Researchers often collect longitudinal data so as to model change over time in a phenomenon and for a population of interest. Inevitably, there will be some variation across individuals in specific time intervals between assessments. By necessity or by choice, a researcher can decide to ignore these individual differences in times of assessments. In this simulation study of growth curve modeling, I investigate how ignoring individual differences in time points when modeling change over time relates to convergence and admissibility of solutions, bias in estimates of parameters, power to detect change over time, and, when there is no change over time, Type I error rate. The simulation factors that I manipulate in this study are magnitude of the individual differences in assessment times that are ignored, magnitude of change over time, number of time points, and sample size. Results show that, in contrast to the correct analysis, ignoring individual differences in time points frequently led to inadmissible solutions, especially with few time points and small samples, regardless of the specific magnitude of individual differences that were ignored. Mean intercept and slope were generally estimated without bias. With few time points and small samples, ignoring individual differences in time points yielded overestimated intercept and slope variances and underestimated intercept-slope covariance and residual variance, more so than when using the correct analysis. When there were more than 3 time points, or when there were 3 time points and sample size was 500, ignoring individual differences in time points yielded overestimated residual variance, but only if individual differences were large. Power and Type I error rate for the linear slope were unaffected by the type of analysis. Overall, this study suggests that it is advisable to account for individual differences in time points whenever possible.
Level of Degree
First Committee Member (Chair)
Second Committee Member
Third Committee Member
growth curve modeling, structural equation modeling, multilevel modeling, longitudinal, simulation, misspecification
Coulombe, Patrick. "Ignoring individual differences in times of assessment in growth curve modeling." (2014). https://digitalrepository.unm.edu/psy_etds/29