Education and Human Sciences, College of (CEHS)

 

Date of this Version

November 2007

Comments

A Dissertation Presented to the Faculty of the Graduate College at the University of Nebraska in Partial Fulfillment of Requirements for the Degree of Doctoral of Philosophy; Major: Educational Studies (Educational Leadership in Higher Education). Under the Supervision of Professor Larry L. Dlugosh Lincoln, Nebraska; October, 2007.
Copyright © 2007 Ann L. McCann.

Abstract

This study explored the extent of and reasons for the adoption of an innovation (REAL) on a university campus (MSU). REAL was an assessment system to improve teaching, student learning and accountability. A mixed methods approach was used, with an electronic survey of faculty (n = 568) and telephone interviews with adopters (n = 9). Using a stratified random sample from all eight colleges, the survey measured perceptions of REAL related to: relative advantage, trialability, result demonstrability, visibility, compatibility, complexity, image and voluntariness. The interview questions also explored the strategies of senior administrative support, collaborative leadership, flexible vision, staff development, and visible actions.

The survey yielded a 41% response rate. All eight scales were reliable with alpha levels > 0.839. All eight scales emerged distinctly in a factor analysis, demonstrating construct validity. Results indicated that REAL was not well known to the faculty. Sixty percent reported hearing of REAL, and only 35% had ever viewed REAL reports. The most common response for how often they had used REAL was “never” (63%). The majority (85%) indicated that REAL had not helped their department improve teaching or learning. The respondents also reported REAL was lacking in the attributes that support adoption, generally disagreeing with the survey items. They did “somewhat agree” (mean = 4.98) that they were not required to use REAL. Perceptions of REAL improved with exposure. Frequent users were more likely to agree with four of the scales (correlations, p < .001). Logistic regression demonstrated that voluntariness was the only scale that consistently predicted REAL use (p < .001). Also, serving on an assessment committee and shown how to use REAL together predicted using REAL (p < .001).

In the interviews, adopters revealed real and potential benefits of REAL. They also revealed REAL was somewhat difficult to use, not compatible with campus culture and not viewed as a means for improvement. Although REAL benefited the campus, it did not possess attributes favorable to adoption. Guidelines were developed for improving adoption of assessment systems.
Adviser: Larry L. Dlugosh

Included in

Education Commons

Share

COinS