Integrated group discussion (IGD) tasks represent a new type of integrated speaking task that requires students to communicate with their peers after comprehending and using provided source materials. With its potential to enhance effective communication and information literacy, IGD stands at the core of language education. Although the IGD task type has become increasingly popular, its construct is still unexplored. In this study, 124 Hong Kong undergraduate students completed an IGD task and a group discussion (GD) task (i.e., without source materials) in Putonghua. Four factors, speech and expression, content with source use, interaction, and language usage, were extracted to represent IGD performance. Two indicators of GD performance (linguistic expression and engagement in interaction) significantly predicted IGD performance, explaining 21.9 % of its total variance. The large proportion of unexplained IGD performance variance suggests that IGD tasks may be worth implementing. The theoretical and practical implications of the findings are discussed in details. Copyright © 2024 Walter de Gruyter GmbH, Berlin/Boston.