Analyzing developmental processes on an individual level using nonstationary time series modeling.

Abstract

Individuals change over time, often in complex ways. Generally, studies of change over time have combined individuals into groups for analysis, which is inappropriate in most, if not all, studies of development. The authors explain how to identify appropriate levels of analysis (individual vs. group) and demonstrate how to estimate changes in developmental processes over time using a multivariate nonstationary time series model. They apply this model to describe the changing relationships between a biological son and father and a stepson and stepfather at the individual level. The authors also explain how to use an extended Kalman filter with iteration and smoothing estimator to capture how dynamics change over time. Finally, they suggest further applications of the multivariate nonstationary time series model and detail the next steps in the development of statistical models used to analyze individual-level data.

The Course of Subjective Sleep Quality in Middle and Old Adulthood and its Relation to Physical Health

“…According to the study by Molenaar, Sinclair, Rovine, Ram, and Corneal (2009), conclusions that are derived from analyses of interindividual variation may often not be generalized to processes that operate on an intraindividual level. Future studies on longitudinal development of subjective sleep quality might apply multilevel modeling approaches that separately analyze within-person development and between-person variation as for instance suggested by Hertzog and Nesselroade (2003).…”

Four Functional Roles for Case Studies in Emerging Adulthood Research

“…There are specialized and complex statistical methods for this process (e.g. Molenaar, Sinclair, Rovine, Ram, & Corneal, 2009), but given that within-case statistical analysis is still not widely used amongst academics and students, here we provide a few simple pointers for how a quantitative analysis package such as SPSS can be used to conduct within-case analysis. It differs in how the rows of the statistics spreadsheet are used.…”

Toward a Unified Framework for the Study of Between-Person and Within-Person Structures: Building a Bridge Between Two Research Paradigms

“…Although this is possible with current SEM software, for T ≥ 80 and with several thousand models to be estimated, SEM is no viable alternative for reasons outlined earlier. This is an active field of research and we are optimistic that future software and/or estimation procedures will overcome these limitations (e.g., see Molenaar, 1994;Molenaar, Sinclair, Rovine, Ram, & Corneal, 2009;Chow, Zu, Shifren, & Zhang, 2011, for work on time-varying parameters). For the time being and for the purpose of this article, however, a work-around is needed.…”