Table 3.
Estimates From Multilevel Regression Models Testing the Association Between Extrinsic Sexual Motives, Stress, and Depressive Symptoms
| Variable | Model 1 | Model 2 | Model 3 | Model 4 | Model 5 | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| b | 95% CI | b | 95% CI | b | 95% CI | b | 95% CI | b | 95% CI | |
| Extrinsic sexual motives | 0.54** | [0.14, 0.94] | 0.57** | [0.17, 0.97] | 0.94** | [0.35, 1.53] | 0.80** | [0.26, 1.35] | 1.14** | [0.45, 1.83] |
| Level of stress | 4.95*** | [4.46, 5.44] | 4.97*** | [4.48, 5.47] | 5.20*** | [4.47, 5.92] | 5.31*** | [4.56, 6.06] | 5.64*** | [4.72, 6.57] |
| Actor is women | 0.08 | [−0.55, 0.72] | 0.09 | [−0.55, 0.72] | 0.15 | [−0.48, 0.78] | −0.04 | [−0.68, 0.61] | −0.66 | [−1.74, 0.41] |
| Partner is women | −0.05 | [−0.70, 0.60] | −0.06 | [−0.71, 0.58] | −0.18 | [−0.84, 0.47] | −0.02 | [−0.67, 0.63] | −0.83 | [−1.82, 0.16] |
| Extrinsic sexual motives × Stress | −0.33 | [−0.87, 0.20] | ||||||||
| Actor women × Extrinsic sexual motives × Stress | −1.14* | [−2.25, −0.03] | ||||||||
| Partner women × Extrinsic sexual motives × Stress | −0.29 | [−1.40, 0.82] | ||||||||
| Actor women × Partner women × Extrinsic sexual motives × Stress | 0.13 | [−2.26, 2.52] | ||||||||
| Constant | 18.01*** | [17.23, 18.79] | 18.04*** | [17.25, 18.82] | 18.08*** | [17.30, 18.86] | 18.00*** | [17.22, 18.78] | 18.27*** | [17.46, 19.09] |
| Log likelihood | −2,365.37 | −2,364.63 | −2,360.33 | −2,362.58 | −2,356.65 |
Notes: b = unstandardized beta coefficient; CI = confidence interval. All models adjusted for age in years, race, education, years lived together, child in household, intrinsic sexual motives. Actor and partner gender were coded as 0 = men and 1 = women. All continuous variables were centered at their sample mean. Constituent interactions terms of the three-way and four-way interaction effects tested in the models are omitted from the table. In this model, actor effect refers to the association between respondent gender and depressive symptoms. Partner effect refers to the association between partner gender and respondent’s depressive symptoms (West et al., 2008). Extrinsic sexual motives are measures of respondent’s motives. We also performed additional analyses that included bisexuality as a covariate, and results remained the same.
*p < .05. **p < .01. ***p < .001.