Shaver, P. R., Belsky, J., & Brennan, K. A. (2000). Comparing measures of adult attachment: An examination of interview and self-report methods. Personal Relationships, 7, 25-43.
For a reprint, please contact Phillip R. Shaver, Ph.D., Department of Psychology, University of California, One Shields Avenue, Davis, CA 95616-8686. [e-mail]Summary
This article examines the relations between the AAI, Bartholomew and Horowitz's self-report attachment measure, and the multi-item romantic attachment scales designed by Collins and Read (1990) in a data set collected by Belsky and colleagues. The research participants were 135 mothers of one-year-old infants who were tested in the Ainsworth Strange Situation. The quantitative coding scales from the AAI were all significantly related to self-report romantic attachment measures, even though the two typologies (from the AAI and from Bartholomew and Horowitz's measure) were not significantly related. The authors conclude, as did Bartholomew and Shaver (1998) and Fraley and Waller (1998), that attachment measures are more precise when analyzed in terms of dimensions rather than types, and that different measures of attachment are related at the level of underlying dimensions, despite differences in focus (child-parent vs. romantic/marital attachments), content (discourse and defensiveness vs. experiences in romantic relationships), and method variance (interview coding, social desirability biases, etc.).
to Self-Report Measures of Adult Attachment