Factor Analysis of a Search Self-Efficacy Scale
Kathy Brennan, Diane E Kelly, Yinglong Zhang · 2016
Participants in information search studies are often asked to characterize their search expertise using questionnaires provided by researchers. These questionnaires often contain ad hoc sets of items in part because there are no valid and reliable measures of search experience. In this paper, we present results of an exploratory factor analysis of 327 responses to a 14-item Search Self-Efficacy scale (SSE) that we have been using as a way to measure search experience in our research. The responses come from eight different interactive information retrieval (IIR) studies, which we have conducted over the last six years, with a variety of participant types: university students, participants from the general adult population and crowd-sourced participants. The purpose of this analysis is to understand the variation in search self-efficacy scores across different types of people and to evaluate the potential of the SSE scale as a tool for measuring search experience. Overall, participants from all eight studies reported similar levels of search self-efficacy; the overall average from all eight studies was 7.47 (items scored on a 10-point scale) with little variance (standard deviation=1.36). Both the lowest and highest scores (7.1 and 7.8) were observed in studies involving the general adult population. A factor analysis showed that the questionnaire items load onto six factors, although only four had sufficient numbers of items loading on them. These four factors represent overall task success, effective use of time, query development skills, and advanced search skills.