ABSTRACT
Change is an important question in psychology (Boker & Wegner, 2007; Collins & Sayer, 2001; Walls & Schafer, 2005). Studying change requires measuring the same individual at multiple times. These measurements often form a time series. Because perfect measures of many psychological constructs such as moods and personality states do not exist, time series of psychological variables are usually contaminated by measurement error. Dynamic factor analysis (DFA) (Browne & Nesselroade, 2005; Browne & Zhang, 2007; Engle & Watson, 1981; Geweke & Singleton, 1981; Molenaar, 1985; Nesselroade, McArdle, Aggen, & Meyers, 2002) is a combination of factor analysis and time-series analysis and is an ideal tool for modeling multivariate time series with measurement error.
