Abstract
Background and aims
Gender nonconformity (GNC), which refers to an individual's expression of gender that does not align with the socially prescribed norms for their biological sex, may be associated with adverse behavioral problems, such as problematic smartphone use (PSU) and problematic internet use (PIU). This study examined the associations between GNC and PSU and GNC and PIU among Chinese adolescents.
Methods
This cross-sectional study utilized data from the 2021 School-based Chinese Adolescents Health Survey, recruiting 23,054 eligible adolescents aged 11 to 21, with an average age of 14.9 (SD: 1.7) years from 504 classes in 84 public high schools across 7 cities in China. Gender nonconformity, PSU/PIU, and demographics were measured. Mixed-effect linear regression models were performed.
Results
Among the participants (51.0% male), 5.3% reported high GNC and 26.9% reported moderate GNC. After adjusting for covariates, high GNC was significantly and positively associated with PSU (Β = 1.11, 95% CI = 0.49–1.72) and PIU severity (Β = 2.16, 95% CI = 1.40–2.93). Stratified analyses indicated that the associations between GNC and PSU differed between males and females, with a significant association observed only among male students (Β = 1.91, 95% CI = 0.97–2.86).
Discussion and conclusions
GNC is positively associated with the severity of PSU and PIU among Chinese adolescents, with male gender-nonconforming adolescents being more vulnerable to PSU. These results highlight the importance of implementing education on gender expression diversity in schools to create an inclusive school environment, which may potentially help prevent PSU and PIU among gender-nonconforming adolescents.
Introduction
The use of smartphones and the internet has become increasingly popular among adolescents worldwide, with widespread access to digital devices and high-speed internet (Ricoy, Martínez-Carrera, & Martínez-Carrera, 2022). In China alone, there were over 191 million adolescent internet users in 2021, with the majority accessing the web through their personal devices (China Internet Network Information Center, 2022). While these technologies offer numerous benefits, excessive use and dependency among young people are growing public concerns (Meng et al., 2022; Pan, Chiu, & Lin, 2020). Two forms of problematic digital use that are common among adolescents are problematic smartphone use (PSU) and problematic internet use (PIU) (Christakis, 2019). Evidence has shown that adolescents who engage in PSU and PIU are at increased risk of physical discomfort (e.g., back, neck, and wrist pain), lower academic performance, and mental health issues (e.g., anxiety, depression, and suicidal behavior) (Shinetsetseg, Jung, Park, Park, & Jang, 2022; van den Eijnden, Koning, Doornwaard, van Gurp, & TerBogt, 2018). Given the high occurrence and harmful effects of PSU and PIU among adolescents, it is crucial to identify those who are most vulnerable and more likely to engage in high levels of problematic digital use.
Gender, in this context, refers to the cultural roles and expectations assigned to individuals based on their biological sex. It is an organizing feature in most societies that shapes norms of socialization (Bewley, McCartney, Meads, & Rogers, 2021). Gender expression, on the other hand, pertains to how individuals display their gender through their physical appearance or behavior (American Psychological Association, 2023). Gender nonconformity (GNC), also known as non-conforming gender expression, occurs when an individual's gender expression deviates from the socially accepted norms associated with their biological sex (Lowry, Johns, & Robin, 2020). It is important to note that gender expression represents the external presentation of one's gender and is distinct from gender identity, which refers to an individual's internal sense of their gender (Loso et al., 2023).
A recent cross-sectional study conducted in Shanghai reported that approximately 30% of Chinese adolescents exhibit moderate or high levels of GNC (Lian & Lou et al., 2022), indicating that the presence of this phenomenon among Chinese adolescents cannot be neglected. Previous evidence has shown that adolescents with GNC expressions usually experience more bullying, victimization, and childhood abuse than their gender-conforming peers (Brown & Stone, 2016; Feijóo & Rodríguez-Fernández, 2021). Moreover, these gender non-conforming adolescents are at an increased risk of developing mental health problems, including depressive and anxiety symptoms (Lin et al., 2021). According to the minority stress theory and the compensatory internet use theory (Kardefelt-Winther, 2014; Meyer, 2003), adolescents with GNC tend to adopt avoidant and inflexible coping styles, such as PSU and PIU, when faced with the aforementioned negative experiences and mental health issues (Brand, 2022; Elhai, Levine, & Hall, 2019).
Despite the growing evidence indicating that adolescents with GNC expressions are more likely to engage in PSU and PIU (Huang, Chang, Lu, & Yen, 2022; Lowry et al., 2018), there is limited research exploring the relationship between GNC and PSU/PIU among school-based Chinese adolescents (Huang et al., 2022). Additionally, previous studies examining these associations have not adequately accounted for important covariates, such as adolescents' depressive and anxiety symptoms, social support status, and school bullying experiences (Long et al., 2016). Furthermore, it remains unclear whether these findings hold true across different biological sexes within the population. Therefore, the present study examined the associations between both GNC and PSU and GNC and PIU among Chinese adolescents.
