Psychometric Evaluation of the Comfort from Companion Animals Scale in a Sexual and Gender Minority Sample

Abstract

Human-animal interaction research is growing in popularity and methodological rigor; however, there remains a need for psychometrically validated measures and inclusion of broader populations. This study addressed these gaps by reporting on the psychometric properties of the Comfort from Companion Animals Scale (CCAS) in a sample of sexual and gender minority emerging adults. Participants included 138 emerging adults between the ages of 18-21 years (M = 19.33 years, SD = 1.11; 38.4% racial/ethnic minority) who identified as a gender (48.6%) and/or sexual minority (98.6%) and who reported living with a companion animal in the past year. We utilized the following analytic methods: (a) confirmatory factor analyses to compare the unidimensional structure of the CCAS with the two alternative models, (b) multiple group analyses to test measurement invariance across demographic groups, and (c) structural equation models to evaluate construct validity. Preliminary analysis found that the majority of participants did not endorse the two lowest response options. To conduct invariance testing, we eliminated items 3, 5, and 8 from the CCAS and collapsed the lowest response options. The results of the confirmatory factor analysis supported the use of this revised unidimensional model. We found evidence of measurement invariance across gender identity, sexual orientation, and race/ethnicity groups. Construct validity was supported by comparing the CCAS with factors on the Pet Attachment and Life Impact Scale; the positive association between the CCAS and anxiety are discussed in the context of prior research. Overall, our findings highlight the importance of validating human-animal interaction measures across samples from diverse backgrounds. We recommend that future studies continue to test the CCAS and other measures of human-animal attachment among diverse samples to delineate which aspects of human-animal interaction may be most beneficial in promoting mental health in vulnerable populations.

Keywords: Attachment; LGBTQ; emotional comfort; human–animal bond; human–animal interaction; measurement.

Figure 1.

Structural Equation Model of associations between CCAS and domains of psychological stress. Note: n=136. Due to including the sexual orientation subgroups used in invariance testing as a covariate, two participants who identified as heterosexual were excluded. β indicates that the results of the SEM have been standardized in standard deviation units, such that a 1 standard deviation increase in CCAS corresponds to a change in the standard deviation of the outcome variables equivalent to the β-coefficient value (e.g., 0.25). To reduce complexity, covariates, residuals, and residual covariances are not reported in the figure. Dotted lines represent nonsignificant relations; solid lines represent significant associations.