Exploring the Relationship between Performance and Response Process Data in Digital Literacy Assessment

Journal of Modern Foreign Psychology. 2024. Vol. 13. No. 1. P. 58–68.

Tkachenko Irina, Tarasova K., Gracheva D.

Measuring complex latent constructs is challenging because of their multi-dimensionality. In this context, computer-based assessments have gained popularity due to its ability to handle large diverse data. The aim of the study is to investigate the interrelationship between performance, time, and actions in computer-based digital literacy assessment. The study involved more than 400 8th-grade schoolchildren (approximately 14—15 years old) from secondary schools in Russia. A subset was obtained from indicators capturing the demonstration of analysis of data, information, and digital content, which is a component of the information literacy in the digital literacy framework. The data was used to create latent models in the structural equation modeling framework. Confirmatory one-factor model for the Performance factor showed a good fit to the data (CFI=1; TLI=1; RMSEA=0). The model with dependencies among indicators demonstrated improved model fit (χ2(18)=510,65; p=0,05) compared to the model without such dependencies. The results suggest that performance, time, and actions are interdependent. The findings underscore the need for a comprehensive approach to assessing digital literacy that accounts for these interdependencies, as well as investigating behavioral patterns of interaction with a large amount of information in the digital environment.
 

Keywords: confirmatory factor analysis, structural equation modeling, response time, evidence-centered design, digital literacy, computer-based assessment, Process data, clicks