Methods
Participants and procedure
This cross-sectional study utilized data from the 2021 School-based Chinese Adolescents Health Survey (SCAHS), an ongoing survey of health-related behaviors among Chinese adolescents in grades 7 to 12. The detailed study design and sampling methods have been previously reported (Lai et al., 2023). In brief, a multi-stage, random cluster sampling approach was used to select participants. Initially, a convenience sample of 7 cities was selected with the assistance of local education bureaus. Subsequently, the number of junior, senior, and vocational high schools was randomly determined based on the proportion of different school types. In each selected school, two classes were randomly chosen from each grade, and all students in these classes were invited to participate in the study, except for those who suffered from severe mental or physical disorders as identified by the head teacher and/or health care physicians. The survey was conducted by trained investigators who were available to address any queries or confusion the participants might have had regarding the structured questionnaire. Participants completed the survey within a single session at school after being informed of the study's objectives and procedures. All participants were expected to independently complete the anonymous questionnaire within 45 min, and the investigators reviewed the completeness of the questionnaires before the participants left the site. Of the total 23,409 students from 504 classes in 84 public high schools (42 junior high schools, 28 senior high schools, and 14 vocational high schools) who were invited to participate, 23,054 completed and submitted the questionnaire, yielding a response rate of 98.5%. Among the included 23,054 adolescents (age range: 11–21 years; mean [SD] age: 14.9 [1.7] years), 51.0% were males.
Measures
PSU and PIU
PSU severity was assessed using the 10-item smartphone addiction scale-short version (SAS-SV) (Kwon, Kim, Cho, & Yang, 2013), and PIU severity was measured using Young's 20-item internet addiction test (IAT) (Young, 1998). SAS-SV and IAT are widely used instruments to assess PSU and PIU with good reliability and validity in the Chinese population (Guo et al., 2020; Luk et al., 2018). Further details are available in the Methods in the Supplementary material.
Gender nonconformity
To assess GNC or gender non-conforming gender expression, we employed a validated measure for gender expression that has been utilized in previous studies (Greytak, 2016; Wylie, Corliss, Boulanger, Prokop, & Austin, 2010), including research conducted among Chinese adolescents (Zhao et al., 2021). Participants were presented with the following statement and question: “A person's appearance, style, dress, or the way they walk or talk may affect how people describe them. How do you think people at school would describe you?” The response options were “very feminine,” “mostly feminine,” “somewhat feminine,” “equally feminine and masculine,” “somewhat masculine,” “mostly masculine,” and “very masculine.” By considering a student's response to this question and their biological sex, a 7-point GNC scale was established (Lowry et al., 2018). The scale ranged from 1, indicating very feminine female students and very masculine male students (representing the most gender conforming individuals), to 7, representing very masculine female students and very feminine male students (representing the most gender non-conforming individuals) (Table S2 in the Supplementary material). However, due to the limited number of mostly or very masculine female students, and the limited number of mostly or very feminine male students, we created a 3-level GNC variable based on previous studies (Lowry et al., 2018, 2020): (1) high GNC (somewhat, mostly, and very masculine female students and somewhat, mostly, and very feminine male students), (2) moderate GNC (equally feminine and masculine female and male students), and (3) low GNC (somewhat, mostly, and very feminine female students and somewhat, mostly, and very masculine male students).
Covariates
Several covariates were also measured, including demographic characteristics (sex, age, household socioeconomic status, family relationship, classmate relationship, teacher-classmate relationship, and academic performance), sexual orientation (heterosexual, lesbian/gay, bisexual, and unsure), depressive symptoms, anxiety symptoms, social support, and being bullied at school the students were attending. Further details on these covariables are available in the Methods and Table S1 in the Supplementary material.
Statistical analysis
Data were described as means (standard deviations, SDs) for continuous variables and as frequencies with percentages for categorical variables. Chi-square and analysis of variance tests were performed to test associations or differences among various GNC and PSU/PIU groups. Univariable mixed-effect linear regression models were employed to estimate the potential associations between GNC and PSU/PIU, with the school unit as the random-effect unit. Unstandardized regression coefficients with 95% confidence intervals (CIs) were reported. Additionally, three adjusted models were developed, each including a different set of covariates. Model 1 included age, sex, household socioeconomic status, family relationship, classmate relationship, teacher-classmate relationship, academic performance, and sexual orientation. Model 2 included the variables in Model 1 plus depressive symptoms scores and anxiety symptoms scores. Model 3 included the variables in Model 2 and additionally added social support and being bullied at school. Biological sex was also tested as an effect modifier in Models 1 through 3, using a product interaction term (biological sex × gender expression). Subgroup analyses were conducted by sex (males and females) to examine potential variations in different subgroups.
To test the robustness of the findings, two sensitivity analyses were also performed: (1) all analyses were repeated using imputed data sets generated through the multiple imputation method by chained equations to address missing data. Ten imputed data sets were created (Lin et al., 2022; Lu, 2017; Y. Zhang et al., 2021); (2) the regression analyses examining the associations between GNC and PSU/PIU were repeated using a binary GNC variable called gender conformity (GC) (GNC score: 1–3) and GNC (GNC score: 4–7) (Lian & Li et al., 2022).
All analyses were performed using Stata, version 17 (StataCrop), from December 2022 to January 2023. All statistical tests were two-sided, and a significance level of p ≤ 0.05 was considered statistically significant.
Ethics
The study procedures were carried out in accordance with the Declaration of Helsinki. This study received ethical approval from the Sun Yat-Sen University School of Public Health Institutional Review Board. Written informed consent explaining the study purposes, processes, benefits, and risks was obtained from each participant and one of their legal guardians.
