Table 5.
Absolute frequencies, outcome prevalences, exposure prevalences, crude and pooled prevalence ratio (PR) estimates, and relative confounding for the analysis of the original data using mother in a paid job as the outcome, father living with the family as the risk factor and social class as confounder (situation 3 original).
| First stratum: High social class | |||||
| Mother employed | No | All | |||
| N | Prev. | N | N | Prev. exp. = 15.3% | |
| Father Present | 66 | 79.5% | 17 | 83 | PR = 1.46 |
| No | 250 | 54.5% | 209 | 459 | M-H weight = 38.28 |
| All | 316 | 58.3% | 226 | 542 | |
| Second stratum: Low social class | |||||
| Mother employed | No | All | |||
| N | Prev. | N | N | Prev. exp. = 24.0% | |
| Father present | 112 | 70.9% | 46 | 158 | PR = 1.88 |
| No | 189 | 37.8% | 311 | 500 | M-H weight = 45.38 |
| All | 301 | 45.7% | 357 | 658 | |
| Combined strata: High and low social class | |||||
| Mother employed | No | All | Prev. exp. = 20.1% | ||
| N | Prev. | N | N | PR (crude) = 1.61 | |
| Father present | 178 | 73.9% | 63 | 241 | PR (M-H) = 1.69 |
| No | 439 | 45.8% | 520 | 959 | Confounding = 4.4% |
| All | 617 | 51.4% | 583 | 1200 | P-value(het) *= 0.01 |
* P-value for testing heterogeneity of the prevalence ratios across strata.