Aims and hypotheses

The aim of this study is to examine the association between self esteem at age 8 and pet ownership, in a large, well characterised both cohort.

Consideration of the literature indicates the complexity of this subject and leads to a set of interlinked hypotheses. Hypotheses to be tested are:

1. Children that live with pets at age 8 will have increased self esteem outcome of global self worth.
2. Children that have lived with pets all of their life will have increased self esteem compared to those that obtained pets at a later life-stage.
3. Children that live with dogs and cats will have increased self esteem compared to those that live with other pets such as fish.
4. The association between self esteem and pet ownership will differ between male and female children.
5. The association between self esteem and pet ownership will be stronger for children that are a single child or the youngest in the family, than those with older siblings, for which other support relationships are available.

Methods

We will use a large existing dataset, the "Avon Longitudinal Study of Parents and Children" (ALSPAC) study. This a prospective study which recruited 14,000 pregnant women with delivery dates between April 1991 and December 1992.

Self-esteem was measured in approximately 7000 children visiting the Children In Focus clinic at age 8years. A 12-item shortened form of Harter's Self Perception Profile for Children (28) was used, comprising the global self-worth and scholastic competence subscales. The measures used valid and reliable procedures explicitly designed to elicit data from children. The children were guaranteed confidentiality.

At a number of time points, the carer of the child (usually the mother) was asked 'do you have any pets' and 'how many of the following pets do you have'. Pet types included cats, dogs, rabbits, rodents (mice, hamster, gerbil etc), birds (budgerigar, parrot etc) and 'other' pets. However, from 2 years onwards a further two categories were added; fish, and turtles/tortoises/terrapins. Pet questions were asked during gestation, and at approximate child age 8 months, 21, 33, 47, 85, 97 and 122 months (up to 10 years old). The pet ownership variables have since been converted into binary responses (yes/no for that pet type) (29). The assumption will be made that the child also lives with the pets reported by the main carer.

The main predictor variables to be used in this study will be pet ownership reported at 8 years by the mother, and pet ownership history recorded throughout childhood, as previously described by the applicant (29). Univariable and multivariable regression analysis will be used to examine for an association between the outcomes of global self worth, or scholastic competence at age 8, with both pet ownership as a whole, and ownership of each pet type. Analysis will account for potential confounding variables available in the dataset which may be associated with self esteem (such as anxiety, depression and social competence), and variables related to pet ownership (as identified in our previous research, for example measures of socioeconomic status (29)). Stratification of samples by gender will be used to compare the association between pet ownership and self esteem in each sample. Stratification by sibling status will allow examination of differences between those children with siblings and those without, and any effect of being the youngest or an only child.