Results
Characteristics of the study population
Table 1 presents the characteristics of the study population stratified by GNC, while Table S3 presents the characteristics stratified by PSU and PIU. Among all participants, 26.9% reported moderate GNC, while 5.3% reported high GNC. Additionally, 39.8% of students were identified as problematic smartphone users, 32.5% were categorized as internet at-risk users, and 652 (2.8%) were identified as problematic internet users. High GNC was found to be more prevalent among female students (6.0%) than male students (4.7%). Moreover, high GNC was more common among lesbian/gay students (25.3%), bisexual students (9.2%), and unsure students (6.4%) compared to heterosexual students (4.2%). High GNC was also more common among adolescents with PSU (6.4%) and those with PIU (problematic internet users: 12.4%; internet at-risk users: 6.8%) than their counterparts.
Characteristics of the study population among different gender expression groups
| Characteristic | Total | Gender Nonconformity, (N = 23,054) | χ2/F | p valuee | ||
| Lowa | Moderateb | Highc | ||||
| Total population, n (%) | 23,054 (100.0) | 15,618 (67.7) | 6,213 (26.9) | 1,223 (5.3) | NA | |
| Age, mean (SD) | 14.9 (1.7) | 14.9 (1.7) | 14.7 (1.7) | 14.6 (1.7) | 81.5 | <0.001 |
| Sex, n (%) | 427.5 | |||||
| Male | 11,761 (51.0) | 8,695 (73.9) | 2,519 (21.4) | 547 (4.7) | <0.001 | |
| Female | 11,293 (49.0) | 6,923 (61.3) | 3,694 (32.7) | 676 (6.0) | ||
| Household socioeconomic statusd, n (%) | ||||||
| High | 7,305 (31.7) | 5,070 (69.4) | 1,848 (25.3) | 387 (5.3) | 47.1 | <0.001 |
| Middle | 13,718 (59.6) | 9,241 (67.4) | 3,803 (27.7) | 674 (4.9) | ||
| Low | 1,990 (8.6) | 1,284 (64.5) | 548 (27.5) | 158 (7.9) | ||
| Family relationshipd, n (%) | ||||||
| Good | 19,121 (83.1) | 13,328 (70.0) | 4,910 (25.7) | 883 (4.6) | 256.2 | <0.001 |
| Average | 3,053 (13.3) | 1,795 (58.8) | 1,022 (33.5) | 236 (7.7) | ||
| Poor | 825 (3.6) | 462 (56.0) | 262 (31.8) | 101 (12.2) | ||
| Classmate relationshipd, n (%) | ||||||
| Good | 18,351 (80.0) | 12,915 (70.4) | 4,577 (24.9) | 859 (4.7) | 348.8 | <0.001 |
| Average | 4,258 (18.6) | 2,482 (58.3) | 1,475 (34.6) | 301 (7.1) | ||
| Poor | 335 (1.5) | 155 (46.3) | 125 (37.3) | 55 (16.4) | ||
| Teacher-classmate relationshipd, n (%) | ||||||
| Good | 16,017 (70.0) | 11,233 (70.1) | 4,032 (25.2) | 752 (4.7) | 179.5 | <0.001 |
| Average | 6,528 (28.5) | 4,076 (62.4) | 2,039 (31.2) | 413 (6.3) | ||
| Poor | 345 (1.5) | 206 (59.7) | 91 (26.4) | 48 (13.9) | ||
| Academic performanced, n (%) | ||||||
| Above average | 8,707 (37.9) | 6,059 (69.6) | 2,230 (25.6) | 418 (4.8) | 54.5 | <0.001 |
| Average | 7,687 (33.4) | 5,225 (68.0) | 2,101 (27.3) | 361 (4.7) | ||
| Below average | 6,590 (28.7) | 4,287 (65.1) | 1,864 (28.3) | 439 (6.7) | ||
| Sexual orientationd, n (%) | ||||||
| Heterosexual | 16,091 (69.8) | 11,919 (74.1) | 3,495 (21.7) | 677 (4.2) | 1,091.9 | <0.001 |
| Lesbian/Gay | 379 (1.6) | 137 (36.1) | 146 (38.5) | 96 (25.3) | ||
| Bisexual | 2,040 (8.8) | 1,011 (49.6) | 841 (41.2) | 188 (9.2) | ||
| Unsure | 1,292 (5.6) | 687 (53.2) | 522 (40.4) | 83 (6.4) | ||
| Being bullied at schoold, n (%) | ||||||
| No | 20,222 (87.8) | 13,912 (68.8) | 5,365 (26.5) | 945 (4.7) | 162.8 | <0.001 |
| Yes | 2,815 (12.2) | 1,693 (60.1) | 845 (30.0) | 277 (9.8) | ||
| Social support score, mean (SD) | 67.9 (15.0) | 69.1 (14.4) | 65.9 (15.6) | 62.8 (17.3) | 181.1 | <0.001 |
| Depressive symptoms scored, mean (SD) | 15.2 (7.9) | 14.9 (7.5) | 15.6 (8.2) | 17.7 (9.6) | 84.2 | <0.001 |
| Anxiety symptoms score, mean (SD) | 3.8 (4.4) | 3.5 (4.1) | 4.1 (4.6) | 5.3 (5.4) | 80.6 | <0.001 |
| PSU, n (%) | ||||||
| Non-PSU | 13,884 (60.2) | 9,510 (68.5) | 3,740 (26.9) | 634 (4.6) | 38.8 | <0.001 |
| PSU | 9,170 (39.8) | 6,108 (66.6) | 2,473 (27.0) | 589 (6.4) | ||
| PIU, n (%) | ||||||
| Non-PIU | 14,908 (64.7) | 10,251 (68.8) | 4,022 (27.0) | 635 (4.3) | 139.0 | <0.001 |
| At-risk users | 7,494 (32.5) | 4,991 (66.6) | 1,996 (26.6) | 507 (6.8) | ||
| PIU | 652 (2.8) | 376 (57.7) | 195 (30.0) | 81 (12.4) | ||
Abbreviations: PSU, Problematic Smartphone Use; PIU, Problematic Internet Use; SAS-SV, Smartphone Addiction Scale Short Version; IAT, Internet Addiction Test.
