The aim of this study was to investigate the relationships between childhood attachment to pets, pet care, compassion to animals, and attitudes towards animals. We first examined socio-demographic factors and focused particularly on pet ownership and types of pets owned. We found that the majority of children scored high on attachment to pets, but these attachment scores differed depending on pet ownership, pet type, and gender of the child. We found associations between attachment to pets and caring behaviour, friendship behaviour, compassion, and attitudes, and examined the direction of these relationships. We will begin by discussing the findings in detail before considering the implications and limitations of the study.
Our results demonstrate that children (at least in this Scottish sample), are highly attached to their pets. The mean score for attachment was 15 (score of 9 was highest and 44 was lowest attachment score possible) and 69.2% of the children scored below 15. Pets are important in children’s lives, with 80% of our sample reporting that they loved pets, 83% of those with pets reported that their pet made them happy, 76% reporting that their pet was their best friend, 62% reporting that they would be lonely without their pet, and 52% reporting that they felt that their pet knew when they were upset and it tried to comfort them. These findings support previous research demonstrating the significance of pets in the lives and social networks of children, that they form close emotional connections to their pets, and that pets may provide a source of affection and comfort [39
]. Children are emotionally expressive towards their pets and are strongly connected to them, often reporting them as one of the most important figures in their lives [70
]. Both the current study and previous work therefore demonstrate that pets are important source of emotional attachment in the lives of children and support the notion that pets should not be overlooked in attachment research.
When we examined strengths of the associations between attachment to pets and caring and friendship behaviour, compassion, and attitudes towards animals, we found that caring behaviour, friendship behaviour, and compassion were significantly associated with attachment scores. Attachment to pets was significantly associated with attitudes towards animals. The finding that caring behaviours (such as spending time with pets, cuddling, stroking, and playing with pets) and friendship behaviours (such as telling secrets to, crying with when sad, and talking to pets) were significantly associated with attachment to pets is consistent with previous findings [71
]. Although we cannot ascertain causation, these findings suggest that children’s participation in pet caring roles at home may possibly foster attachment to their pets, which may have positive outcomes for the child (e.g., improved well-being and quality of life [37
]) as well as the animal (e.g., better care and welfare [27
]). From direct experience, children learn to be nurturing, and develop the ability to recognise, understand, and share the feelings of others [73
]. This may also explain why attachment to pets is related to children’s beliefs about animal minds [64
], the attribution of emotions to pets [75
], and predicts a more reliable and consistent ability to interpret animal behaviour and facial expressions [76
]. The finding that compassion towards animals (such as feeling upset and wanting to help when an animal is hurt or upset) is associated with attachment scores, suggests that children’s attachment to their pets is connected with empathetic and compassionate orientations towards pets. Again, although we have not directly tested causation here, this makes sense, as those who are securely attached to others are more likely to develop compassion in their relationships [78
]. Humane education or other activities that facilitate empathy and compassion could therefore potentially promote positive attachment to pets. Promoting compassion and empathy towards animals has important implications for prosocial behaviour towards other children, as animal-directed empathy can generalise to human-directed empathy [79
Furthermore, we found that attachment to pets significantly predicted positive attitudes to animals, and from previous research we know that attitudes are associated with empathy, pro-social, and humane treatment of animals, greater concern for animal welfare, and less cruelty [11
]. Our findings support previously identified links between pet attachment, empathy, positive attitudes to animals, and a prosocial orientation and behaviour [58
]. Attachment to pets is a relational concept rather than an attitude or ideology [84
]. High attachment to pets can promote positive child-pet relationships, whilst low attachment to pets may have a negative impact on child-pet relationships [35
]. High attachment facilitates nurturance and humane behaviour, whereas low attachment has been related to higher acceptance of animal cruelty [14
]. Animal cruelty and neglect may be associated with a lack of emotional attachment between child and their pet.
Considering socio-demographic factors in attachment to pets, we found that girls were significantly more attached to their pets than boys. Girls also scored significantly higher on caring behaviour, friendship behaviour, and compassion, which supports previous findings in other studies of children [37
] and adults [85
]. However, other studies have found no gender difference in pet attachment [63
] or in care related behaviour towards pets [86
]. These mixed findings may be explained by different populations and measurement tools. We found no significant difference in attachment scores between low, medium, or high family affluence, which is consistent with previous findings that also found no difference [37
], although Westgarth et al. [63
] found that deprivation increases with the number of dogs owned. However, the internal reliability of the Family Affluence Scale used in this study was low and results should be interpreted with caution. This measure is widely used in child health research and has been developed specifically as a child and adolescent self-report research tool. We also found no significant difference in attachment to pets between younger and older children, however, research with adolescents shows that attachment to pets decreases between 11 and 15 years of age [37
]. It may therefore be worth investigating attachment to pets in a wider age range in the future. We did find, however, that older children scored higher on caring behaviour, possibly due to increased responsibility for pet care within the home [27
Our findings show that children with pets (or who had pets in their past) scored higher on all attachment items than children without pets or those who had never had a pet. These findings suggest that children who grow up with pets have an early opportunity to form pet attachments, become emotionally connected to their pets through direct experience, spend quality time with their pets, and experience a ‘sharing of significant moments’, which is consistent with previous research [37
]. For example, children view pets as confidantes for secrets, demonstrating this emotional connection [69
]. Although pet ownership alone is important, it seems to be that having a pet that a child feels is their own has the most influence, as we found that children with their own pet scored higher on pet attachment. This finding supports the argument that a close relationship and emotional bond is more important than merely the presence of a pet within a home [82
], possibly due to increased direct involvement and responsibility for the care of their pet, thus facilitating attachment.
