Analysing the effects of individual characteristics and self-efficacy on users' preferences for system features in relevance judgment

Introduction. The design of information systems has traditionally focused on algorithmic relevance. While scholars have called for a user-centred perspective on relevance, less is known about what system features are important in supporting subjective relevance judgment of different individuals. Method. Drawing from Kuhlthau's information search process and Norman's task completion model, an integrated model conceptualising information seeking on two levels (process and sub-task) were developed. Two hundred and seventy-seven users were surveyed on system features they considered important at various stages of an information seeking task. Analysis. Descriptive statistics and exploratory factor analysis were used to group system features. The relationships between system feature preference and individual characteristics were tested using multiple regressions. Results. The top features were identified and factor analysis resulted in a seven-factor solution. Except for sex, the multiple regression analyses found statistically significant relationships between individual characteristics such as selfefficacy in information seeking and respondent's system feature preference. Conclusions. Information systems design will benefit from understanding users' varying information seeking selfefficacy and their needs for features supporting subjective relevance judgment. With its strong tradition of studying diverse human behaviour, the information behaviour field has much to offer a user-centred and context-aware approach to information system design.