a Female students who describe themselves as very/mostly/somewhat feminine; male students who describe themselves as very/mostly/somewhat masculine.
b Students who describe themselves as equally feminine and masculine.
c Female students who describe themselves as very/mostly/somewhat masculine; male students who describe themselves as very/mostly/somewhat feminine.
d Missing data: 41 for household socioeconomic status, 55 for family relationship, 110 for classmate relationship, 164 for teacher-classmate relationship, 70 for academic performance, 3,252 for sexual orientation (2,905 chose “unwilling to answer” and 347 chose “neither men nor women”), 17 for being bullied, 55 for depressive symptoms scores.
e p value was based on the Chi-square test for categorical variables and analysis of variance for continuous variables as appropriate.
Associations of GNC with PSU and PIU
The unadjusted mixed-effect linear regression models demonstrated that moderate and high GNC was associated with the severity of PSU (moderate GNC: Β = 0.71, 95% CI = 0.41–1.01; high GNC: Β = 2.84, 95% CI = 2.25–3.43) and PIU (moderate GNC: Β = 0.93, 95% CI = 0.54–1.32; high GNC: Β = 5.50, 95% CI = 4.73–6.28) (Table S4 in the Supplementary material). As shown in Table 2, after adjusting for covariates in Model 3, students with high GNC were more likely to have a higher level of PSU/PIU severity (PSU: Β = 1.11, 95% CI = 0.49–1.72; PIU: Β = 2.16, 95% CI = 1.40–2.93), compared with those with low GNC.
Associations of gender nonconformity with PSU and PIUa
| PSU | PIU | |||||
| Β (95% CI) | Β (95% CI) | |||||
| Model 1 | Model2 | Model3 | Model 1 | Model2 | Model3 | |
| Gender nonconformity | ||||||
| Low | Ref | Ref | Ref | Ref | Ref | Ref |
| Moderate | 0.26 (−0.71, 0.59) | 0.11 (−0.21, 0.43) | 0.04 (−0.27, 0.36) | 0.20 (−0.24, 0.63) | −0.07 (−0.41, 0.39) | −0.16 (−0.55, 0.24) |
| High | 1.82 (1.04, 2.32)b | 1.27 (0.65, 1.89)b | 1.11 (0.49, 1.72)b | 3.45 (2.61, 4.28)b | 2.46 (1.69, 3.23)b | 2.16 (1.40, 2.93)b |
| R2 | 0.10 | 0.18 | 0.19 | 0.10 | 0.26 | 0.27 |
Abbreviations: Β, unstandardized regression coefficient; 95% CI, 95% confidence interval; Ref, Reference; R2: coefficients of determination, R square.
a PSU was measured by the SAS-SV scores, ranging from 10 to 60, with higher scores indicating the severity of PSU; PIU was measured by the IAT scores, ranging from 20 to 100, with higher scores indicating the severity of PIU.
Model 1 adjusted for sex, age, household socioeconomic status, family relationship, classmate relationship, teacher-classmate relationship, academic performance, and sexual orientation.
Model 2 adjusted for the covariates in Model 1, plus depressive symptoms scores and anxiety symptoms scores.
Model 3 adjusted for the covariates in Model 2, plus social support scores and being bullied.
b p < 0.001.
Associations of GNC with PSU/PIU stratified by sex
The associations between interaction items (biological sex × gender expression) and PSU/PIU were found to be statistically significant (Table S5 in the Supplementary material). Therefore, separate sex-stratified analyses were conducted for male and female adolescents. Among male students, a significant association was observed between high GNC and PSU severity (Fig. 1). Male students with high GNC reported a significantly higher level of PSU (Β = 1.91, 95% CI = 0.97–2.86 in Model 3) than those with low GNC (Table 3). Table 4 demonstrated that the associations between high GNC and PIU were significant for both female students (Β = 2.03, 95% CI = 1.00–3.07 in Model 3) and male students (Β = 2.33, 95% CI = 1.19–3.50 in Model 3).
Sex and gender expression interaction with the severity of problematic smartphone use (PSU).
Note: The PSU severity was measured by the SAS-SV scale. Based on the mean value of SAS-SV scores in this population, the ordinate starts at 28. Highly gender conforming refers to male adolescents who reported very, mostly, and somewhat masculine or female adolescents who reported very, mostly, and somewhat feminine. Gender nonconforming refers to male adolescents who reported very, mostly, and somewhat feminine or female adolescents who reported very, mostly, and somewhat masculine. The mixed-effect linear regression model was adjusted for age, household socioeconomic status, family relationship, classmate relationship, teacher-classmate relationship, academic performance, and sexual orientation
Citation: Journal of Behavioral Addictions 12, 3; 10.1556/2006.2023.00040
Associations of gender nonconformity with PSUa stratified by sex
| Male (n = 9,837) | Female (n = 9,704) | |||||
| Β (95% CI) | Β (95% CI) | |||||
| Model 1 | Model2 | Model3 | Model 1 | Model2 | Model3 | |
| Gender nonconformity | ||||||
| Low | Ref | Ref | Ref | Ref | Ref | Ref |
| Moderate | 0.75 (0.23, 1.26)b | 0.39 (−0.11, 0.89) | 0.32 (−0.18, 0.81) | −0.17 (−0.60, 0.26) | −0.15 (−0.56, 0.26) | −0.21 (−0.62, 0.20) |
| High | 2.81 (1.82, 3.80)c | 2.10 (1.15, 3.05)c | 1.91 (0.97, 2.86)c | 0.82 (−0.18, 1.67) | 0.41 (−0.40, 1.21) | 0.28 (−0.53, 1.09) |
| R2 | 0.10 | 0.17 | 0.18 | 0.10 | 0.19 | 0.19 |
Abbreviations: Β, unstandardized regression coefficient; 95% CI, 95% confidence interval; Ref, Reference; R2: coefficients of determination, R square.
