Self-Concepts in Education Research Paper

1. Definition Of The Term ‘Self-Concept’

The term ‘self-concept’ is used in two interrelated ways. Talking about the self-concept of a person has to be differentiated from referring to a number of different self-concepts of this person. Usage of the term in self-concept research implies that ‘the’ self-concept may be defined as the total set of a person’s cognitive representations of him or herself which are stored in memory. In the plural sense, different self-concepts of a person are subsets of such representations relating to different self-attributes (like abilities, physical appearance, social status, etc.). In both variants, the term implies that self-concepts are self-representations which are more or less enduring over time, in contrast to situational self-perceptions. The definition is open as to whether these representations refer to factual reality of personal attributes, or to fictitious (e.g., possible or ideal) attributes.

There are two definitional problems which remain conceptually unresolved and impede scientific progress. First, it is unclear whether self-related emotions should be included or not. One may argue that emotions are different from cognition and should be regarded as separate constructs. From such a perspective, (cognitive) self-concepts and (emotional) self-related feelings might be subsumed under umbrella constructs like self-esteem, comprising both cognitive and affective facets, but should not be mixed up conceptually. On the other hand, prominent measures of the construct take more integrative perspectives by combining cognitive and affective self-evaluative items. One example are H. Marsh’s widely used Self-Description Questionnaires (SDQ: Marsh 1988) measuring academic self-concepts both by cognitive items (e.g., ‘I am good at mathematics’) and by affective items (e.g., ‘I enjoy doing work in mathematics).’

Second, there is no common agreement on the conceptual status of self-related cognitions linking the self to own actions and the environment. An example are self-efficacy beliefs pertaining to own capabilities to be successful in solving specific types of tasks, thus cognitively linking the self (own capabilities) to own actions (task performance) and to environmental demands (a domain of tasks; cf. Pajares 1996; see Self-efficacy: Educational Aspects). It may be argued that such self-representations should be regarded as part of a person’s self-concept as well. However, terms like self-concept, on the one hand, and self-efficacy beliefs, control beliefs etc., on the other, have been used as if they were relating to conceptually distinct phenomena, and have been addressed by different traditions of research. These research traditions tend to mutually ignore each other in spite of overlapping constructs and parallel findings, implying that more ‘cross-talk’ among researchers should take place in order to reduce conceptual confusion and the proliferation of redundant constructs which still prevails at the beginning of the twenty-first century (Pajares 1996).

2. Facets And Structures Of Self-Concepts

Self-related cognitions can refer to different attributes of the self, and can imply descriptive or evaluative perspectives pertaining to the factual or nonfactual (e.g., ideal) reality of these attributes. Research has begun to analyze the structures of these different representational facets.

2.1 Representations Of Attributes: Hierarchies Of Self-Concepts

Self-concepts pertaining to different attributes may vary along two dimensions: the domain of attributes which is addressed, and their generality. It may be theorized that self-concepts are hierarchically organized along these two dimensions in similar ways as semantic memory networks can be assumed to be, implying that more general self-concepts are located at top levels of the self-concept hierarchy, and more specific self-concepts at lower levels. In educational research, the hierarchical model put forward by Shavelson et al. (1976) stimulated studies on facets of self-concepts differing in generality. This model implied that a ‘general self-concept’ is located at the top of the hierarchy; an ‘academic self-concept’ pertaining to one’s academic capabilities as well as social, emotional, and physical nonacademic self-concepts at the second level; self-concepts relating to different school subjects and to social, emotional, and physical subdomains at the third level; and more specific self-concepts implying evaluations of situation-specific behavior at lower levels.

Studies showed that correlations between academic self-concepts pertaining to different subjects tend to be near zero, in contrast to performance which typically is correlated positively across academic domains. An explanation has been provided by Marsh’s internal– external frame of reference model (I E model) positing that the formation of academic self-concepts may be based on internal standards of comparison implying within-subject comparison of abilities across domains, as well as external standards implying between- subjects social comparison with other students (Marsh 1986). Applying internal standards may lead to negative correlations between self-concepts pertaining to different domains (e.g., perceiving high ability in math may lead to lowered estimates of verbal abilities, and vice versa). External standards would imply positive correlations since achievement in domains like mathematics and languages is positively correlated across students. The model assumes that students use both types of standards, implying that opposing effects may cancel out. Accordingly, the Shavelson et al. (1976) model has been reformulated by assuming that students hold math-related and verbal-related academic self-concepts, but no general academic self-concept (Marsh and Shavelson 1985).

