Statistical Issues in the Study of Temporal Data: Daily Experiences

Abstract

This article reviews statistical issues that arise in temporal data, particularly with respect to daily experience data. Issues related to nonindependence of observations, the nature of data structures, and claims of causality are considered. Through the analysis of data from a single subject, we illustrate concomitant time-series analysis, a general method of examining relationships between two or more series having 50 or more observations. We also discuss detection of and remedies for the problems of trend, cycles, and serial dependency that frequently plague temporal data, and present methods of combining the results of concomitant time series across subjects. Issues that arise in pooling cross-sectional and time-series data and statistical models for addressing these issues are considered for the case in which there are appreciably fewer than 50 observations and a moderate number of subjects. We discuss the possibility of using structural equation modeling to analyze data structures in which there are a large number (e.g., 200) of subjects, but relatively few time points, emphasizing the different causal status of synchronous and lagged effects and the types of models that can be specified for longitudinal data structures. Our conclusion highlights some of the issues raised by temporal data for statistical models, notably the important roles of substantive theory, the question being addressed, the properties of the data, and the assumptions underlying each technique in determining the optimal approach to statistical analysis.

Rumination, emotion, and forgiveness: Three longitudinal studies.

“…However, when the number of observations per person provides adequate degrees of freedom, it becomes possible to specify more rigorous models that enable one to evaluate whether rumination on a given day is associated with avoidance and revenge motivation on the next day, even after controlling for previous levels of avoidance and revenge motivation. Establishing that changes in a putative x variable (e.g., rumination) precede changes in a putative y variable (i.e., forgiveness) would provide stronger (although by no means definitive) support for the proposition that the rumination-forgiveness association is a causal one (Finkel, 1995;West & Hepworth, 1991). Therefore, we conducted a third study using data from a 21-day diary study that allowed us to evaluate the temporal nature of the relations among our variables.…”

Daily negative interactions and mood among patients and partners dealing with multiple sclerosis (MS): The moderating effects of emotional support

“…Additionally, a diary method has some methodological advantages as compared to single-point self-report questionnaires. First, the problem of retrospection is minimized by reducing the time between actual occurrence and report (West & Hepworth, 1991). Second, by examining the temporal precedence of events within individuals, participants can be used as their own controls, thereby reducing the influence of confounding variables (West & Hepworth, 1991).…”

Chronic strain, daily work stress and pain among workers with rheumatoid arthritis: Does job stress make a bad day worse?

“…In all analyses, we controlled for day of the week by including four uncentered dummy variables (with Friday as the reference day) to model day of the week (West & Hepworth, 1991). The day variables (time-varying covariates) were included in all the analyses and modeled as fixed effects (no level 2 variance component) (Bryk & Raudenbush, 1992, p. 151).…”