Evidence to Support Faculty Development in Japan: Development of the Self-Evaluation Scale of Teaching Behavior in Nursing Lectures

Purpose:

Lectures have been used in nursing education as an effective means to teach nursing knowledge and theory, and to connect these to practice. However, lectures tend to make students passive and limit their use of reflection. These limitations require faculty members to maintain and enhance the quality of their teaching activities in nursing lectures. Yet, no scales have been developed for faculty self-evaluation of teaching activities in this area. Using such a scale would enable the assessment of the quality of teaching activities and support faculty development. The purpose of this study was to develop and evaluate the Self-Evaluation Scale of Teaching Behavior in Nursing Lectures. 

Methods:

The study consisted of four phases: a) item generation based on the findings of qualitative and inductive research (Goto et al., 2010) that conceptualized faculty’s behavior in nursing lectures, b) assessment of content validity and refinement of the scale by a panel of experts and a pilot study, c) item analysis and selection through a survey, and d) evaluation of the scale’s validity and reliability.

Results:

A 5-point Likert scale was used, comprising 63 items classified into 9 subscales that were based on 9 concepts of teaching behavior in nursing lectures. In the first survey, the instrument packets were mailed to 1373 faculty members of 116 randomly sampled nursing schools in Japan. In total, 652 (47.5%) scales were returned, and 574 valid datasets were analyzed. From the data, 36 items were selected based on the results of an item analysis. A factor analysis of the 36 items extracted 9 factors that were reflective of the 9 subscales. The known-groups technique was used to examine construct validity. Faculty members with teaching experience of more than 20 years had higher scores on the scale than faculty members with teaching experience of 2 years or less (t = -4.27, p < .001). Cronbach’s alpha of the scale was 0.95 and Cronbach’s alpha of the ten subscales ranged from 0.71 to 0.83. In the second survey, the same instrument packets were mailed to 71 faculty members (using convenience sampling) to examine test-retest reliability. In all, 21 (29.6%) scales were returned, and 21 valid datasets were analyzed. The test-retest reliability was 0.80 (p < .001).  

Conclusion:

The results indicate that this scale possesses good construct validity, high internal consistency reliability, and high stability. It is useful for supporting faculty development.