Development and psychometric exploration of the online teaching self-efficacy scale

ABSTRACT With over 3.1 million students enrolling in at least one online higher education course in the fall semester of 2005, online education is emerging as one of the most prominent and pervasive forms of distance education (Allen & Seaman, 2006). The resulting implication for postsecondary institutions is that the necessary instructors are needed to meet the growth of online education. The challenge presented with this need revolves around the consideration of the integral pedagogical foundations involved in online instruction. The literature presents a potential link between the ambiguity of online pedagogy and instructors� self-beliefs to teach online (Easton, 2003). However, a measure to assess the specific dimensions of online teachers� self-efficacy beliefs is not available. Additionally, current teacher self-efficacy scales do not encompass the domain specific tasks of online instruction. The study had two primary purposes; the first was to develop the Online Teaching Self-Efficacy Inventory, a series of five scales to examine the teaching self-efficacy beliefs of online instructors. The second portion of the study explored the psychometric properties of the inventory scales. The key research questions were these: (1) How many factors are needed to summarize the pattern of correlations in the correlation matrices for the inventory scales? (2) Are the inventory scales reliable? (3) Within the inventory scales, how much variance does each factor account for?
(4) How can the factors be interpreted from the derived variables of the Online Teaching Self-Efficacy Inventory scales? An exploratory factor analysis for each of the five online teaching self-efficacy inventory scales was conducted to ascertain item composition using factor extraction and factor rotation analyses. The analyses resulted in identifying items for each scale whose removal enhanced the instrument�s factor structure. The resulting inventory has 47 items allocated to five scales:(1) Web-Based Course Structure;(2) The Online Alignment of Objectives, Instruction, and Assessment;(3) Course Content Migration; (4) Virtual Interaction; and (5) Selection of Technological Resources. Internal consistency and reliability analyses were conducted to determine the extent to which items within each scale measured the same construct as other items within that scale. All of the retained items corresponding with each of the five inventory scales had factor loadings of at least 0.32. The alpha reliability coefficient for each scale ranged from 0.84 to 0.95. The development of the Online Teaching Self-Efficacy Inventory relied extensively on literature pertaining to online instruction and expert content validation techniques. It treats online instruction and associated instructional self-efficacy beliefs as distinctive entities encompassed by task-specific competencies and dimensions. The results of the study provide an initial measure for examining the self-efficacy beliefs of online instructors. Implications of the study and directions for future research are presented.

This dissertation won 1st Place in the Texas Tech University Outstanding Thesis and Dissertation Award, Social Sciences, 2010.