Effects of position and number of relevant documents retrieved on users' evaluations of system performance

  • Authors:

  • Diane Kelly;Xin Fu;Chirag Shah

  • Affiliations:

  • University of North Carolina at Chapel Hill, Chapel Hill, NC;University of North Carolina at Chapel Hill, Chapel Hill, NC;University of North Carolina at Chapel Hill, Chapel Hill, NC

  • Venue:
  • ACM Transactions on Information Systems (TOIS)
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Information retrieval research has demonstrated that system performance does not always correlate positively with user performance, and that users often assign positive evaluation scores to search systems even when they are unable to complete tasks successfully. This research investigated the relationship between objective measures of system performance and users' perceptions of that performance. In this study, subjects evaluated the performance of four search systems whose search results were manipulated systematically to produce different orderings and numbers of relevant documents. Three laboratory studies were conducted with a total of eighty-one subjects. The first two studies investigated the effect of the order of five relevant and five nonrelevant documents in a search results list containing ten results on subjects' evaluations. The third study investigated the effect of varying the number of relevant documents in a search results list containing ten results on subjects' evaluations. Results demonstrate linear relationships between subjects' evaluations and the position of relevant documents in a search results list and the total number of relevant documents retrieved. Of the two, number of relevant documents retrieved was a stronger predictor of subjects' evaluation ratings and resulted in subjects using a greater range of evaluation scores.