. 2015 Dec 1;122(4):426–435. doi: 10.1016/j.prevetmed.2015.06.020

Table 3.

The numbers of each type of dog purpose reported by owners from different household types.

	Family	More than one adult	Retired	Single adult	Not reported
Household pet^a	1288	1231	205	153	64^a, ^c
Working dog^a	84^a, ^c	132^a, ^d	21	9	7
Assistance dog^b	8	11	10^b, ^d	3	1
Multi-purpose^b	7	9	2	2	0
Show dog	3	4	0	3	0
Breeding dog	1	1	0	0	0
Other^b	8	8	2	4	2
Not reported^a	515^a, ^d	350^a, ^c	47^a, ^c	61	51^a, ^d
Total	1914	1746	287	235	125

χ² tests performed with Bonferroni correction. For example, 84 of 1914 dogs in families were working dogs compared with 169 of 2393 in other household types. Due to low numbers in many categories, only household pet, working dog and purpose not reported categories were assessed for associations.

Fisher’s exact tests performed with Bonferroni correction. For example, 8 of 1914 dogs in families were assistance dogs compared with 25 of 2393 in other household types. Due to very low numbers, show and breeding dog categories were not considered.

Negative association, P < 0.003.

Positive association, P < 0.003.