2.2 Different Perspectives On Attributes

Self-representations can imply descriptive as well as evaluative accounts of attributes (e.g., ‘I am tall’ vs. ‘I am an attractive woman’). In academic self-concepts, this distinction may often be blurred because descriptions of academic competence may use comparison information implying some evaluation as well (e.g., ‘I am better at math than most of my classmates’). Any subjective evaluation of academic competence may use a number of different standards of evaluation, e.g. (a) social comparison standards and (b) intraindividual comparison across domains as addressed by Marsh’s I/E model (see above), as well as (c) intraindividual comparison of current and past competence (implying an evaluation of one’s academic development), (d) mastery-oriented comparison relating one’s competence to content-based criteria of minimal or optimal performance in an academic domain, and (e) cooperative standards linking individual performance to group performance.

Beyond existing attributes (‘real self’), self-representations can pertain to attributes of the self which do not factually exist, but might exist (‘possible selves’), are wanted by oneself (‘ideal self’), wanted by others (‘ought self’), etc. Concepts of nonreal selves may be as important for affect and motivated self-development as concepts of the real self. For example, perceived discrepancies between ideal and real self may be assumed to trigger depressive emotion, whereas discrepancies between ought and real self may give rise to anxiety (Higgins 1987).

3. Self-Concepts And Academic Achievement

The relation between self-concept and academic achievement is one of the most often analyzed problems in both self-concept and educational research. In the first stage of research on this problem, connections between the two constructs were analyzed by correlational methods based on cross-sectional field studies. Results implied that self-concepts and achievement may be positively linked. However, the magnitude of the correlation proved to depend on the self-concept and achievement constructs used. For example, in the meta-analysis provided by Hansford and Hattie (1982), the average correlation between self-concept and achievement/performance measures was r = 0.22 for general self-esteem, and r = 0.42 for self-concept of ability. When both self-concepts and achievement are measured in more domain-specific ways, correlations tend to be even higher (e.g., correlations for self-concepts and academic achievement in specific school subjects). This pattern of relations implies that the link gets closer when self-concept and criterion measure are matched, and when both are conceptualized in domain-specific ways.

In the second stage, researchers began to analyze the causal mechanisms producing these relations. From a theoretical perspective, the ‘skill development’ model maintained that academic self-concepts are the result of prior academic achievement, whereas the ‘self-enhancement’ approach implied that self-concepts influence students’ achievement (cf. Helmke and van Aken 1995). In a number of longitudinal studies, both hypotheses were tested competitively by using cross-lagged panel analysis and structural equations modeling. Results showed that both hypotheses may be valid, thus implying that self-concepts and achievement may be linked by reciprocal causation: Prior academic achievement exerted effects over time on subsequent academic self-concepts, and self-concepts influenced later achievement, although effect sizes for the latter causal path were less strong and consistent (cf. Helmke and van Aken 1995, Marsh and Yeung 1997). This evidence suggests that academic self-concepts are partly based on information about own academic achievement, and may contribute to later academic learning and achievement.

Not much is known about the exact nature of these mechanisms. Investigators have just begun to explore possible mediators and moderators of self-concept achievement relations. Judging from theory and available evidence, the following may be assumed.

3.1 Effects Of Achievement On Self-Concepts

Academic feedback of achievement (e.g., by grades) may underly the formation of self-representations of academic capabilities. The match between feedback and self-representations may depend on the cumulativeness and consistency of feedback, its salience, the reference norms used, and the degreee of congruency with competing information from parents, peers, or one’s past. This implies that the impact of achievement on self-concepts may differ between schools, teachers, and classrooms using different standards and practices of feedback (cf. Helmke 1992). Finally, beyond exerting effects of academic self-concepts, achievement feedback can be assumed to influence students’ general sense of self-worth (Pekrun 1990), thus affecting students’ overall psychological health and personality development as well.

