Evaluating a New Measure of Adult Attachment: The Tripartite Attachment Battery
The Tripartite Attachment Battery (TAB) is a new set of self-report scales of individual differences in adult attachment characteristics. It measures: (a) attachment security (Secure Attachment Scale), (b) attachment anxiety and avoidance, as well as their corresponding secondary attachment strategies (Organized Insecurity Scale), and (c) disorganized attachment (Disorganized Attachment Scale). The primary goal of the current research was to refine the TAB and examine its psychometric properties. In Manuscript 1, new items were added to improve TAB subscales that were previously found to have poor internal consistency reliabilities. Data were collected from 538 university and community members via an online survey. The factor structure, internal consistency reliability, and construct validity of the TAB measures were evaluated. The factor analyses supported subscales similar to those identified in previous research (i.e., four Secure Attachment Scale subscales, six Organized Insecurity Scale subscales, and a unidimensional Disorganized Attachment Scale). They all had internal consistencies that were acceptable or better. There was support for the construct validity of the higher-order TAB scales based on their relationship with childhood maltreatment and current psychological functioning variables. There was also support for the construct validity of specific TAB subscales based on their relationships with variables theoretically related to the constructs measured by each subscale. In Manuscript 2, a short-form version of the TAB (TAB-SF) was created to increase the practicality of its use. Using data from Manuscript 1, items for the TAB-SF were selected in a preliminary study. In the main study, 365 community members completed the TAB-SF via an online survey. Factor analyses of the TAB-SF scales supported subscales similar to the TAB. However, one Secure Attachment Scale subscale and one Organized Insecurity Subscale were not supported. The TAB-SF had internal consistencies and test-retest reliabilities that were adequate or better, and its correlations with two measures of psychological functioning were supportive of their construct validity. The TAB-SF contained 38 fewer items than the TAB. The TAB and TAB-SF are promising measures of adult attachment. Directions for using and further refining these measures are discussed along with the implications of the findings for attachment theory.
Attachment Theory, Attachment Behavior, Attachment Security, Measurement
Doctor of Philosophy (Ph.D.)