Supplementary file 1_Development and validity evidence of the dog–human attachment scale.pdf

The present study aimed to develop and provide validity evidence for the Dog–Human Attachment Scale (DHAS), designed to assess dogs’ attachment styles toward their caregivers. Content validity of the proposed items was confirmed by experts. To examine the internal structure of the instrument, data from 1,014 Brazilian dogs were analyzed based on owner responses to the DHAS, with 713 cases used for exploratory factor analysis (EFA) and 301 for confirmatory factor analysis (CFA). Parallel analysis and exploratory factor analysis revealed a three-factor structure: Anxiety, Avoidance, and Insecurity, capturing distinct aspects of canine attachment. The CFA supported the adequacy of the three-factor model. Test–retest procedures demonstrated good temporal stability. Evidence of internal structure validity was found, including acceptable composite reliability and ordinal McDonald’s omega coefficients for all three factors, as well as no evidence of differential item functioning (DIF) across guardian gender. Finally, latent profile analysis based on participants’ scores on the three dimensions identified an optimal three-profile solution in the sample, corresponding to insecure-anxious (16.2%), disorganized (47.3%), and insecure-avoidant (36.5%) attachment styles. These findings highlight the utility of the DHAS in distinguishing variations in dog–human attachment patterns. The instrument offers a reliable and valid tool for advancing research and clinical practice, contributing to a deeper understanding of how attachment mechanisms shape canine emotional regulation and behavior in relation to caregivers and influence the quality of the human–dog relationship.