For the MMR group, the children (71 girls; 76 boys) were aged 1–4 years (mean = 2.71; SD = 0.75;

median = 3). For the dTaP/IPV group, the children (50 girls; 58 boys) were aged 1–4 years (mean = 2.72; SD = 0.76; median = 3). Self-reported uptake of primary immunisation was high. For children in the MMR group, 132 (89.8%) had received the first MMR, three (2%) had received the separate measles, mumps and rubella components, and nine (6.1%) were not immunised against these diseases; 138 (93.9%) had completed vaccinations against diphtheria, tetanus, pertussis, polio and Hib before 1 year of age; three (2%) were not immunised and for four children (2.7%) the parents indicated that this was unknown (information was not provided for the remaining children). For children in the dTaP/IPV group, 98 PD0332991 molecular weight (90.7%) had received the first GW3965 mouse MMR, one had received the separate components,

seven (6.5%) were not immunised and for one child the parent indicated that this was unknown; 105 (97.2%) had completed vaccinations against diphtheria, tetanus, pertussis, polio and Hib, one child was not immunised against these diseases and one parent indicated this was unknown (one parent did not provide uptake information). Parents in the two groups differed only in terms of sex, χ2(1, n[MMR] = 147, n[dTaP/IPV] = 108) = 5.543, exact p = 0.024 and number of children, U = 6621.500, n[MMR] = 147, also n[dTaP/IPV] = 108, p = 0.012; with more fathers in the dTaP/IPV group and those in the MMR group having more children. No differences were found on other parent or child characteristics (p > 0.05). The items measuring each TPB component should correlate with each other and exhibit high internal consistency [12]. Thus, it was necessary to determine whether the items designed to measure each component (Table 1) fulfilled these requirements. Firstly, because parents were asked to complete an identical set of questions about either MMR or dTaP/IPV, the following check was conducted

for each TPB component in turn to determine whether the two datasets had a similar structure and could be combined in order to conduct reliability analysis [22]: (1) Combining the two datasets (MMR and dTaP/IPV), the raw scores for items in the TPB component were subjected to principal components analysis (PCA) with a forced single-factor solution; Table 3 shows that for each TPB component, the correlation between the two sets of loadings was high (close to 1) and the constant was not significantly different from zero. This indicated that even though the absolute values in the dTaP/IPV and MMR groups might differ, the interrelationship between items was similar. Thus, reliability statistics could be examined within the combined dataset.