Addressing Autocorrelation in Time Series Data: A Comparison of Four Analytic MethodsUsing Data from College Course Evaluations

Abstract

The sequential nature of observations in time series make them inherently prone to autocorrelation. This can be problematic because autocorrelation violates a major assumption associated with many conventional statistical methods. Although numerous analytic techniques address autocorrelation, the literature is generally devoid of discussions that contrast the benefits and disadvantages of various methods. This paper provides readers with a brief introduction to autocorrelation and related concepts, and uses empirical data from college course evaluations to contrast the results of four commonly used methods for adjusting autocorrelation in social science research. Implications of results and recommendations for choosing between these strategies are discussed.