View article

Time series data and time series designs have a long history in psychological research. Time series designs are distinguished from other designs by the properties of time series data and by the necessary reliance on a statistical model to control threats to validity. In the long run, a time series is a realization of a latent causal process. This chapter elaborates on the specific forms of the general model and on the set of rules for building models at a later point. For present purposes, the general model allows three design variations:(a) descriptive time series designs,(b) correlational time series designs, and (c) experimental or quasi-experimental time series designs. The chapter lists the threats to validity that are relevant to time series studies. Time series designs differ from other approaches in that common threats to validity are controlled by a statistical model.(PsycInfo Database Record (c) 2023 APA, all rights reserved)