How do Information Source Selection Strategies Influence Users' Learning Outcomes'

Learning-related type of tasks has attracted much research attention recently but it is still not clear what factors would influence users learning outcomes and how. In this study, we conducted a user experiment to assess searchers learning outcomes and examine how information source selection strategies would influence their learning outcomes. In this experiment, thirty-two college students conducted search for two types of learning tasks: receptive tasks and critical tasks. Participants were asked to write down what they knew about the task before and after the search. For data analysis, we proposed a comprehensive assessment method, which used both quantitative measures (i.e. knowledge points, knowledge facets, knowledge scope, etc.) and qualitative measures to assess users' learning outcomes. Our results demonstrated that searchers' information source preferences influence their learning outcomes; i.e., encyclopedia-preferred sessions had better relevance of written summaries in receptive tasks and Q&A preferred sessions led to better relevance in critical tasks. Furthermore, searchers had two types of information source selection strategies: task-adaptive strategy and non-task-adaptive strategy. The results showed that searchers with task-adaptive strategy could gain better learning outcomes, e.g. knowledge points, facets, scope, depth, relevance and analyticity. This study highlighted the importance of information source selection strategies in learning-related type of tasks, and knowing how to select suitable information sources for different types of tasks may benefit the learning outcome for searchers.