Elements of performance and satisfaction as indicators of the usability of digital spatial interfaces for information-seeking: Implications for ISLA

This study explored the overall usability of two Internet-accessible digital library interfaces for information-seeking in order to provide guidance for the continued development of Project ISLA, a geospatial library under development at the University of Southern California. The interfaces studied were those of the Journal Storage Project, JSTOR, a subset of the University of Michigan's Digital Library Project, and the prototype interface of Project Alexandria, an effort of the University of California at Santa Barbara. Twenty-three university students participated in the study which employed convenience sampling, typical for identifying target users in the study of usability. Participants were given a pre-screening questionnaire and an abilities test to assess their individual differences in the areas of computer and Internet experience, gender, and spatial ability. Four subtests of spatial ability from The Kit of Factor-Referenced Cognitive Tests were given to assess participants' spatial aptitude. Descriptive in nature, the study employed usability validation testing to assess the ability of participants to complete predetermined benchmark task scenarios for the performance measures; task complexity increased with each subsequent scenario for each of the two interfaces. The Questionnaire for User Interaction Satisfaction{dollar}rm sp{lcub}TM{rcub}{dollar} was used to measure the satisfaction variables based upon predetermined benchmark measures. At the established benchmarks, several usability concerns were raised. The study found that those with higher spatial ability as an individual difference outperformed their low spatial ability counterparts on both interfaces; participants with low spatial ability required more time to complete each task scenario than those with high spatial ability. Likewise, those with high spatial ability rated both interfaces higher in satisfaction than their low ability counterparts. Findings are consistent with the findings of Chen & Rada's 1996 23-study meta-analysis, as well as their finding that task complexity shares an inverse relationship with task performance. Recommendations are made for ISLA's digital geospatial graphical user interface (GUI) which specifically address issues of differences in spatial ability, as well as temporal and other issues encountered in the use of Internet-based GUIs.