Matching theory to research: Applying the random intercept actor–partner interdependence model to parent–adolescent relationship quality

Objective Parent–adolescent relationships go through changes during adolescence but also have properties that remain stable. Background The unit of analysis (between person vs. within person) can significantly impact the conclusions drawn from research investigating characteristics of parent–adolescent relationships. Methods The current study utilized data from the Longitudinal Study of Child Abuse and Neglect (LONGSCAN; Runyan et al., 1998) to examine the parent–adolescent relationship throughout adolescence at 12, 14, and 16 years in a sample of 478 parent–adolescent dyads. We used a cross‐lagged actor–partner interdependence model and a random intercept cross‐lagged actor–partner interdependence model to compare how results change when examining between‐person versus within‐person associations. Results We found strong, positive actor effects for both mothers and adolescents as well as significant partner effects in the cross‐lagged actor–partner interdependence model. Within the random intercept cross‐lagged actor–partner interdependence model, the partner effects disappeared and there were only within‐person effects in adolescents. Conclusion Results of the current study highlight the importance of considering the unit of analysis in studying relational dynamics. Implications Researchers are recommended to carefully consider whether the unit of analysis matches between theory and statistical analysis in their studies.