Measurement of attachment in human-animal interaction research

Abstract

Abstract Researchers in the field of human-animal interaction often investigate the construct of the human-animal bond in order to measure contextual information about companion-animal relationships beyond simply living with a pet. This practice allows researchers to go beyond a simple manifest variable (e.g., dog-owning vs. cat-owning participants) to investigate how specific relationships with companion animals may impact outcomes. However, the current measures of the relationship between people and their pets (i.e., attachment) are flawed. Specifically, their initial validations were based on homogenous samples, the original conceptualizations of the human-animal bond were fragmented, and some used improper statistical techniques in determining the underlying factor structure of the measure. The current study utilized confirmatory factor analysis ( n = 589) to examine the factor structure and underlying psychometric properties of the Lexington Attachment to Pets Scale and the Short Attachment to Pets Scale. The factor analysis revealed that the Lexington Attachment to Pets Scale and the Short Attachment to Pets Scale both have issues with model fit in a sample of undergraduate students at a public Southwestern University. This article offers recommendations for future research investigating the human-animal bond.