3.2 Effects Of Self-Concepts On Achievement

Self-concepts may influence the formation of self-efficacy and success expectations when being confronted with academic tasks. These expectations may underly academic emotions and motivation, which may in turn influence effort, strategies of learning, and on-task behavior directly affecting learning and performance (Pekrun 1993). Research has shown that self-concepts implying moderate overestimation of own capabilities may be optimal for motivation and learning gains, whereas an underestimation may be rather detrimental. However, self-evaluations precisely reflecting reality may also be suboptimal for learning and achievement, indicating that there may be a conflict between educational goals of teaching students to be self-realistic vs. highly motivated (Helmke 1992).

4. Development Of Self-Concepts

4.1 Basic Needs Underlying Self-Concept Development

Two classes of basic human needs seem to govern the development of self-representations (cf. Epstein 1973). One are general needs for maximizing pleasure and minimizing pain, from which needs for self-enhancement follow, implying motivation to establish and maintain a positive view of the self. The second category comprises needs to perceive reality and foresee the future in such ways that adaptation and survival are possible, thus implying needs for self-perceptions which are consistent with reality, and consistent with each other (needs for consistency). Self-enhancement and consistency may converge when feedback about the self is positive. However, they may be in conflict when feedback is negative: The need for self-enhancement would suggest not accepting negative feedback, whereas needs for reality-oriented consistency would imply endorsing it.

4.2 Developmental Courses Across The Life Span

The development of self-concepts is characterized by the interplay of mechanisms driven by these conflicting basic needs. In general, in the absence of strong information about reality, humans tend to overestimate their capabilities in self-enhancing ways. At the same time, they search for self-related information, and tend to endorse this information even if it is negative on condition that such negative information is salient, consistent, and cumulative. For example, it has repeatedly been found that many children entering school drastically overestimate their competences to master academic demands, but adjust their self-evaluations downward when diverging cumulative feedback is given (Helmke 1992). The process of realistically interpreting self-related feedback may be impeded in young children by their lack of sufficiently sophisticated metacognitive competences. Nevertheless, the interplay of beginners’ optimism and experts (relative) realism may characterize not only the early elementary years, but later phases of entering new institutions as well (university, marriage, a new job etc.), thus implying a dynamic interplay of self-related optimism and realism across much of the life span.

4.3 Impact Of Educational Environments

The influence of social environments was addressed early in the twentieth century by symbolic interactionism postulating that interactions with significant others may shape the development of self-concepts (cf. the concept of the ‘looking glass self,’ Cooley 1902). Generally, a number of environmental variables may be influential. Self-concepts can develop according to direct attributions of traits and personal worth by other persons on condition that such attributions are consistent with other sources of information and interpreted accordingly. A second source of self-related information are indirect, implicit attributions which are conveyed by others’ emotional and instrumental behavior towards the developing person. Of specific importance are acceptance and support by others implying attributions of personal worth, thereby influencing the development of a person’s general self-esteem (Pekrun 1990).

Beyond attributions, social environments may define situational conditions for the development of knowledge, skills, motivation, and behavior, which may in turn contribute to self-perceptions of own competences. For example, instruction may build up knowledge and skills influencing knowledge-specific academic self-concepts; support of autonomy may foster the acquisition of self-regulatory abilities and, thereby, the development of related self-concepts of abilities; and consistent behavioral rules may enhance the cognitive predictability of students’ environments which may also positively affect their overall sense of competence.

As outlined above, concerning academic self-concepts, feedback of achievement implying information about abilities and effort may be of specific importance, and the effects of such feedback may depend on the reference norms used. Feedback given according to competitive, interindividually referenced norms implies that the self-concepts of high achievers may benefit, whereas low achievers may have difficulties protecting their self-esteem. In contrast, intraindividual, mastery-oriented and cooperative norms may be better suited to foster low achievers’ self-concepts as well. Finally, classroom conditions may also be influential. Specifically, grouping of students may influence students’ relative, within-classroom achievement positions, thus affecting their academic self-concepts when competitive standards of grading are used. For example, being in a low-ability class would help an average student to maintain positive academic self-concepts, whereas being in a class of highly gifted students would enforce downward adjustment of self-evaluation (‘big-fish-little-pond effect,’ Marsh 1987).

