BRIEF REPORTS The Increased Construct Validity and Clinical Utility of Assessing Relationship Quality Using Separate Positive and Negative Dimensions Richard E. Mattson Auburn University Dawnelle Paldino and Matthew D. Johnson State University of New York at Binghamton The Positive and Negative Quality in Marriage Scale (F. D. Fincham & K. J. Linfield, 1997) is a self-report measure that separately assesses positive and negative dimensions of relationship quality. Fincham and Linfield found that ratings of positive and negative marital quality accounted for unique variance in maladaptive attributions and self-reports of dyadic behavior beyond that accounted for by the Marital Adjustment Test (H. J. Locke & K. M. Wallace, 1959), a widely used measure of marital quality that combines these dimensions. The current study expanded on these findings using a different measure of relationship quality and observed dyadic behavior with a sample of engaged couples (N = 43). The results indicate that a two-dimensional approach to measuring self-evaluations of relationship quality is more informative than a one-dimensional approach during the engagement period. Keywords: marital quality, marital satisfaction, measurement, premarital, engaged One way to assess marital quality is through a spouse’s subjec- tive, evaluative judgment of his or her marriage or partner (Fin- cham & Bradbury, 1987). These evaluations are theorized to represent the common way in which underlying marital dysfunc- tion is expressed across couples (Jacobson, 1985). Measures as- sessing these evaluations typically contain rating scales anchored by positive (e.g., satisfied) and negative (e.g., dissatisfied) end- points on a single dimension. Responses on these scales are summed into a single index reflecting the spouse’s overall evalu- ation of the quality of the marriage. However, in a study based on advances in attitude assessment (e.g., Thompson, Zanna, & Grif- fin, 1995), Fincham and Linfield (1997) wrote that marital quality might not reduce to a single and undifferentiated evaluation of the marriage. Rather, they suggested that marital quality might com- prise at least two separate evaluative judgments representing pos- itive and negative dimensions. To test this assumption, Fincham and Linfield (1997) developed the Positive and Negative Quality in Marriage Scale (PANQIMS), which measures positive and negative subjective evaluations of the marriage or partner using separate items for each dimension. Using confirmatory factor analysis and a sample of married couples, Fincham and Linfield demonstrated that a two-factor structure (i.e., positive and negative dimensions) of the PANQIMS provided a better fit to the data than did a single latent construct. It was also demonstrated that the PANQIMS provided additional utility over the Marital Adjustment Test (MAT; Locke & Wallace, 1959), a commonly used one-dimensional measure of marital quality. Spe- cifically, the separate positive and negative dimensions accounted for unique variance in self-reports of dyadic behavior and mal- adaptive attributions for partner behaviors beyond what was ac- counted for by MAT scores (for another example, see Menchaca & Dehle, 2005). In the present study, we attempted to expand on the psychomet- ric findings regarding the PANQIMS in three ways. First, this study used a sample of engaged couples. In doing so, we expanded on the generalizability of the Fincham and Linfield (1997) findings by assessing couples at a relationship stage that has been suggested as a target for future research (e.g., Karney & Bradbury, 1995). Furthermore, the importance of studying this stage is underscored by research suggesting that dysfunctional dyadic interactions dur- ing the engagement period predict the development of marital problems (e.g., Markman, 1981). As the couples in this study were not married, the term relationship quality is used because it more accurately denotes the construct being measured. Second, the PANQIMS was compared to a measure of rela- tionship quality that more accurately tests the psychometric utility of a two-dimensional assessment over a one-dimensional one. Specifically, comparing the PANQIMS to the MAT may not truly test the added utility of a two-dimensional approach because the MAT includes both evaluations of marital quality and self-reports of dyadic behavior, whereas the PANQIMS uses evaluations only. The comparison of one- and two- dimensional approaches is therefore confounded by differences in the types of questions used to assess relationship quality. Therefore, the PANQIMS was evaluated in relation to an ad- aptation of the Semantic Differential (SMD; Osgood, Suci, & Richard E. Mattson, Department of Psychology, Auburn University; Dawnelle Paldino and Matthew D. Johnson, Department of Psychology, State University of New York at Binghamton. This study was conducted while Richard E. Mattson was at Binghamton University, State University of New York. Zachary Collins, Peggy Echt, Laura Frame, Mark Lenzenweger, and Stephen Lisman reviewed drafts of this article. Courtney Centrelli, Zachary Collins, Peggy Echt, Laura Frame, Brianna Friedman, Patrick Harvey, Kristin Heron, Diana Interrante, Hey Bruce Li, Matthew Price, and Rebecca Osterhout aided in the collection and coding of data. We gratefully acknowledge their assistance. Correspondence concerning this article should be addressed to Richard E. Mattson, Department of Psychology, Auburn University, 226 Thach, Auburn, AL 36849-5214. Psychological Assessment Copyright 2007 by the American Psychological Association 2007, Vol. 19, No. 1, 146 –151 1040-3590/07/$12.00 DOI: 10.1037/1040-3590.19.1.146 146 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.