Date of Award:

5-2010

Document Type:

Dissertation

Degree Name:

Doctor of Philosophy (PhD)

Department:

Instructional Technology and Learning Sciences

Committee Chair(s)

Andrew Walker

Committee

Andrew Walker

Committee

Nick Eastmond

Committee

Brett E. Shelton

Committee

Brian Belland

Committee

Kerstin E. E. Schroder

Abstract

Online learning research is largely devoted to comparisons of the learning gains between face-to-face and distance students. While student learning is important, comparatively little is known about student satisfaction when engaged in online learning and what contributes to or promotes student satisfaction. Emerging research suggests there are a few strong predictors of student satisfaction, and other predictors that may or may not predict student satisfaction. None of the existing research examines predictors together, or statistically controls for course differences. This study examines the influence of various factors on student satisfaction including three types of interaction, Internet self-efficacy, and self-regulated learning.

Participants (N = 180) include both undergraduate and graduate students attending exclusively online classes in education. Students responded to an online survey adapted from several different scales. A pilot test of the survey and procedures showed strong validity and reliability for the sample. To control for course differences, data analysis focused on a hierarchical linear model (HLM) with student and class level variables. Results indicate learner-instructor interaction and learner-content interaction are significant predictors of student satisfaction when class-level variables are excluded. Of the class-level predictors, only the program from which the course was offered moderates the effect of learner-content interaction on student satisfaction. There is no direct impact of class-level predictors on student satisfaction. Learner-content interaction is the sole significant predictor when class-level predictors are added to the model. Supporting analyses for the HLM, results, limitations, and significance of the findings are reported and discussed.

Checksum

ba2e70020deb42d064b5286f8a8d89af

Comments

This work made publicly available electronically on August 30, 2010.

Share

COinS