Skip to main content

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).

From: Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio

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

  1. * P-value for testing heterogeneity of the prevalence ratios across strata.