a PSU was measured by the SAS-SV scores, ranging from 10 to 60, with higher scores indicating the severity of PSU.
Model 1 adjusted for age, household socioeconomic status, family relationship, classmate relationship, teacher-classmate relationship, academic performance, and sexual orientation.
Model 2 adjusted for the covariates in Model 1, plus depressive symptoms scores and anxiety symptoms scores.
Model 3 adjusted for the covariates in Model 2, plus social support scores and being bullied.
b p < 0.05.
c p < 0.001.
Associations of gender nonconformity with PIUa stratified by sex
| Male (n = 9,837) | Female (n = 9,704) | |||||
| Β (95% CI) | Β (95% CI) | |||||
| Model 1 | Model2 | Model3 | Model 1 | Model2 | Model3 | |
| Gender nonconformity | ||||||
| Low | Ref | Ref | Ref | Ref | Ref | Ref |
| Moderate | 0.76 (0.10, 1.42)c | 0.18 (−0.42, 0.79) | 0.44 (−0.56, 0.64) | −0.35 (−0.93, 0.22) | −0.27 (−0.80, 0.25) | 0.41 (−0.94, 0.12) |
| High | 3.88 (2.63, 5.13)b | 2.61 (1.45, 3.77)b | 2.33 (1.19, 3.50)b | 3.09 (1.97, 4.22)b | 2.33 (1.29, 3.37)b | 2.03 (1.00, 3.07)b |
| R2 | 0.10 | 0.25 | 0.27 | 0.11 | 0.26 | 0.27 |
Abbreviations: Β, unstandardized regression coefficient; 95% CI, 95% confidence interval; Ref, Reference; R2: coefficients of determination, R square.
a PIU was measured by the IAT scores, ranging from 20 to 100, with higher scores indicating the severity of PIU.
Model 1 adjusted for age, household socioeconomic status, family relationship, classmate relationship, teacher-classmate relationship, academic performance, and sexual orientation.
Model 2 adjusted for the covariates in Model 1, plus depressive symptoms scores and anxiety symptoms scores.
Model 3 adjusted for the covariates in Model 2, plus social support scores and being bullied.
b p < 0.001
c p < 0.05
Sensitivity analyses
The results from the imputed data analyses yielded similar findings, further supporting the positive association between high GNC and the severity of PSU and PIU (Table S6 in the Supplementary material). Additionally, significant associations were observed between the interaction items (biological sex × gender expression) and PSU/PIU (Table S7 in the Supplementary material). Furthermore, high GNC was associated with elevated PSU and PIU levels in both male and female students (Tables S8 and S9 in the Supplementary material). In the sensitivity analysis using the binary variable of GNC, the associations of GNC with PSU and PIU remained significant in Model 1 and Model 2, except when adjusting for social support scores and experiences of being bullied in Model 3, where the associations were no longer significant (Table S10 in the Supplementary material).
Discussion
This study represents one of the initial investigations into the associations between GNC and the severity of PSU/PIU in a sizable school-based sample of adolescent students in mainland China. Among the participants, 39.8% of adolescents reported experiencing PSU, while 35.3% reported being either internet at-risk users (32.5%) or problematic internet users (2.8%). These occurrence rates of PSU and PIU were consistent with a recent report on Chinese adolescents (M. X. Zhang, Chen, Tong, Yu, & Wu, 2021), but higher than the pooled rate (21.6%) from a meta-analysis covering adolescents from 64 countries (Meng et al., 2022). These findings indicate that excessive smartphone and internet use is a prevalent issue among Chinese adolescents, highlighting the need for effective interventions. Moreover, among the participants in this study, 26.9% of adolescents reported moderate levels of GNC, while 5.3% reported high levels. These findings are consistent with a recent study conducted in Shanghai, which suggested that 3 out of 10 adolescents experience moderate or high levels of GNC (Lian & Lou et al., 2022). These results highlight the importance of paying attention to GNC among Chinese adolescents. Furthermore, consistent with previous research (Becker, Ravens-Sieberer, Ottová-Jordan, & Schulte-Markwort, 2017; Kawabe, Horiuchi, Ochi, Oka, & Ueno, 2016; Méndez, Jorquera, & Ruiz-Esteban, 2020), the presence of GNC/PSU/PIU varied based on demographic characteristics (e.g., sex, age, household socioeconomic status, and academic performance), sexual orientation (heterosexual, lesbian/gay, bisexual, and unsure), depressive symptoms, anxiety symptoms, social support, and experiences of bullying at school. Therefore, it is important to consider these variables as covariates when examining the association between GNC and PSU/PIU. To the best of our knowledge, this study is the first to comprehensively explore the associations between GNC and PSU/PIU among school-based Chinese adolescents while considering a series of covariates.
