Development and validation of the teacher writing to learn scale

Abstract

Despite numerous efforts by educators and U.S. government agencies to improve the public education system, students continue to struggle with writing, mathematics, science and reading. Researchers and educators have employed a wide range of interventions, but proficiencies are still not at desired levels. One intervention that lacks empirical research is writing to learn (WTL). Social constructivist learning theory and cognitive learning theory of information processing provide an explanation as to why WTL promises to be an effective tool for improving content knowledge and writing skills. Further, the theoretical literature on WTL and the research on general writing mirror such theories of learning. However, despite over thirty years of theoretical and inductive research, little research examines the generalizability of WTL's effectiveness on writing and other content areas. Before measuring the effects of WTL on students, it is necessary to address teacher knowledge and efficacy of WTL. Therefore, the purpose of this proposed study is to develop an instrument to measure teacher knowledge and efficacy of WTL in the content areas of mathematics, science, social studies and language arts (which includes reading). Using the theories of self-efficacy (Bandura, 1977) and using the literature on effective teaching of writing, WTL and theories of learning. This study began with item development using the literature and teacher input. Next, experts were used to test content validity and appropriate item response. The result was a six factor model to be tested empirically. Internal consistency measures using alpha and omega, exploratory factor analysis and confirmatory factor analysis were used to check the response processes of the measure. The scale was correlated with other measures and differences tests were used to examine attributes of respondents. Results indicated problems with the first, second and last factors. The remaining two factors, perceived relevance of writing to the content and efficacy of teaching with writing showed the best fit indices, though future research is needed to refine them. The final two factors negatively correlated with writing apprehension, positively with teacher efficacy (with little explained variance) and positively correlated with number of years teaching. Difference tests indicate a strong difference between content areas of teachers on both factors and a small difference in efficacy to teach writing given gender. No differences were found between urban, rural and suburban teachers and none were found between middle school and high school teachers. This research adds to the body of work by developing a measure of teacher readiness to use WTL. However, future research is needed to refine the instrument to a usable state so that intervention research and staff development can use it.