5. Summary Of Implications For Educational Practice

Fostering students’ self-concepts may be beneficial for their achievement and personality development. Furthermore, since positive self-esteem may be regarded as a key element of psychological health and wellbeing, nurturing self-esteem may be regarded as an educational goal which is valuable in itself. From the available research summarized above, it follows that education may contribute substantially to self-concept development. Concerning parents as well as academic institutions like the school, acceptance and support may be of primary importance for the development of general self-esteem. In addition, giving students specific feedback on achievement, traits, and abilities may help in shaping optimistic self-concepts and expectancies which nevertheless are grounded in reality. In designing feedback, it may be helpful to use mastery-oriented, individual, and cooperative reference norms instead of relying on social comparison standards and competitive grading. Furthermore, educational environments may help students’ self-concept development by providing high-quality instruction, consistent normative and behavioral structures implying predictability, as well as sufficient autonomy support fostering students’ sense of controllability and competence.

Bibliography:

1. Bracken B A (ed.) 1996 Handbook of Self-concept. Wiley, New York
2. Cooley C H 1902 Human Nature and the Social Order. Scribner, New York
3. Covington M V 1992 Making the Grade. Cambridge University Press, New York
4. Covington M V, Beery R 1976 Self-worth and School Learning. Holt, Rinehart and Winston, New York
5. Epstein S 1973 The self-concept revisited or a theory of a theory. American Psychologist 28: 404–16
6. Hansford B C, Hattie J A 1982 The relationship between self and achievement performance measures. Review of Educational Research 52: 123–42
7. Hattie J 1992 The Self-concept. Erlbaum, Hillsdale, NJ
8. Helmke A 1992 Selbst ertrauen und schulische Leistungen. Hogrefe, Gottingen, Germany
9. Helmke A, van Aken M A G 1995 The causal ordering of academic achievement and self-concept of ability during elementary school: A longitudinal study. Journal of Educational Psychology 87: 624–37
10. Higgins E T 1987 Self-discrepancy: A theory relating self and aff Psychological Review 94: 319–40
11. James W 1893 Psychology. Fawcett, New York
12. Kulik J A, Kulik C L 1992 Meta-analytic findings on grouping processes. Gifted Child Quarterly 36: 73–7
13. Marsh H W 1986 Verbal and math self-concepts: An internal external frame of reference model. American Educational Research Journal 23: 129–49
14. Marsh H W 1987 The big-fish-little-pond effect on academic self-concept. Journal of Educational Psychology 79: 280–95
15. Marsh H W 1988 Self Description Questionnaire I (SDQ): Manual and Research Monograph. Psychological Corporation, San Antonio, TX
16. Marsh H W, Shavelson R J 1985 Self-concept: Its multifaceted, hierarchical structure. Educational Psychologist 20: 107–25
17. Marsh H W, Yeung A S 1997 Causal effects of academic self-concept on academic achievement: Structural equation models of longitudinal data. Journal of Educational Psychology 89: 41–54
18. Pajares F 1996 Self-efficacy beliefs in academic settings. Review of Educational Research 66: 543–78
19. Pekrun R 1990 Social support, achievement evaluations, and self-concepts in adolescence. In: Oppenheimer L (ed.) The Self-concept. Springer, Berlin, pp. 107–19
20. Pekrun R 1993 Facets of adolescents’ academic motivation: A longitudinal expectancy-value approach. In: Maehr M, Pintrich P (eds.) Advances in Motivation and Achievement. JAI Press, Greenwich, CT, Vol. 8, pp. 139–89
21. Shavelson R J, Hubner J J, Stanton G C 1976 Self-concept: Validation of construct interpretations. Review of Educational Research 46: 407–41
22. Skaalvik E M, Rankin R J 1995 A test of the internal external frame of reference model at different levels of math and verbal self-perception. American Educational Research Journal 32: 161–84