The purpose of the current study was to develop and test the psychometric properties of the Family Genogram Interview (FGI), designed to improve standardization of genogram interviewing and to target assessment of four family emotional processes articulated in Bowen Family Systems Theory (Kerr and Bowen, 1988). The FGI was designed to incorporate assessment of both (a) factual questions about basic family demographics and nodal events using standardized genogram symbols (McGoldrick, et al., 1999), and (b) four dimensions of family emotional processes (i.e., marital conflict, emotional cutoff, symptoms in a spouse, & focus on a child), thought to be indicators of the systemic functioning in family systems (Bowen, 1978; Kerr & Bowen, 1988).
Stage 1 of the current study involved creating and revising the FGI. Items were written to create the FGI, and the subscale psychometrics were then evaluated. Based on reliability analyses at the item level, 4 FGI were removed for inadequate reliability. At the subscale level the FGI-Focus on a Child subscale was removed due to low internal consistency reliability. Based on these analyses the final version of the FGI was established and used in subsequent analyses in Stage 2.
In Stage 2 the inter-interviewer reliability of the participant scores in the FGI was determined to be on average at 95% agreement, indicating the FGI can be reliably administered across participants. Construct validity assessment of the FGI subscale scores yielded mixed results, with two subscales, FGI-Marital Conflict and Symptoms in a Spouse, corresponding with existing measures of similar constructs. However the third subscale, FGI-Emotional Cutoff, did not demonstrate a relationship to a similar existing measure as hypothesized. Contrary to expectations, the FGI subscales also did not correspond with the Differentiation of Self Inventory (DSI). However, there were also some informative, yet unpredicted results lending support for the construct validity of the FGI-EC subscale. Consistent with Bowen Theory, the FGI-EC significantly correlated with the DSI-ER and MSI-FD subscales. The mixed results indicate there appears to be some preliminary support for the construct validity of three remaining FGI subscales (i.e., Marital Conflict, Symptoms in a Spouse, and Emotional Cutoff. The clearest limitation is that there was little evidence that emerged to support the notion that FGI subscales tap into systemic family emotional processes, which they were originally designed to assess. Using a cluster analysis, two groups were found to exist in the participant scores on the FGI. The two groups differed on FGI-Marital Conflict and Emotional Cutoff scores, with one group having high FGI-Marital Conflict scores and low FGI-Emotional Cutoff scores and the other group displaying the opposite. The groups did not differ on demographic characteristics or self-reported levels of differentiation of self, suggesting that different patterns may exist among individuals at similar levels of differentiation. Limitations of this study and directions for future studies are discussed.