Previous research has suggested that mental health problems, such as depressive symptoms and anxiety symptoms (Arrivillaga, Rey, & Extremera, 2022; Kósa et al., 2022; Ueno, Ito, Murai, & Fujiwara, 2020), as well as adverse social environments (e.g., low social support status and being bullied at school) (Gong et al., 2022; Liu, Liu, & Yuan, 2021; Vessey, Difazio, Neil, & Dorste, 2022), may contribute to the occurrence of PSU and PIU among adolescents. A recent study conducted among young sexual minorities demonstrated a positive association between GNC and PSU, even after adjusting for internalized sexual stigma and sexual orientation (Huang et al., 2022). Similarly, our study observed that, after accounting for covariates, moderate GNC did not show a significant association with PSU and PIU. However, high GNC was found to be independently and strongly associated with increased severity of PSU and PIU among general Chinese adolescents. These associations remained significant even after adjusting for covariates such as mental health status and social environment. It is worth noting that the association between high GNC and PIU was more pronounced compared to PSU. Furthermore, when considering covariates such as mental health status and social environment in the models, the unstandardized regression coefficient between GNC and PSU/PIU was slightly attenuated, suggesting that depressive symptoms, anxiety symptoms, low social support status, and experiences of being bullied may also contribute to the severity of PSU and PIU among Chinese adolescents. Moreover, it is also important to highlight that high GNC students with mental health problems or adverse social environments may also exhibit higher levels of PSU and PIU.
Several potential explanations can account for the observed association between high GNC and the severity of PSU and PIU among adolescents. Although all students may experience social and psychological stressors, high GNC adolescents face even greater challenges. First, high gender-nonconforming adolescents may encounter prejudice and stigma due to their gender presentation not aligning with socially expected gender norms, including physical appearance, clothing and accessories choice, and communication style (Hong & Garbarino, 2012; Reisner, Greytak, Parsons, & Ybarra, 2015). When subjected to gender-based bullying, they are more likely to turn to smartphones and the internet to seek online social support and escape from unfriendly social environments in reality. Consequently, smartphone and internet use may play a crucial role in their daily lives. On the one hand, the online world provides a relatively inclusive environment for gender expression diversity (Vessey et al., 2022). Adolescents may seek acceptance and recognition for their gender expression, as well as form new friendships or engage in social interactions with strangers. On the other hand, smartphones and internet activities can provide them with pleasurable experiences, which help them escape negative moods they may face in the real world (Meng et al., 2020). The observed stronger association with PIU may be attributed to the higher occurrence rate of PIU among the gender-nonconforming participants. These findings highlight that gender-nonconforming students may experience elevated levels of stress related to GNC, leading them to PSU/PIU as a coping strategy, similar to how increased PSU/PIU among sexual minority youths may occur as a coping mechanism in response to social and minority stress (Huang et al., 2022).
Additionally, as previous research suggests (Huang et al., 2022), while there were some similarities in the associations between GNC and PSU/PIU among female and male students, there are also differences. This study found that GNC was not significantly associated with PSU among female students, while high GNC was positively associated with severe PSU among male students. These results may be related to the fact that gender-nonconforming male students may experience more severe minority stressors than gender-nonconforming female students. In Chinese culture, masculine gender role expectations and the stigma towards feminine males are deeply embedded (Wang, Yu, Yang, Drescher, & Chen, 2020), resulting in greater societal tolerance towards girls behaving in a “boyish” manner compared to boys behaving in a “girly” manner (Zucker, 2019). These findings suggest that gender-nonconforming male students may face more social stress and may be more likely to turn to digital technologies as a compensatory coping behavior for the relief of their real-life stress compared with gender-nonconforming female students (Lowry et al., 2018). Further research is needed to better understand the links between gender norms, biological sex, social stressors, and PSU/PIU across the spectrum of gender conformity.
Some limitations of our study should be acknowledged. First, due to the nature of the cross-sectional design, we cannot establish a definite causal relationship between GNC and PSU/PIU among Chinese adolescents. Therefore, further longitudinal studies are needed to explore this issue. Second, our study sample was limited to middle and high school students, so our findings cannot be generalized to other populations, such as young adults in university. Third, since we collected data from schools, gender-nonconforming adolescents not attending school were not included in this study. These individuals may have more severe issues with problematic digital use than the school-based sample. Fourth, self-report questionnaires were used to collect data, which may have resulted in ambiguity in interpreting the gender expression questions, and reporting of PSU and PIU behaviors could be influenced by peer pressure, leading to information bias. Future studies should consider using digital trace data to capture smartphone and internet usage patterns more accurately. Fifth, although participants with high GNC scores might be the transgender population, there were no transgender individuals in our study sample of Chinese adolescents.
Conclusions
In summary, our study found a significant association between GNC and the severity of PSU and PIU among Chinese adolescents in school settings. Moreover, our results revealed that high levels of GNC were associated with more severe PSU/PIU, particularly among male students. Accordingly, our findings highlight the need to pay greater attention to GNC in schools and promote educational initiatives that support gender expression diversity. To foster a nondiscriminatory and inclusive school environment, addressing the vulnerabilities of gender-nonconforming adolescents, particularly male students who are more likely to exhibit higher levels of PSU, is essential. As such, school psychologists and specialized support services should play a vital role in facilitating real-life interactions among gender-nonconforming students, their families, classmates, and teachers, thereby effectively preventing PSU and PIU in this vulnerable population.
