Table 3. Bivariate analyses of household characteristics and the occurrence of a violent death in the previous year.
Dependent Variable | Sample size of dataset used | Violent death (nā=ā103) Odds ratio (95% CI) | P-value* |
Sex of head of household | 12 247 | ||
Male headed household | 1.0 | ||
Female headed household | 1.81 (1.14, 2.87) | 0.01 | |
Age of head | 12 247 | 1.02 (1.01, 1.03) | 0.01 |
Race of head of household | 10 320 | ||
Black/African | 1.0 | ||
Coloured | 0.40 (0.16, 0.97) | 0.04 | |
White/Asian** | 0.07 (0.01, 0.48) | 0.01 | |
Asset index quintile | 12 017 | ||
Poorest quintile | 1.0 | ||
2nd poorest quintile | 0.99 (0.53, 1.86) | 0.99 | |
Middle quintile | 0.57 (0.27, 1.19) | 0.14 | |
2nd richest quintile | 1.72 (0.93, 3.18) | 0.09 | |
Richest quintile | 0.22 (0.07, 0.70) | 0.01 | |
Employment of head | 12 030 | ||
No work for payment in the past 7 days | 1.0 | ||
Worked for payment in the past 7 days | 0.36 (0.20, 0.63) | <0.001 | |
Years of education of head | 12 247 | 0.95 (0.91, 0.99) | 0.01 |
P-value for Pearson's chi-square test of homogeneity
Owing to the small sample size White and Indian/Asian population groups were combined for analysis