Skip to main content

View full-text article in PMC

. 2020 Nov 11;9(11):3625. doi: 10.3390/jcm9113625

Table 3.

Predictive models of pornography use and cybersex behaviors: stepwise logistic regression (n = 1500).

Criterion: Downloading Sexually Explicit Material. Fitting Indexes: H‒L (p-Value) = 0.193; N-R² = 0.155; AUC = 0.748 (95% CI: 0.709 to 0.787).
Predictor	Contrast	B	SE	p	OR	95%CI
Sex	Male vs. Female	1.554	0.211	<0.001	4.730	3.126	7.157
Sexual orientation				0.011
	Homosex.vs. Heterosex.	−0.774	0.761	0.309	0.461	0.104	2.050
	Ambisex. vs. Heterosex.	1.147	0.372	0.002	3.149	1.519	6.530
	Non-defined vs. Heterosex.	−0.293	0.560	0.601	0.746	0.249	2.235
Sexual interest/Internet beh.	Yes vs. No	0.916	0.191	<0.001	2.498	1.718	3.632
1st sexual experience at age…				0.006
	Linear trend	0.222	0.289	0.442	1.249	0.709	2.200
	Quadratic trend	−0.053	0.297	0.858	0.948	0.530	1.697
	Cubic trend	1.086	0.360	0.003	2.961	1.462	5.997
	Quartic trend	−0.561	0.347	0.106	0.571	0.289	1.126
Criterion: using social media to send sexual content. Fitting indexes: H‒L (p-value) = 0.755; N-R² = 0.181; AUC = 0.776 (95% CI: 0.728 to 0.825).
Predictor		B	SE	p	OR	95%CI
Sex	Male vs. Female	0.989	0.221	<0.001	2.690	1.744	4.149
Drugs use/abuse				0.022
	Linear trend	0.415	0.260	0.111	1.514	0.909	2.523
	Quadratic trend	0.025	0.320	0.936	1.026	0.548	1.920
	Cubic trend	0.603	0.368	0.101	1.827	0.889	3.755
Sexual interest/Internet behavior	Yes vs. No	1.705	0.210	<0.001	5.504	3.647	8.306
He/she has been abused	Yes vs. No	1.372	0.308	<0.001	3.943	2.156	7.210
Criterion: using social media to send self-sexual material. Fitting indexes: H‒L (p-value) = 0.554; N-R² = 0.190; AUC = 0.790 (95% CI: 0.740 to 0.841).
Predictor		B	SE	p	OR	95%CI
Sexual orientation				0.001
	Homosex. vs. Heterosex	0.842	0.560	0.133	2.320	0.774	6.960
	Ambisex. vs. Heterosex	1.289	0.360	<0.001	3.630	1.791	7.356
	Non-defined. vs. Heterosex.	0.750	0.464	0.106	2.116	0.853	5.252
Sexual interest/Internet behavior.	Yes vs. No	1.295	0.225	<0.001	3.650	2.349	5.669
1st sexual experience at age…				<0.001
	Linear trend	0.670	0.325	0.039	1.955	1.034	3.697
	Quadratic trend	−0.120	0.328	0.716	0.887	0.466	1.689
	Cubic trend	1.023	0.404	0.011	2.782	1.261	6.135
	Quartic trend	−0.714	0.374	0.056	0.490	0.235	1.019
He/she has been abused	Yes vs. No	1.021	0.310	0.001	2.776	1.512	5.098
Forced to share sex content	Yes vs. No	0.595	0.247	0.016	1.813	1.117	2.941
Criterion: participation in sexual chats. Fitting indexes: H‒L (p-value) = 0.878; N-R² = 0.045; AUC = 0.642 (95% CI: 0.582 to 0.702).
Predictor		B	SE	p	OR	95%CI
Sex	Male vs. Female	0.684	0.227	0.003	1.983	1.270	3.095
Sexual interest/Internet behavior.	Yes vs. No	0.588	0.224	0.009	1.801	1.161	2.795
Forced to share sex content	Yes vs. No	0.907	0.251	<0.001	2.477	1.515	4.047
Criterion: use of erotic telephone lines. Fitting indexes: H-L (p-value) = 0.744; N-R² = 0.083; AUC = 0.703 (95% CI: 0.646 to 0.761).
Predictor		B	SE	p	OR	95%CI
Sex	Male. vs. Female	0.730	0.231	0.002	2.074	1.319	3.263
Drugs use/abuse				0.019
	Linear trend	0.566	0.280	0.043	1.762	1.018	3.050
	Quadratic trend	−0.356	0.302	0.238	0.701	0.388	1.265
	Cubic trend	0.147	0.332	0.659	1.158	0.604	2.218
Age (years-old)		−0.234	0.101	0.021	0.791	0.649	0.965
Frequency sexual experiences				0.001
	Linear trend	−0.215	0.342	0.530	0.807	0.413	1.576
	Quadratic trend	−0.713	0.315	0.024	0.490	0.264	0.909
	Cubic trend	0.784	0.530	0.139	2.190	0.774	6.194
	Quartic trend	0.708	0.445	0.112	2.031	0.849	4.860

H-L: Hosmer‒Lemeshow. N-R²: Nagelkerke’s pseudo-R² coefficient. AUC: Area under the ROC curve. B: Logistic parameter. SE: Standard error. OR: odds ratio.