We analyzed how effort in searching is associated with search output and task outcome. In a field study, we examined how students' search effort for an assigned learning task was associated with precision and relative recall, and how this was associated to the quality of learning outcome. The study subjects were 41 medical students writing essays for a class in medicine. Searching in Medline was part of their assignment. The data comprised students' search logs in Medline, their assessment of the usefulness of references retrieved, a questionnaire concerning the search process, and evaluation scores of the essays given by the teachers. Pearson correlation was calculated for answering the research questions. Finally, a path model for predicting task outcome was built. We found that effort in the search process degraded precision but improved task outcome. There were two major mechanisms reducing precision while enhancing task outcome. Effort in expanding Medical Subject Heading (MeSH) terms within search sessions and effort in assessing and exploring documents in the result list between the sessions degraded precision, but led to better task outcome. Thus, human effort compensated bad retrieval results on the way to good task outcome. Findings suggest that traditional effectiveness measures in information retrieval should be complemented with evaluation measures for search process and outcome.

No abstract

No abstract

We describe information production and use in a client information system (CIS) by social workers in child protection. Data consists of interviews and observations with think aloud material. Information production and use are embedded with other work tasks and go throughout the whole client process. CIS does not support social workers' tasks sufficiently. Our findings imply recommendations for developing CIS for better fitting to work practices in child protection.

Abstract. Our study focuses on the information needed for overviews by social workers in child protection, and how information is presented in client information systems (CIS). Data consists of semi-structured interviews and social workers' observations while they used CIS in their daily work. The analysis was structured by the concept of temporal trajectory. We identified three major interconnected information strands: concern, child's private life and institutional actions. Their temporal lengths and aspects are disparate. CIS offers modest temporal overviews of clients' cases. Representing information strands as timelines on the interface would provide better overviews.