Unpacking the Multispecies Family: Predicting Pets as Family Members Using the General Social Survey

ABSTRACT The multispecies family has grown rapidly over the past 30 years in the United States. Scholarly understanding of pets as legitimate family members is increasing, but most work has been qualitative in nature. Statistical modeling of these dynamics has been bound by a lack of access to large‐scale, nationally representative datasets paywalled by industry actors like the American Veterinary Medical Association and American Pet Products Association, yet quantitative analysis of this growing family type is greatly needed. In answer to this need, the 2018 General Social Survey (GSS) included several questions regarding pets and relationship dynamics. The objective of this paper is to use these data to model select variables as predictive of who considers their pet part of the family in the United States. Using binary logistic regression, I measure the likelihood that children ever born, gender, Catholicism, and singlehood predict respondent assignment of family status to household pets. Respondents who had never had children were 107% more likely than those with at least one child to almost always assign family status to their pets. Women were 99% more likely to almost always label their pets as family. Catholics were about 40% less likely to do so than non‐Catholics. Singlehood was not significant. No statistical research considers how household structure (as modeled here) impacts the assignment of family to pets in the United States. Furthermore, these findings advance research needed for future work examining how pet ownership impacts fertility intentions.