|
Article Excerpt
The psychometric properties of a scale assessing the writing self-efficacy of 1,258 students from Grades 4 to 11 were analyzed with exploratory factor analysis. Two factors emerged, 1 designating basic grammar skills and 1 designating more advanced composition skills. The Writing Self-Efficacy Scale (F. Pajares & G. Valiante, 1999) functioned equally well at each academic level assessed.
**********
Researchers in the area of writing have traditionally focused either on the skills that writers bring to this craft or on the instructional practices teachers use to improve their students' composition skills (e.g., Faigley, 1990; Hairston, 1990). Primarily, these efforts have been aimed at understanding the cognitive processes underlying students' compositions (see Hayes, 2006). More recently, however, researchers have pointed out that the relationship between students' cognitive skills and the manner in which they engage text is mediated by the interpretations students make about the skills they possess (Hidi & Boscolo, 2006; Pajares & Valiante, 2006). To account for this phenomenon, investigators have explored the self-beliefs that underlie student motivation in writing. Findings from this avenue of inquiry have led researchers to suggest that students' beliefs about their own writing competence, or self-efficacy beliefs, are instrumental to their ultimate success as writers (see Pajares, 2003). As a consequence, a number of researchers have explored the relationship between students' self-efficacy beliefs about writing, other motivation variables related to writing, and various writing outcomes.
Research findings have consistently shown that writing self-efficacy beliefs and writing performances are related. Most early self-efficacy studies were conducted on college students, and effect sizes for the influence of writing self-efficacy on writing performance ranged from beta = .32 to beta = .42 (e.g., McCarthy, Meier, & Rinderer, 1985; Meier, McCarthy, & Schmeck, 1984). Another consistent finding was that neither writing apprehension nor other motivation variables were typically predictive of writing performance in regression models that included self-efficacy. More recent findings support these results (Bruning & Horn, 2000; Pajares, 2003; Pajares & Valiante, 1997, 1999, 2001, 2006; Rankin, Bruning, & Timme, 1994; Wachholz & Etheridge, 1996; Zimmerman & Bandura, 1994). In general, results reveal that writing self-efficacy makes an independent contribution to the prediction of writing outcomes and mediates between previous and subsequent achievement in writing. This is the case even when powerful covariates such as preassessed writing ability or previous writing performance are included in statistical models. Standardized regression coefficients and direct effects between writing self-efficacy and writing outcomes in multiple regression and path analyses that control for such preperformance assessments have ranged from .19 to .40. Correlations between writing self-efficacy and writing performances typically range from .30 to .50, often depending on the age and competence of the students.
Writing self-efficacy is also associated with motivation constructs that are themselves related to writing performances. These concepts have been prominent in self-efficacy studies and have included writing self-concept (see Crain, 1996; Marsh, 1989), writing apprehension (Pajares & Valiante, 1997, 1999, 2001), self-efficacy for self-regulation (Harris & Graham, 1992; Zimmerman & Bandura, 1994), students' perceived value of writing (Shell, Colvin, & Bruning, 1995; Shell, Murphy, & Bruning, 1989), and achievement goal orientations (Pajares, Britner, & Valiante, 2000).
The relationship between gender and self-efficacy has been a focus of research in the area of writing, and findings have not been consistent. Some researchers have found no gender differences in writing self-efficacy across academic grades (Shell et al., 1995). Others have found that girls report stronger confidence in their writing capabilities than do boys, at least through middle school (Pajares, Miller, & Johnson, 1999; Pajares & Valiante, 2001). These differences may begin at early ages (Crain, 1996). Researchers have observed that girls experience a drop in their academic motivation and perceptions of competence as they reach high school (Bruning & Horn, 2000), perhaps because they begin to encounter classroom structures that emphasize a masculine form of discourse (Cleary, 1996).
Some findings have shed light on the development of writing self-efficacy beliefs as a function of schooling. Shell et al. (1995) assessed the writing self-efficacy of students in Grades 4, 7, and 10 and found no differences in their confidence that they possessed various grammar, usage, and composition skills. Again, because older students are in better possession of those skills, one wonders why confidence in skills does not increase proportionately. Other researchers have reported that students in the 1st year of middle school have stronger confidence in their writing skills than do students in subsequent middle school grades even though older students have greater writing competence (Pajares & Valiante, 1999). This pattern of decreasing confidence in language arts skills is consistent with findings from expectancy-value researchers who have reported that students' self-concepts of ability in English decrease from the start of Grade 6 to the end of Grade 7 (Wigfield, Eccles, MacIver, Reuman, & Midgley, 1991).
ASSESSING WRITING SELF-EFFICACY
Bandura (2006) has provided clear guidelines regarding how self-efficacy beliefs should be operationalized and measured. Because efficacy beliefs vary in level, strength, and generality, these dimensions are important in determining how instruments should be constructed. Imagine that a researcher is interested in assessing the essay-writing self-efficacy of middle school students and wishes to compare that with students' capability to write an essay. In any given domain, there are different levels of task demands that researchers may tap. In writing an essay, these can range from the lower level of writing a simple sentence with proper punctuation to the higher level of organizing sentences into a paragraph so as to clearly express an idea. Students are asked to rate the strength of their belief in their capability to perform each of the levels identified. Hence, a self-efficacy scale should provide multiple items of varying difficulty that collectively assess the domain of essay writing. In addition, items should be prototypical of essay writing rather than minutely specific features of writing (e.g., confidence to form letters), and they should be worded in terms of can, a judgment of capability, rather than...
|
