Chapter 01

Multilevel Models for Intensive Longitudinal Data

Theodore A. Walls, Hyekyung Jung, and Joseph E. Schwartz

In the initial chapter in this volume, Walls, Jung and Schwartz undertake several ambitious goals intended to orient both applied statisticians and for methodologists working in many scientific fields to the content of this volume. As such, they first provide a synopsis of the emergence of intensive longitudinal data, particularly as it has progressed in behavioral science. They follow this summary with a brief exegesis of multilevel modeling approaches, with emphasis on the equivalence of models expressed in terms of "levels" with those expressed by a single linear combination. They also review model estimation options for the multilevel model and direct the reader to extensive resources for this kind of modeling. Finally, they develop an example which demonstrates the application of the multilevel model for intensive longitudinal data when calendar or clock time is not salient and, rather, when a theoretically specified time-graded association is of interest. This application utilizes diary data collected from Indian adolescents regarding their choice and control beliefs over many within-day and over days reports. They expand the model to demonstrate both the use of diagnostics for residuals and the incorporation of covariates.

In order to read some of these files, you must have Adobe Reader installed on your computer. You can download a free copy of this program from the Adobe website.

Website Terms and Conditions and Privacy Policy
Please send comments or suggestions about this Website to