Predictors of relationship satisfaction during the COVID-19 pandemic

Prior work and theory suggest many vulnerabilities, stressors, and adaptive processes shape relationship satisfaction. In the current research, we used machine learning to understand which constructs have greater predictive importance for perceived changes in satisfaction since the pandemic began and satisfaction over the prior week. In a large sample collected at the beginning of the pandemic (N = 1873; Study 1), relationship processes were most predictive, explaining up to 70% of variance in satisfaction. Feeling appreciative of one's partner and being satisfied with quality time spent with one's partner were consistently top predictors of satisfaction. We also examined whether these important predictors were associated with changes in relationship satisfaction across the first year of the pandemic in a longitudinal subsample (N = 618; Study 2). Appreciation and satisfaction with quality time were associated with high and relatively stable relationship satisfaction over time.

Automated summary

This research used machine learning to identify the key factors contributing to changes in relationship satisfaction during the pandemic. Relationship processes were found to be the most important predictors, explaining up to 70% of the variance in satisfaction. Appreciation of one's partner and satisfaction with quality time spent together consistently emerged as top predictors of satisfaction. Longitudinal analysis further showed that appreciation and satisfaction with quality time were associated with high and relatively stable relationship satisfaction over time.