Combining cognitive theory and data driven approaches to examine students’ search behaviors in simulated digital environments

Large-scale Assessments in Education

An IERI – International Educational Research Institute Journal

Table 5 Spearman Correlations (ρ) of Inquiry Task Performance Variables with Students’ Proportion of Search Sessions Classified into Each Cluster

	Proportion sessions in cluster 1 (n = 99)	Proportion sessions in cluster 2 (n = 162)	Proportion sessions in cluster 3 (n = 48)	Proportion sessions in cluster 4 (n = 10)
Total task score
Inquiry task total score (max: 100)	0.310	− 0.325	0.035	0.019
Task phase-level subscores
Task phase: setup (max: 12)	0.228	− 0.251	0.066	− 0.017
Task phase: free roam (max: 51)	0.209	− 0.260	0.044	0.098
Task phase: conclusion (max: 37)	0.203	− 0.259	0.117	− 0.005
Construct subscores
Subconstruct: planning (max: 6)	0.100	− 0.240	0.207	0.083
Subconstruct: locating (max: 22)	0.150	− 0.197	0.084	0.015
Locating: Questioning (max: 7)	− 0.053	− 0.089	0.202	0.083
Locating: Searching (max: 4)	0.407	− 0.254	− 0.226	− 0.041
Locating: Choosing Sources (max: 5)	0.095	− 0.112	0.070	− 0.053
Locating: Saving Sources (max: 6)	0.117	− 0.176	0.112	0.001
Subconstruct: Evaluating (max: 35)	0.260	− 0.275	− 0.019	0.099
Evaluating: Importance (max: 7)	0.215	− 0.257	0.029	0.095
Evaluating: Usefulness (max: 14)	0.247	− 0.236	− 0.038	0.052
Evaluating: Trustworthiness (max: 14)	0.225	− 0.245	− 0.025	0.118
Subconstruct: Synthesis (max: 37)	0.203	− 0.259	0.117	− 0.005

Values exceeding |.20| appear in boldface. Subscores for the Free Roam phase and the Locating and Evaluating subconstructs reflect students’ performance on the simulated web search tool, in addition to a simulated library search tool and simulated conversations with virtual characters (Coiro et al., 2018, 2019)