22 Time Series Analysis for Psychological Research

Research Methods in Psychology


  1. Wayne F. Velicer PhD1,
  2. Peter C. Molenaar PhD2

Published Online: 26 SEP 2012

DOI: 10.1002/9781118133880.hop202022

Handbook of Psychology, Second Edition

Handbook of Psychology, Second Edition

How to Cite

Velicer, W. F. and Molenaar, P. C. 2012. Time Series Analysis for Psychological Research. Handbook of Psychology, Second Edition. 2:IV:22.

Author Information

  1. 1

    University of Rhode Island, Cancer Prevention Research Center, Kingston, Rhode Island, USA

  2. 2

    The Pennsylvania State University, University Park, Pennsylvania, USA

Publication History

  1. Published Online: 26 SEP 2012


Time series analysis is a statistical methodology appropriate for longitudinal research designs that involve single subjects that are measured repeatedly at regular intervals over a large number of observations. The focus is on within-person variability rather than between-person variability. A time series analysis can investigate the underlying naturalistic process, the pattern of change over time, or evaluate the effects of either a planned or unplanned intervention. Advances in information systems technology are making make time series designs an increasingly feasible method for studying important psychological phenomena. The unique characteristic of all time series designs is the presence of serial dependency in the data. Beyond describing the basic model and applications, this chapter examines issues such as generalization, cyclic or seasonal data, the problem of missing data, and new methods of measurement. Traditional univariate time series analysis is extended to procedures that allow for multivariate time series analysis, including the role of covariates, modeling within a structural equation modeling format, and the patterns of intra-individual differences across time within a dynamic factor analysis model. The methods described in this chapter represent a unique series of methods that can address a new set of research questions.


  • time series analysis;
  • dynamic factor analysis;
  • idiographic methods;
  • serial dependency