Out Drinking the Joneses: Neighborhood Factors Moderating the Effects of Drinking on Relationship Quality over the First Four Years of Marriage.

Neighborhood quality has been cross-sectionally linked to both relationship behaviors and relationship well-being. Consistent with the Vulnerability Stress-Adaptation model of relationship functioning (Karney & Bradbury, 1995), we hypothesized that associations between social behaviors (e.g., drinking) and relationship quality could be moderated by neighborhood factors. Specifically, we characterized neighborhoods along multiple dimensions using multiple methods (self-report, census) to investigate how neighborhood factors might clarify ambiguous effects of alcohol use on marital functioning. A nationally recruited sample of 303 newlywed couples completed a baseline assessment around the time of marriage and was then assessed yearly across the first 4 years of marriage (94% retention). Three level HLM slope-intercept models were used to model changes in relationship satisfaction across the first 4 years of marriage. Results suggested that, for couples living in highly disordered neighborhoods, positive shifts in overall levels of drinking within specific waves of assessment were associated with corresponding negative shifts in satisfaction whereas in neighborhoods without perceived disorder, this effect was reversed. For couples living in neighborhoods with low levels of domestic structures (high census rates of single renters without children), within-couple discrepancies favoring higher rates of husband drinking in specific waves predicted poorer relationship quality for both partners in those same waves whereas those same discrepancies predicted higher satisfaction in high domesticity neighborhoods (high census rates of married homeowners with children). The findings provide insight into the different roles of alcohol use in relationship maintenance and highlight the importance of using external context to understand intradyadic processes.

Testing the ruler with item response theory: increasing precision of measurement for relationship satisfaction with the Couples Satisfaction Index.

The present study took a critical look at a central construct in couples research: relationship satisfaction. Eight well-validated self-report measures of relationship satisfaction, including the Marital Adjustment Test (MAT; H. J. Locke & K. M. Wallace, 1959), the Dyadic Adjustment Scale (DAS; G. B. Spanier, 1976), and an additional 75 potential satisfaction items, were given to 5,315 online participants. Using item response theory, the authors demonstrated that the MAT and DAS provided relatively poor levels of precision in assessing satisfaction, particularly given the length of those scales. Principal-components analysis and item response theory applied to the larger item pool were used to develop the Couples Satisfaction Index (CSI) scales. Compared with the MAS and the DAS, the CSI scales were shown to have higher precision of measurement (less noise) and correspondingly greater power for detecting differences in levels of satisfaction. The CSI scales demonstrated strong convergent validity with other measures of satisfaction and excellent construct validity with anchor scales from the nomological net surrounding satisfaction, suggesting that they assess the same theoretical construct as do prior scales. Implications for research are discussed.