The Ambiguous Loss Scale: Development and Validation with People Experiencing Pet Loss

Background and Purpose: Ambiguous loss is defined by Boss (2009) as an experience of grief and loss that lacks a culturally-defined ceremony to provide resolution. This lack of culturally-recognized closure can cause pain that affects the way individuals cope and grieve, making it difficult to find resolution. Many types of ambiguous loss have been identified, including missing-in-action soldiers, kidnapped children, miscarriage, and pet loss. While each of these is unique, they share three underlying dimensions: (a) frozen grief, (b) psychological distress, and (c) difficulty coping. However, existing research typically measures ambiguous losses using mental health (e.g., depressive) symptoms or using separate instruments related to the experience (e.g., measuring grief after miscarriage with the Perinatal Grief Scale) rather than considering ambiguous loss to share dimensions across experiences. This study details the development of an ambiguous loss measure and its validation with people experiencing the death or loss of a pet.

Methods: We developed a 22-item Ambiguous Loss Scale with three theoretically-justified subscales: (a) psychological stress, (b) frozen grief, and (c) coping. We established content validity by incorporating feedback from three expert panelists’ review. We recruited participants via convenience sampling and collected data via REDCap. Of the 750 participants who screened in, participants who did not complete the full survey (n=303) did not statistically differ from those that did (n=447) based on gender, relationship status, race, household income, or type of pet lost. We randomly split our sample into two subsamples (Sample 1 N=224, Sample 2 N=223) and conducted a series of confirmatory factor analyses in each to test the fit of three theoretically-informed models: (a) a simple one-factor model, (b) a correlated two-factor model, and (c) a correlated three-factor model.

Results: Our results indicated that, in both samples, the correlated three-factor model was the best fit for these data. After modifications, the final instrument contained 14 items (Sample 1: χ2(74)=278.277, p<0.001; CFI=0.964, TLI=0.955; RMSEA=0.111, SRMR=0.043; Sample 2: χ2(74)=253.776, p<0.001; CFI=0.972, TLI=0.966; RMSEA=0.104, SRMR=0.037). Each item significantly loaded on a factor, with coefficients ranging from 0.593 to 0.904. Configural, metric, and scalar invariance were established, suggesting measurement equivalence across gender and age groups. Evidence of construct validity was found with significant relationships in expected directions with the Pet Bereavement Questionnaire with correlation coefficients ranging from 0.409 to 0.755.

Conclusions and Implications: These findings support the use of the Ambiguous Loss Scale among people who have experienced pet loss. The hypothesized three-factor model was supported in both samples providing evidence that this scale may adequately measure ambiguous loss among pet owners. Future studies are needed to replicate and build on these findings in order to assess the utility of the Ambiguous Loss Scale in other samples with varied experiences of ambiguous loss.