Table 3 Multilevel binary logistic regression results on the predictors of premarital sexual intercourse among young women in sub-Saharan Africa
From: Prevalence and predictors of premarital sexual intercourse among young women in sub-Saharan Africa
Variables | Null model | Model I aOR (95% CI) | Model II aOR (95% CI) | Model III aOR (95% CI) |
|---|---|---|---|---|
Age | ||||
 15–19 |  | Reference (1.0) |  | Reference (1.0) |
 20–24 |  | 4.39*** (4.25–4.55) |  | 4.49*** (4.34–4.65) |
Level of education | ||||
 No formal education |  | Reference (1.0) |  | Reference (1.0) |
 Primary |  | 1.09** (1.03–1.15) |  | 1.34*** (1.26–1.42) |
 Secondary/higher |  | 1.49*** (1.41–1.58) |  | 1.63*** (1.54–1.72) |
Religion | ||||
 Christianity |  | 1.49*** (1.26–1.75) |  | 1.74*** (1.47–2.06) |
 Islam |  | 0.67*** (0.56–0.79) |  | 0.66*** (0.56–0.78) |
 Traditional |  | Reference (1.0) |  | Reference (1.0) |
 No religion |  | 2.44*** (1.99–2.98) |  | 2.44*** (1.99–3.00) |
Employment status | ||||
 Not working |  | 0.77*** (0.75–0.80) |  | 0.75*** (0.73–0.78) |
 Working |  | Reference (1.0) |  | Reference (1.0) |
Wealth index | ||||
 Poorest |  | Reference (1.0) |  | Reference (1.0) |
 Poorer |  | 0.99 (0.94–1.04) |  | 0.99 (0.94–1.05) |
 Middle |  | 0.96 (0.91–1.01) |  | 0.94* (0.89–0.99) |
 Richer |  | 0.83*** (0.78–0.87) |  | 0.77*** (0.73–0.82) |
 Richest |  | 0.59*** (056–0.62) |  | 0.55*** (0.52–0.58) |
Frequency of reading newspaper or magazine | ||||
 Not at all |  | 1.06 (0.89–1.25) |  | 0.98 (0.83–1.17) |
 Less than once a week |  | 0.89 (0.75–1.05) |  | 086 (0.72–1.02) |
 At least once a week |  | 0.98 (0.83–1.17) |  | 0.87 (0.73–1.04) |
 Almost every day |  | Reference (1.0) |  | Reference (1.0) |
Frequency of listening to radio | ||||
 Not at all |  | 0.82*** (0.74–0.91) |  | 0.90* (0.81–0.99) |
 Less than once a week |  | 0.73*** (0.66–0.81) |  | 0.80*** (0.72–0.89) |
 At least once a week |  | 0.69*** (0.62–0.76) |  | 0.79*** (0.71–0.88) |
 Almost every day |  | Reference (1.0) |  | Reference (1.0) |
Frequency of watching television | ||||
 Not at all |  | 0.42*** (0.39–0.45) |  | 0.50*** (0.46–0.53) |
 Less than once a week |  | 0.51*** (0.47–0.55) |  | 0.56*** (0.52–0.60) |
 At least once a week |  | 0.55*** (0.51–0.59) |  | 0.56*** (0.52–0.60) |
 Almost every day |  | Reference (1.0) |  | Reference (1.0) |
Place of residence | ||||
 Urban |  |  | Reference (1.0) | Reference (1.0) |
 Rural |  |  | 0.74*** (0.72–0.76) | 0.73*** (0.70–0.76) |
Sub-region | ||||
 West Africa |  |  | 0.34*** (0.32–0.37) | 0.61*** (0.56–0.67) |
 East Africa |  |  | 0.28*** (0.26–0.31) | 0.32*** (0.29–0.35) |
 Central Africa |  |  | 0.65*** (0.60–0.71) | 0.78*** (0.71–0.86) |
 Southern Africa |  |  | Reference (1.0) | Reference (1.0) |
Random effect results | ||||
 PSU variance | 0.04 (0.04–0.05) | 0.04 (0.04–0.06) | 0.05 (0.04–0.06) | 0.05 (0.04–0.06) |
 ICC | 0.0126461 | 0.0132099 | 0.0141608 | 0.0137314 |
 LR test | 250.07 (X = 0.0000) | 214.16 (X = 0.0000) | 262.74 (X = 0.0000) | 203.97 (X = 0.0000) |
 Wald chi-square |  | 11,593.86*** | 2934.43*** | 13,307.89*** |
 Model fitness |  |  |  |  |
 Log-likelihood | − 58,809.755 | − 51,886.136 | − 57,288.897 | − 50,442.408 |
 AIC | 117,623.5 | 103,816.3 | 114,589.8 | 100,936.8 |
 N | 87,924 | 87,924 | 87,924 | 87,924 |
 Number of groups | 1567 | 1567 | 1567 | 1567 |