Funding sources
This research was supported by a grant from the Natural Science Foundation of China Natural Science Foundation of Guangdong Province (Dr. Guo, 2022A1515012333) and the Science and Technology Planning Project of Guangzhou (Dr. Guo, 202102020136).
Authors’ contribution
LG contributed to the conception and design of the work. XYZ, YWY, and LG performed the analysis. All authors contributed to the interpretation of data. WQJ, YTH, CHH, YLH, and CYL accessed and verified the underlying data reported. XYZ and YWY wrote the initial draft of the manuscript. All authors critically revised the manuscript. All authors had full access to all the data in the study and had final responsibility for the decision to submit for publication.
Conflict of interest
We declare no competing interests.
Acknowledgment
We appreciate all participants and schools who participated in the School-based Chinese Adolescents Health Survey.
Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1556/2006.2023.00040.
References
American Psychological Association. (2023). Understanding transgender people, gender identity and gender expression. Retreved 2023.06.18 from https://www.apa.org/topics/lgbtq/transgender-people-gender-identity-gender-expression.
Arrivillaga, C., Rey, L., & Extremera, N. (2022). Psychological distress, rumination and problematic smartphone use among Spanish adolescents: An emotional intelligence-based conditional process analysis [Journal Article; Research Support, Non-U.S. Gov’t]. Journal of Affective Disorders, 296, 1–8. http://doi.org/10.1016/j.jad.2021.09.021.
Becker, I., Ravens-Sieberer, U., Ottová-Jordan, V., & Schulte-Markwort, M. (2017). Prevalence of adolescent gender experiences and gender expression in Germany [Journal Article]. Journal of Adolescent Health, 61(1), 83–90. http://doi.org/10.1016/j.jadohealth.2017.02.001.
Bewley, S., McCartney, M., Meads, C., & Rogers, A. (2021). Sex, gender, and medical data [Editorial]. Bmj-British Medical Journal, 372, n735. http://doi.org/10.1136/bmj.n735.
Brand, M. (2022). Can internet use become addictive? [Journal Article]. Science, 376(6595), 798–799. http://doi.org/10.1126/science.abn4189.
Brown, C. S., & Stone, E. A. (2016). Gender stereotypes and discrimination: How sexism impacts development [Journal Article; Review]. Advances in Child Development and Behavior, 50, 105–133. http://doi.org/10.1016/bs.acdb.2015.11.001.
China Internet Network Information Center. (2022, 2022-12-1). The 2021 national underage internet use research report. Retreved 2023.6.18 from https://www.cnnic.net.cn/6/86/116/index.html.
Christakis, D. A. (2019). The challenges of defining and studying “digital addiction” in children [Journal Article]. Jama-Journal of the American Medical Association, 321(23), 2277–2278. http://doi.org/10.1001/jama.2019.4690.
van den Eijnden, R., Koning, I., Doornwaard, S., van Gurp, F., & Ter Bogt, T. (2018). The impact of heavy and disordered use of games and social media on adolescents’ psychological, social, and school functioning [Journal Article]. Journal of Behavioral Addictions, 7(3), 697–706. http://doi.org/10.1556/2006.7.2018.65.
Elhai, J. D., Levine, J. C., & Hall, B. J. (2019). The relationship between anxiety symptom severity and problematic smartphone use: A review of the literature and conceptual frameworks [Journal Article; Systematic Review]. Journal of Anxiety Disorders, 62, 45–52. http://doi.org/10.1016/j.janxdis.2018.11.005.
Feijóo, S., & Rodríguez-Fernández, R. (2021). A meta-analytical review of gender-based school bullying in Spain [Introductory Journal Article; Journal Article; Meta-Analysis; Review]. International Journal of Environmental Research and Public Health, 18(23) http://doi.org/10.3390/ijerph182312687.
Gong, Z., Lv, Y., Jiao, X., Liu, J., Sun, Y., & Qu, Q. (2022). The relationship between COVID-19-related restrictions and fear of missing out, problematic smartphone use, and mental health in college students: The moderated moderation effect of resilience and social support [Journal Article; Research Support, Non-U.S. Gov’t]. Frontiers in Public Health, 10, 986498. http://doi.org/10.3389/fpubh.2022.986498.
Greytak, E. A. G. A. (2016). Identifying transgender and other gender minority respondents on population-based surveys. In H. J. L. (Ed.), Best practices for asking questions to identify transgender and other gender minority respondents on population-based surveys (2014).
Guo, W., Tao, Y., Li, X., Lin, X., Meng, Y., Yang, X., … Li, T. (2020). Associations of internet addiction severity with psychopathology, serious mental illness, and suicidality: Large-sample cross-sectional study [Journal Article]. Journal of Medical Internet Research, 22(8), e17560. http://doi.org/10.2196/17560.
Hong, J. S., & Garbarino, J. (2012). Risk and protective factors for homophobic bullying in schools: An application of the social–ecological framework. Educational Psychology Review, 24, 271–285. http://doi.org/10.1007/s10648-012-9194-y.
Huang, M. F., Chang, Y. P., Lu, W. H., & Yen, C. F. (2022). Problematic smartphone use and its associations with sexual minority stressors, gender nonconformity, and mental health problems among young adult lesbian, gay, and bisexual individuals in Taiwan [Journal Article; Research Support, Non-U.S. Gov’t]. International Journal of Environmental Research and Public Health, 19(9). http://doi.org/10.3390/ijerph19095780.
Kardefelt-Winther, D. (2014). A conceptual and methodological critique of internet addiction research: Towards a model of compensatory internet use. Computers in Human Behavior, 31, 351–354. http://doi.org/10.1016/j.chb.2013.10.059.
