Learners’ Self-Perception and Motivation

Generalizing from previous studies, it is found that research on an individuals’

perception of his own competence (e.g. Harter, 1998) is very similar to that on expectancy-value models (Pintrich & Schunk, 2002). To be more specific, the construct of self-perception is analogous to the definition of task-specific self-concept proposed by Eccles and Wigfield.

From educational psychologists’ perspectives, students’ positive attitude toward learning and positive self-perception of their competence has significant impact on their motivation and the enhancement of their academic achievement (e.g., Harter, 1981; Bandura, 1994). The concept of self-perception was originated from Bem’s

‘Self-perception theory’ (Bem, 1972). Bem proposed that an individual comes to know their internal states partially by inferring the inner states from observations of their own behavior and the outcomes resulted from the behavior.

Harter (1992) presented a model that reflects factors influencing motivational orientation. The model suggests that the initial mastery urge starts the learning process, followed by the learning behaviors that coped with the urge. These behaviors result in certain degree of success or failure, along with learners’ perceptions of their control over the outcomes. The learning behaviors also bring in feedback from the significant others. All of consequences are postulated to have a direct impact on learners’

40

perceived competence, which sequentially provoke affective reactions, and eventually, the chain of the factors forms into the learners’ motivational orientation.

In general, previous research work on self-perception has shown that learners with positive self-perceptions of their academic competence and positive expectancies will be more likely to perform better and be more involved in academic tasks. Due to the importance of self-perception in learning motivation, studies exploring the developmental differences among various gender-groups, age-groups, and ethnic- groups in the construct have been initiated. The current study, which aims to mend the gap between such studies and their exploration in the research filed of SLA, is done in a single EFL context (i.e. Taiwan) where most students are from a uni-ethnic background; therefore, the variation of ethnic groups is eliminated, leaving the relationship between gender, age (grade-level) and self-perception for inspection.

2.4.3.1 The nature of gender in self-perception

Previous studies have shown that age and gender play a significant role in learning motivation and engagement. Pintrich and Schunk (2002) observed that, in most studies where gender difference was found for perception of competence, female learners had lower self-perception than males (Wigfield et al, 1996). However, in actual achievement outcomes, gender difference was either not found, or female learners usually performed better than male learners. Many researchers explained the phenomena by claiming that males tend to be more self-congratulatory and females tend to be more modest (Eccles, Adler, & Meece, 1984; Wigfield et al., 1996).

Several cross-sectional and longitudinal studies on gender differences in the academic performance have reached a general gender stereotype in that, higher math self-concepts and values exist in males, while higher language self-concepts and values are held in females (e.g., Jacobs et al., 2002; Marsh et al., 2006; Trautwein,

41

2003; Watt, 2004; Wigfield, Eccles, & Pintrich, 1996). Eccles, Wigfield and their colleagues attributed such stereotype to the fact that gender differences exist because of the culture or social stereotyping of the academic activities (Eccles et al., 1998).

Therefore, male learners are found to have higher self-perceptions on their performance on math and sports, two generally stereotyped as more masculine activities; while female learners tend to have higher self-perceptions of their competence in reading and English, two commonly stereotyped as more feminine activities (Eccles, 1983; Eccles & Harold, 1991; Eccles et al., 1989; Eccles et al., 1998; Wigfield et al., 1991). However, such gender stereotype has been largely challenged (e.g., Marsh, 1993) and further inspection on the nature of the gender differences is needed.

2.4.3.2. The nature of grade-level in self-perception

It has been claimed that learners’ competence perceptions and values generally decline through school years (Jacobs et al., 2002). Research has revealed that a decline in learner motivation and engagement is generally found to emerge after middle school. Jacobs et al. (2002) carried out a longitudinal study that explored the effects of grade-level in learning motivation, and discovered that learners’

self-perceptions of competence and subjective task values declined as they grew older.

(Jacobs, Lanza, Osgood, Eccles, & Wigfield, 2002). Martin (2001, 2003e, 2004) has also confirmed such declines in learning motivation and engagement in the middle school years. However, most of these studies were cross-sectional in nature instead of longitudinal.

A longitudinal cohort-sequential study examining the development of adolescents’ self-perceptions, values, and task perceptions was conducted by Watt (2004) in NSW Australia with students from Grades 7 through 11. She adopted

42

questionnaires assessing students’ self-perceptions (talent and success expectances), values (intrinsic and utility values), and task perceptions (difficulty and effort required) in relation to both math and English. The results displayed that self-perceptions and values declined through adolescence, which is in consistence with former research.

Pintrich and Schunk (2002) proposed three general explanations for the drop of learners’ self- perception across grade levels: First, the adoption of the Likert-scale in the self-rating questionnaire can pose as an estimation problem for younger learners may be more likely to use the endpoints of the Likert scale while older learners tend to respond across the continuum of the scale. Second, as learners grow older, their

‘internal cognitive psychological mechanisms’ also develops; younger learners may have higher self-perceptions while older learners may be more modest in their perception of competence. Third, the change of the learning environment around the learners, such as class structure, evaluation criteria, teachers’ demand, can result in the drop of self-perception.