A dynamic social relations model for clustered longitudinal dyadic data with continuous or ordinal responses.

Area of Science:

• Social psychology
• Quantitative psychology
• Statistical modeling

Background:

• Social relations models (SRMs) are used to analyze clustered dyadic interaction data.
• Existing SRMs have limitations in handling longitudinal data and dynamic structures.
• Distinguishing relationship effects from temporal fluctuations and measurement error is challenging.

Purpose of the Study:

• To propose an extension of social relations models for longitudinal dyadic data.
• To incorporate dynamic structures allowing for analysis of temporal dependencies.
• To investigate how individuals respond to their partner's previous behavior.

Main Methods:

• Developed an extended social relations model for longitudinal, clustered dyadic data.
• The model accommodates continuous, binary, or ordinal response variables.
• Applied the model to Canadian family data on conflict discussion tasks.

Main Results:

• The extended model successfully disentangles relationship effects from temporal fluctuations and measurement error.
• Demonstrated the ability to investigate partner effects across previous observations.
• The application provided insights into dyadic interactions within families.

Conclusions:

• The proposed extension offers a robust framework for analyzing dynamic social interactions in longitudinal dyadic studies.
• This approach enhances understanding of relationship dynamics and individual responses within pairs.
• The model is applicable to various types of dyadic data and research questions.