Table 3.
Hierarchical Logistic Regression analysis of adolescents’ individual attributes and their social network variables predicting risky sexual behavior (N = 806)
| variables | Odds ratio (95% confidence interval) | |
|---|---|---|
| Model I | Model II | |
| Age of respondents | 1.04(.84 – 1.30) | 1.35(1.05 – 1.74)* |
| Sex of respondents | ||
| Male® | 1 | 1 |
| Female | 2.48(1.56 – 3.93)* | 2.50 (1.50 – 4.18)** |
| Residence | ||
| Rural® | 1 | 1 |
| Urban | .872(.55 – 1.38) | 1.42(.84 – 2.40) |
| Living arrangement | ||
| With both parents® | 1 | 1 |
| With single parents | 5.13(3.05 – 8.72)** | 5.00(2.82 – 8.87)** |
| With others | 1.39(.51 – 3.79) | 1.75(.62 – 4.95) |
| Drinking alcohol | ||
| Never ® | 1 | 1 |
| Sometimes | 3.37(1.45 – 3.87)** | 2.42(1.43 – 4.13)** |
| Quite often | 3.60(1.30 – 9.98)* | 3.68(1.27 – 10.70)* |
| Chewing khat | ||
| Never ® | 1 | 1 |
| Sometimes | 1.85(1.06 – 3.23)* | 1.88(1.04 – 3.40)* |
| Quite often | 2.48(.78 – 7.72) | 2.25(.66 – 6.59) |
| Network size | – | .84 (.72 – .97)* |
| Network members’ sexual norm | – | 1.61(1.04 – 2.50)* |
| Network tie strength | – | 1.12(1.06 – 1.19)** |
| Network homogeneity | – | 1.58(.98 – 2.55)* |
| Model G2 (−2log likelihood) | 316.478 | 228.702 |
| Degree of freedom | 9 | 13 |
| Changed Chi-square | 90.985** | 133.532** |
Note: *p < .05; **p < .01; ® = reference category