Psychometric evaluation of the Swedish Quality of Dyadic Relationships scale - homogeneity and construct validity.

RATIONALE AND AIMS: The Quality of Dyadic Relationships is a self-assessment scale used to evaluate various aspects of relationship quality. Psychometric evaluation by the developers of the instrument has led to a nontested amended version. Further psychometric testing is thus warranted, and the aim of this study was to evaluate homogeneity, construct validity (in terms of concurrent, discriminant and known-groups validity) and any floor and ceiling effects of the Quality of Dyadic Relationships. METHODS: Forty-seven cohabitant couples (47 women with a mean age of 30.0 years and 47 men with a mean age of 31.5 years) answered the Quality of Dyadic Relationships, the Relationship Assessment Scale (to test concurrent validity) and the Perceived Stress Scale (to test discriminant validity). Homogeneity (internal consistency) was calculated by Cronbach's alpha. Concurrent and discriminant validity were estimated as correlations between Quality of Dyadic Relationships and the other instruments. Assessment of known-groups validity was based on the variables of parental status and gender. Floor and ceiling effects were evaluated according to frequency distribution. RESULTS: The overall homogeneity was good with acceptable Cronbach's alpha values (alpha > 0.70) for all subscales but dyadic sexuality. Concurrent validity and discriminant validity were found. Known-groups validity was indicated by significant differences between individuals with different parental status on the total QDR index, where the ones without children scored higher. No difference between the genders was found. No significant floor effects were found, but a significant ceiling effect was found in the subscale dyadic sensuality, with 27.7% of respondents scoring maximum. CONCLUSION: In all, the QDR showed promising psychometric properties and may be used for screening and follow-up purposes. However, it can benefit from further development, as suggested by the ceiling effect in the subscale dyadic sensuality and the low internal consistency in the subscale dyadic sexuality.