MODELING DIARY AND DYADIC DATA
<p>Dr. Jean-Philippe Laurenceau, Psychology, University of Delaware<br>Dr. Niall Bolger, Psychology, Columbia University</p>
This workshop is for social psychologists, family and medical sociologists, gerontologists, social workers, communication researchers, and researchers of marketing and organizational behavior.
Diary methods allow researchers to examine dyadic processes within the context of daily life in a way that is typically not possible with more traditional methods. Dyadic data acquired via diary methods present several data analytic challenges stemming from various sources of interdependence in these data. Traditional statistical methods (e.g., ANOVA, regression) typically focus on observations that are independently sampled, whereas dyadic diary data are inherently non-independent. Not only is there non-independence between members of the dyad, but also non-independence of observations within each dyad member. The hierarchical linear model (HLM) provides a conceptual framework and a flexible set of analytic tools to study interpersonal and family processes. One major application focuses on modeling longitudinal data where intensively measured (e.g., daily) time series data are clustered within persons. A second application concerns the analysis of dyads, where individual responses are clustered within couples, sibships, caregiving dyads, or other matched pairs. This workshop will consider the formulation of statistical models for these two applications (analysis of diary data, analysis of dyadic data) and their intersection (analysis of dyadic diary data).
The course will meet 8 hours per day, devoting equal time to lecture-demonstration and a computer lab. The focus will be on using the multilevel regression modeling framework (e.g., HLM) with some application using SEM. Participants should have strong backgrounds in multiple regression analysis.
A list of likely topics includes:
* Introduction to multilevel models for diary data
* Allowing for auto-correlated error terms
* Testing mediation in diary data
* Dyadic consensus and discrepancy
* Partners as multivariate outcomes
* Actor-partner interdependence model
* Exchangeable vs. distinguishable dyads
* Extending multilevel modeling to dyadic diary data
* Separating state and trait effects with centering
* Dichotomous outcomes
* Psychometrics of diary measures