In relation to pet type, we found that the type of pet that a child has influenced the degree of child-pet attachment. Children with pet dogs scored the highest on attachment, which makes sense given children demonstrate high attachment to dogs, view dogs as attachment figures, and have more direct contact with dogs inside and outside of the home [63
]. Dogs are also more likely to read and adapt their behaviour in response to human emotional signals [88
]. Children also demonstrated high attachment to cats, followed by small mammals, fish/reptiles/amphibians, and other (e.g., horse), with the lowest scores being shown for birds. Pet dogs and cats were the only significant predictors of child-pet attachment. The finding that higher attachment is shown for dogs and cats is consistent with previous research in adults [89
] and children [83
] and may be explained by the wider range of behaviours that can be displayed between children and their pet dog or cat compared to other animals. Dogs and cats may also be more receptive to our emotions and express their emotions and behaviour more clearly than other pet animals, facilitating a closer connection or bond. However, most attachment research uses measures that are based on direct interactions with pet dogs (e.g., groom and walk) rather than emotional aspects (e.g., joy and love) that can be expressed for all pets [89
], which may explain these findings.
Limitations and Future Directions
This was a large-scale questionnaire study using self-report data and thus might be subject to response biases such as social desirability, potential peer influence, and demand characteristics. Although we did not measure behaviour directly there is a strong evidence base for the link between attitudes and behaviour [90
]. Self-report questionnaire methods are a tried and tested approach for children of this age range, but it is possible that a minority of the younger children included in the sample may have needed some teacher support in completing the questionnaire (provided as part of the data collection procedure). However, only two children were under the age of 7 years and teachers were instructed to help children only with reading items and were discouraged from interpreting items or suggesting answers to minimise teacher effects. Future research would benefit from using a combination of self-reports, parent reports, observational, and behavioural methods to allow data triangulation and ensure accuracy of findings.
There are also some limitations with the measures used in this research. In the humane behaviour measure, the sub-scale “aggression towards pets” consisted of only one item and was excluded from analysis, however, we know from previous studies that low attachment to pets is related to cruelty to animals [14
]. The SAPS, used to measure pet attachment, was highly reliable within our sample and in those which validated its use [37
] and is a short measure of attachment developed for use in large survey research. It should be noted that it does not cover all aspects of attachment relationships, focusing more on the emotional aspects of pet attachment such as ‘crying with’ (comfort) and feeling lonely without them (separation anxiety) and the friendship aspects of pet attachments, such as playing with and spending time with their pet every day. Other aspects of attachment relationships (e.g., secure base behaviour) are not included, which is a potential limitation. To enable all children to answer the SAPS they could answer about a pet they had or a hypothetical pet (this could be a pet they used to have or one owned by a relative). Our research indicates differences in attachment between children who currently have pets and those who do not. Further analysis could explore in more depth differences in pet attachment between current pet owners and those who currently do not have a pet, but perhaps have had a pet previously, wish to own a pet, or have a strong connection to a pet owned by someone else (relative, friend, or neighbour). It would also be interesting to look at differences between how children respond to the items within SAPS in relation to different pets (e.g., dog owners and cat owners [91
]). In terms of the analysis exploring associations between pet care and attachment, it should be noted that two items in SAPS (“I spend time every day playing with my pet” and “I talk to my pet quite a lot”) are similar to those on the pet care measure (such as “play with” and “talk to”) and this may have influenced the findings. Despite these limitations, SAPS provides interesting insights into children’s attachment to pets, and is linked to positive outcomes such as quality of life [37
] and humane behaviour (as seen in the current study). It would be interesting to use this measure in samples outside the UK to examine its cross-cultural reliability and to complement SAPS with other measures of pet attachment using mixed methods approaches. Although research on attachment to pets is expanding, there are still inconsistent results [92
] which may be due to the use of different measures of attachment. Developing and refining age-appropriate measures for child-animal interaction research remains a priority for research.
In a short survey it is not possible to capture data on all variables of potential interest. In this study we did not consider family dynamics such as dual or single-parent families and sibling status, which could have influenced attachment scores (children in single parent families and youngest children show greater attachment to their pets [63
]). Although we used a diverse sample from across Scotland that included a variety of ethnicities and religions we did not include measures of ethnicity, religion, or cultural background, which have been shown to influence human-animal interactions [85
]. Future research might also consider the impact of pet loss and grief as an indicator of pet attachment and the impact it may have on children’s development and mental health, including anxiety and depression [94