Table 4.
Proportion (%) of purchases that were healthy in each period in the subsample according to purchaser and characteristics of the purchaser
| Pre-intervention | Signage | Signage + taste | Signage + taste + price | Post-intervention | P value for interaction with period | Overall mean for all periods | P value for main effect | |
|---|---|---|---|---|---|---|---|---|
|
Purchaser |
|
|
|
|
|
NS |
|
0.0068 |
| Adult only |
39.8 |
40.8 |
42.4 |
45.3 |
46.8 |
|
43.5† |
|
| Child only |
26.6 |
27.3 |
47.5 |
35.8 |
40.8 |
|
35.8‡ |
|
| Both present |
30.6 |
35.4 |
40.3 |
46.8 |
35.6 |
|
39.0‡ |
|
|
Weight status
a
|
|
|
|
|
|
0.0014 |
|
0.0125 |
| Non-overweight |
32.6 |
44.7† |
46.1 |
40.5† |
45.9 |
|
42.1 |
|
| Overweight |
35.7 |
27.5‡ |
35.6 |
50.8‡ |
43.4 |
|
39.7 |
|
|
Sex
a
|
|
|
|
|
|
0.0094 |
|
NS |
| Male |
30.3 |
38.6 |
33.6† |
50.3† |
42.5 |
|
41.3 |
|
| Female |
35.3 |
37.5 |
50.6‡ |
38.0‡ |
46.0 |
|
41.3 |
|
| Overall mean | 33.7 | 37.9 | 43.9* | 43.3* | 44.9* | 41.3 | 0.0048 |
n = 1032 items purchased by adults alone, n = 400 by children alone, and n = 1080 by adults and children together; A chi-square analysis was used to test for main effects and interactions.
aThe weight status and sex of the purchaser were recorded for transactions involving adults only and children only. They were not recorded when both adults and children were present during the transaction.
*Significantly different from pre-intervention, p < 0.05.
†,‡Values within a column with different superscripts are significantly different for that effect, p < 0.05.