Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation




Educational, School, and Counseling Psychology

First Advisor

Dr. H. Thompson Prout


There are hundreds of thousands of children living in foster care in the United States on any given day. Mental health professionals may be called upon to assist with evaluating the parental capacity of these children’s parents in order to inform reunification decisions. One of the key parental capacity domains to be evaluated is the relationship between parent and child (Schmidt et al., 2007). The Working Model of the Child Interview coding scheme (WMCI; Zeanah et al., 1996) is one tool for evaluating this relationship. There is a significant practice-to-research gap with this measure. To date, no peer-reviewed studies have established the WMCI coding scheme’s psychometric properties; the need for such research with this measure has been cited in the literature (Sprang, Clark, & Bass, 2005).

In this dissertation, the literature was reviewed on attachment theory and internal working models of caregiving, the WMCI and other measures of similar constructs, the importance of establishing a measure’s construct and criterion-validity, and forensic standards for measures. Then, it outlined a series of research questions aimed at exploring the psychometric properties of the WMCI Coding Scheme with mothers who have maltreated. Next, the methodology was described. Based on the results of the 403 biological mothers who completed the WMCI as part of court-appointed evaluations following child maltreatment, the items of the WMCI Coding Scheme were best conceptualized using two-factors: Quality and Content. This finding was consistent with the manual. Significant differences in item-level scores existed for all three WMCI descriptive classifications and for overall factor scores. Due to the instability of the two-item Content factor, this dissertation explored the use of a WMCI Total Score (combining the two factors) and using only the WMCI Quality factor and items. No significant relationships existed between any demographic characteristics and WMCI factor scores. Weak, negative correlations with other measures provided some evidence of convergent validity. In conclusion, some potential clinical/research implications for the WMCI Coding Scheme were made and limitations and future directions were described.