Development of the Dog Attachment Insecurity Screening Inventory (D-AISI): A Pilot Study on a Sample of Female Owners

SIMPLE SUMMARY: The Strange Situation Procedure is a laboratory test originally designed to assess the quality of a child’s attachment bond to their mother and is widely used in dogs to assess their attachment bond towards the owner. However, the SSP is time consuming and limits the amount and variety of obtainable data. In order to overcome these limitations, we adapted a three-dimensional parent-report scale, named the Attachment Insecurity Screening Inventory (AISI) 6–12, originally developed to assess 6- to 12-year-old children’s attachment insecurity, to dog–owner dyads, and we assessed scale consistency and validity. A first statistical analysis performed on the responses provided by 524 female owners to the online questionnaire revealed five scale dimensions named “physical contact”, “control”, separation anxiety”, “owner as emotional support”, and “owner as a source of positive emotion”. However, a further forced extraction of three components resulted in subscales that mirrored the ones reported for the original AISI in terms of item composition (i.e., ambivalent, avoidant, and disorganized). The three subscales also had satisfactory to good measures of internal reliability. The final scale was named the Dog Attachment Insecurity Screening Inventory (D-AISI). Although promising, it needs to be refined and tested for more validity measures. ABSTRACT: To date, the Strange Situation Procedure is the only tool available to investigate the quality of the dog’s attachment bond towards the owner. This study aimed to adapt a parent-report scale, named the Attachment Insecurity Screening Inventory (AISI) 6–12, originally designed to assess 6- to 12-year-old children’s attachment insecurity, to dog–owner dyads and assess measures of consistency and validity. The online questionnaire was completed by 524 female dog owners. Principal component analysis (PCA) revealed five components named, respectively, “physical contact”, “control”, “separation anxiety”, “owner as emotional support”, and “owner as a source of positive emotion”. Because of the three-factor structure of the original AISI, a PCA with a pre-fixed set of three factors was also performed. The resulting subscales mirrored the ones found for the original scale (i.e., ambivalent, avoidant, and disorganized), although four items did not fit the model. Internal reliability appeared to be satisfying for the ambivalent and the disorganized subscales, and good for the avoidant subscale. The theoretical background and the results of this study suggest that the three-dimensional model represents a better solution for the interpretation of the Dog Attachment Insecurity Screening Inventory (D-AISI). Although promising, this scale requires refinement and assessment of additional validity measures.