|
|
|
| Longitudinal Item Response Times Models for Assessing Change in Latent Processing Speed |
| CHEN qi-peng1 ZHAN pei-da1,2,3 |
| 1.School of Psychology, Zhejiang Normal University, Jinhua 321004, China; 2. Intelligent Laboratory of Child and Adolescent Mental Health and Crisis Intervention of Zhejiang Province, Jinhua 321004, China; 3. Key Laboratory of Intelligent Education Technology and Application of Zhejiang Province 321004, China |
|
|
|
|
Abstract In psychological and behavioral studies, measuring change over time is essential to a developmental study. These changes can be captured by longitudinal latent variable models. With the spread of computerized assessment, learning, and experimental systems, it has become common to collect process data, such as reaction or response times (RTs), in addition to traditional item response accuracy (RA) data. RT data is used as a complement to RA data, describes the total time taken by individuals to solve problems, and can be used to analyze the latent processing speed of individuals. However, existing longitudinal models mainly focus on longitudinal RA data and pay little attention to longitudinal RT data. While most of the existing RT models are limited to analyzing cross-sectional RT data and cannot track the development of students' latent processing speed over time. To this end, in this study, four longitudinal RT models based on two commonly used longitudinal modeling methods (i.e., multivariate normal distribution modeling and latent growth curve modeling) were proposed to achieve objective assessing of individual latent processing speed change. The measurement models are consistent across the four models, with differences mainly in the structural model, which describes how the latent processing speed changes over time. Overall, the proposed four modeling approaches, which are extensions of the lognormal RT model (van der Linden, 2006) in the longitudinal setting from the perspective of structural equation modeling, can provide the development trajectories of latent processing speed. The results of the empirical study and simulation study mainly indicated that the four longitudinal RT models proposed in this paper have practical applicability and good psychometric properties, which not only enrich the analysis methods of longitudinal RT data in psychological and educational measurement but also expand the application scope of longitudinal latent variable models.
|
|
Published: 14 March 2023
|
|
|
|
|
|
|
