Paper Info
Title
How Does Team Composition Affect Knowledge Gain of Users in Collaborative Web Search?
Abstract
Studies in searching as learning (SAL) have revealed that user knowledge gain not only manifests over a long-term learning period, but also occurs in single short-term web search sessions. Though prior works have shown that the knowledge gain of collaborators can be influenced by user demographics and searching strategies in long-term collaborative learning, little is known about the effect of these factors on user knowledge gain in short-term collaborative web search. In this paper, we present a study addressing the knowledge gain of user pairs in single collaborative web search sessions. Using crowdsourcing we recruited 454 unique users (227 random pairs), who then collaboratively worked on informational search tasks spanning 10 different topics and information needs. We investigated how users' demographics and traits, and the interaction between these factors could influence their knowledge gain. We found that in contrast to offline collaboration cases, user demographics such as gender, age, etc. do not significantly effect users' knowledge gain in collaborative web search sessions. Instead, our results highlight the presence of labor division of queries and particular interaction patterns in communication that facilitate knowledge gain in user pairs. Based on these findings, we propose a multiple linear regression model to predict the knowledge gain of users in collaborative web search sessions from the perspective of team composition.
Year
DOI
Venue
2020
10.1145/3372923.3404784
HT '20: 31st ACM Conference on Hypertext and Social Media Virtual Event USA July, 2020
DocType
ISBN
Citations 
Conference
978-1-4503-7098-1
0
PageRank 
References 
Authors
0.34
0
3
Name
Order
Citations
PageRank
Luyan Xu112.38
Xuan Zhou230622.57
Ujwal Gadiraju3282.51