Objective  To evaluate the reliability and validity of the Problematic Internet Pornography Use Scale-Chinese Version (PPUS-C) and to provide evidences for application of the scale among Chinese college students.

  Methods  Eight psychology researchers translated the original PPUS into Chinese first and then one English professional translated the PPUS-C back into English. Several minor linguistic modifications were performed for several words and expressions in original PPUS. We conducted a survey among a total of 1 026 students randomly selected in 7 universities in Fujian, Shandong, Liaoning, Guangdong, and Shaanxi province in November 2016. PPUS-C, the Attitudes Towards Sexuality Scale (ATSS), Self Rated Health Measurement Scale (SRHMS), and Online Sexual Activities (OSA) were used in the survey. Internal consistency coefficient, test-retest reliability, construct validity, and criterion validity were adopted to evaluate the reliability and validity of the PPUS-C.

  Results  Item analysis demonstrated that the correlation coefficients between scores of 12 items and their factor scores ranged from 0.83 to 0.93 for the PPUS-C (all P < 0.001), indicating a good item discrimination of the scale. Exploratory factor analysis revealed a cumulative variance contribution rate of 81.85% for the 4 common factors of the PPUS-C. The factor structures for all the items of the PPUS-C were generally consistent with those of the original PPUS, except for the 6th item. Confirmatory factor analyses on the structure of original PPUS produced a Tucker-Lewis index (TLI）of 0.959, comparative fit index (CFI）of 0.967, and a root mean square error of approximate (RMSEA）of 0.046; the exploratory factor analyses on the structure of PPUS-C revealed a TLI of 0.957, CFI of 0.966, and RMSEA of 0.046; the results suggest a better fitting of the structure of original PPUS so the reliability and validity evaluation were performed according to the structure original PPUS. For the PPUS-C, the overall Cronbach′s α coefficient of was 0.95; the test-retest reliability coefficient was 0.89; the factor Cronbach′s α of coefficient and test-retest coefficient were 0.81 and 0.61 for distress and functional problems, 0.88 and 0.70 for excessive use, 0.91 and 0.66 for control difficulty, and 0.88 and 0.62 for use for escaping/avoiding negative emotion, respectively, suggesting a good reliability of the PPUS-C. For all the participants, the average overall score was 9.36 ± 9.48 for PPUS-C, 38.00 ± 5.97 for ATSS, 99.99 ± 15.62 for mental health, 92.08 ± 17.91 for social health, and 22.14 ± 12.54 for OSA; the overall score of PPUS-C was significantly associated with the scores of the ATSS and OSA, with the correlation coefficients of 0.13 and 0.57 (both P < 0.001) and negatively associated with the scores of mental health and social health (r = – 0.15 and r = – 0.11, both P < 0.001); the results indicated a good criterion validity of the PPUS-C. Compared to the female participants, the males possessed significantly higher overall score of PPUS-C (11.13 ± 9.98 vs. 7.17 ± 7.91) and PPUS-C factor scores of distress and functional problems (2.73 ± 2.74 vs. 1.99 ±2.30), excessive use (3.17 ± 2.82 vs. 2.07 ± 2.23), control difficulty (2.66 ± 2.93 vs. 1.62 ± 2.28), and use for escaping/avoiding negative emotion (2.56 ± 2.77 vs. 1.48 ± 2.02) (P < 0.001 for all). Multi-group invariance test demonstrated that the PPUS-C is of equivalent significance and latent structure when applied in both male and female participants.

  Conclusion  The compiled Problematic Internet Pornography Use Scale-Chinese Version is of good reliability and validity and applicable among college students with Chinese cultural background.