Recent developments in theories and data collection methods have made intensive longitudinal data (ILD) increasingly relevant and available for organizational research. New methods for analyzing ILD have emerged under the multilevel modeling framework. In this article, we first delineate features of ILD (including autoregressive relationships, trends, cycles/seasons, and between-subject variability in temporal trends). We discuss the analytic challenges for handling ILD using traditional analytic tools familiar to organizational researchers (e.g., growth models, single-subject time series analyses). We then introduce a statistical approach for handling ILD from the multilevel modeling framework: dynamic structural equation modeling (DSEM). We provide three examples using simulated data sets to demonstrate how to apply DSEM to examine ILD with a software program familiar to organizational researchers (i.e., Mplus). Finally, we discuss issues related to applying DSEM, including centering, missing data, and sample size.
Bibliographical noteFunding Information:
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Le Zhou?s work on this research was supported by the Lawrence Fellowship from the Carlson School of Management, University of Minnesota and the National Science Foundation (Grant No. 1533151). Mo Wang?s work on this research was supported in part by the Lanzillotti-McKethan Eminent Scholar Endowment.
© The Author(s) 2019.
- dynamic structural equation modeling
- experience sampling method
- intensive longitudinal data
- multilevel modeling
- time series