RELIABILITY AND VALIDITY OF SELF, SPOUSE, AND OBSERVER ASSESSMENTS OF OPEN COMMUNICATION IN MARRIAGE
Abstract
This study explores the relationships among self, spouse, and observer measures of open communication in marriage. A theoretical domain of open communication is presented which integrates message content, communicator style, and target components. Reliability, construct, and criterion related validity are examined for measures of the domain. The sample consists of 150 married individuals (75 couples) who completed self-report scales of open communication and marital quality. A subsample of 26 married subjects (13 couples) voluntarily participated in a behavioral observation phase in which they discussed a topic pre-rated by them as "difficult to discuss openly." The videotaped interaction of the 26 spouses was rated by trained observers for open content and open style characteristics using a behavioral observation system developed for this study. In addition, spouses rated their own and their partner's behavior for openness using a simplified version of the behavioral rating system. The results are presented in relation to three psychometric concerns. First, all measures demonstrate acceptable reliability as measured by Cronbach's alpha. In addition, the measures seem to reflect similar construct component structures. The internal structures support a theoretical domain of open communication which includes content characteristics (topic intimacy, verbal immediacy, content valence, and general content openness), and style characteristics (non-verbal, valenced, emotional, receptive, and general style openness). Second, the component and subcomponent correlations across measures are low suggesting that different information is provided about open communication by different measurement perspectives. Third, only the self-report measure of open communication yields significant relationships with the criterion variable--marital quality. Nondistressed spouses report higher levels of general content openness, negative open style, emotional open style, and general style openness than distressed spouses. Self, spouse, and observer behavioral observation assessments do not consistently discriminate between high and low levels of marital quality. Methodological procedures require that the findings be considered descriptive rather than estimates of population characteristics. The results are discussed in terms of factors that affect measurement of open communication such as cross-situational consistency of behavior, objective self-awareness, social desirability, and perceptual differences among communicators, spouse, and observers. Methodological implications for the development of sound, practical measures of open communication are suggested.
Degree
Ph.D.
Subject Area
Social research
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.