaires include a set of items asking questions about achievement and interest in different subjects (e.g., “How good do you think you are in Swedish”, “How good do you think you are in Mathematics”, “How interested are you in Mathematics”, with five response options). Five items asking about self-perceptions of academic achievement in different subjects and five items asking about interest in different subjects were selected as possible indicators of academic self-concept. The student questionnaire also included questions about perceptions of competence in solving more specific tasks, e.g.: “How well do you read and understand a text in Swedish?”, “How well do you spell in Swedish?”, “How well do you understand an English film without subtitles?”, “How well can you speak to someone in English?”, “How well can you calculate percentages?”, and “How well can you calculate area and circumference”. For these items too there were five response choices. At least three items were selected from each of the three areas Swedish, English and mathematics. Method of analyses Confirmatory factor analysis (CFA) was used to investigate the dimensionality of the different domains and the amount of correlation between self-concept and self-efficacy factors. Structural equation models (SEM) were used to investigate relations between self-concept and self-efficacy on the one hand and grades and achievement tests on the other hand. For methodological purposes individual factor scores also were estimated for all factors and results from analyses of these were compared with the SEM results.

The CFA analyses showed that it was possible to identify separate self-concept factors for achievement and interest. Separate self-efficacy factors also were found for each of the three domains Swedish, English and mathematics. However, these factors were highly correlated, and it proved possible to subsume them under a general self-efficacy factor in a bi-factor model, which also identified specific self-efficacy factors for the domains of English and mathematics. The results also showed that the academic self-concept factor was almost perfectly correlated with the general self-efficacy factor. This result may tentatively be interpreted to suggest that the differences between the concepts of self-concept and self-efficacy may primarily be a matter of choice of indicators of self-perceived competence. SEM analyses in which relations between self-concept/self-efficacy factors are related to achievement variables are currently under way.

Bandura, A. (1997). Self-Efficacy: The Exercise of Control, Freeman, New York. Bandura, A., Barbaranelli, C., Caprara, G. V., and Pastorelli, C. (1996). Multifaceted impact of self-efficacy beliefs on academic functioning. Child Development, 67, 1206–1222. Bong, M., & Skaalvik, E.M. (2003). Academic Self-Concept and Self-Efficacy: How Different Are They Really? Educational Psychology Review, 15(1), 1-40. Grygiel, P., Modzelewski, M., & Pisarek, J. (2016). Academic self-concept and achievement in Polish primary schools: cross-lagged modelling and gender-specific effects. European Journal of Psychology in Education, DOI: 10.1007/s10212-016-0300-2. Marsh, H. W. (1986). Verbal and math self-concepts: An internal/external frame of reference model. American Educational. Research Journal, 23, 129–149. Marsh, H.W. (1987). The big-fish-little-pond effect on academic self-concept. J. Educ. Psychol. 79, 280–295. Marsh, H.W., & O´Mara, A. (2008). Reciprocal effects between academic self-concept, self-esteem, achievement and attainment over seven adolescent years: unidimensional perspectives of self-concept. Personality and Social Psychology Bulletin, 34(4). 542-552. Rosenberg, M. (1979). Conceiving the Self. New York: Basic Books, New York.