Kawabe, K., Horiuchi, F., Ochi, M., Oka, Y., & Ueno, S. (2016). Internet addiction: Prevalence and relation with mental states in adolescents [Journal Article]. Psychiatry and Clinical Neurosciences, 70(9), 405–412. http://doi.org/10.1111/pcn.12402.
Kósa, G., Feher, G., Horvath, L., Zadori, I., Nemeskeri, Z., Kovacs, M., … Tibold, A. (2022). Prevalence and risk factors of problematic internet use among Hungarian adult recreational Esports players [Journal Article; Research Support, Non-U.S. Gov’t]. International Journal of Environmental Research and Public Health, 19(6). http://doi.org/10.3390/ijerph19063204.
Kwon, M., Kim, D. J., Cho, H., & Yang, S. (2013). The smartphone addiction scale: Development and validation of a short version for adolescents [Journal article; validation study]. Plos One, 8(12), e83558. http://doi.org/10.1371/journal.pone.0083558.
Lai, W., Li, W., Guo, L., Wang, W., Xu, K., Dou, Q., … Lu, C. (2023). Association between bullying victimization, coping style, and mental health problems among Chinese adolescents [Journal Article; Research Support, Non-U.S. Gov’t]. Journal of Affective Disorders, 324, 379–386. http://doi.org/10.1016/j.jad.2022.12.080.
Lian, Q., Li, R., Liu, Z., Li, X., Su, Q., & Zheng, D. (2022). Associations of nonconforming gender expression and gender identity with bullying victimization: An analysis of the 2017 youth risk behavior survey [Journal Article; Research Support, Non-U.S. Gov’t]. Bmc Public Health, 22(1), 650. http://doi.org/10.1186/s12889-022-13071-6.
Lian, Q., Lou, C., Zhong, X., Zhang, J., Tu, X., Fang, Y., … Zuo, X. (2022). Nonconforming gender expression and insufficient sleep among adolescents during COVID-19 school closure and after school reopening [Journal Article; Research Support, Non-U.S. Gov’t]. Bmc Public Health, 22(1), 2037. http://doi.org/10.1186/s12889-022-14463-4.
Lin, L., Cao, B., Chen, W., Li, J., Zhang, Y., & Guo, V. Y. (2022). Association of adverse childhood experiences and social isolation with later-life cognitive function among adults in China [Journal Article; Research Support, Non-U.S. Gov’t]. Jama Network Open, 5(11), e2241714. http://doi.org/10.1001/jamanetworkopen.2022.41714.
Lin, Y., Xie, H., Huang, Z., Zhang, Q., Wilson, A., Hou, J., … Chen, R. (2021). The mental health of transgender and gender non-conforming people in China: A systematic review [Journal Article; Systematic Review]. Lancet Public Health, 6(12), e954–e969. http://doi.org/10.1016/S2468-2667(21)00236-X.
Liu, C., Liu, Z., & Yuan, G. (2021). Cyberbullying victimization and problematic Internet use among Chinese adolescents: Longitudinal mediation through mindfulness and depression [Journal Article]. Journal of Health Psychology, 26(14), 2822–2831. http://doi.org/10.1177/1359105320934158.
Long, J., Liu, T. Q., Liao, Y. H., Qi, C., He, H. Y., Chen, S. B., & Billieux, J. (2016). Prevalence and correlates of problematic smartphone use in a large random sample of Chinese undergraduates [Journal Article]. Bmc Psychiatry, 16(1), 408. http://doi.org/10.1186/s12888-016-1083-3.
Loso, H. M., Locke, D. S., Chaarani, B., Ivanova, M., Garavan, H., Johns, M. M., & Potter, A. S. (2023). Associations between gender nonconformity, school environments, family conflict, and emotional and behavioral health among children ages 10–11 [Journal Article; Research Support, N.I.H., Extramural]. Journal of Adolescent Health, 72(6), 869–876. http://doi.org/10.1016/j.jadohealth.2023.02.008.
Lowry, R., Johns, M. M., Gordon, A. R., Austin, S. B., Robin, L. E., & Kann, L. K. (2018). Nonconforming gender expression and associated mental distress and substance use among high school students [Journal Article; Multicenter Study]. Jama Pediatrics, 172(11), 1020–1028. http://doi.org/10.1001/jamapediatrics.2018.2140.
Lowry, R., Johns, M. M., & Robin, L. E. (2020). Violence victimization, substance use disparities, and gender-nonconforming youth [Journal Article]. American Journal of Preventive Medicine, 58(5), e159–e169. http://doi.org/10.1016/j.amepre.2019.12.021.
Lu, K. (2017). Number of imputations needed to stabilize estimated treatment difference in longitudinal data analysis [Journal Article]. Statistical Methods in Medical Research, 26(2), 674–690. http://doi.org/10.1177/0962280214554439.
Luk, T. T., Wang, M. P., Shen, C., Wan, A., Chau, P. H., Oliffe, J., … Lam, T. H. (2018). Short version of the Smartphone Addiction Scale in Chinese adults: Psychometric properties, sociodemographic, and health behavioral correlates [Journal Article]. Journal of Behavioral Addictions, 7(4), 1157–1165. http://doi.org/10.1556/2006.7.2018.105.
Méndez, I., Jorquera, H. A., & Ruiz-Esteban, C. (2020). Profiles of mobile phone problem use in bullying and cyberbullying among adolescents [Journal Article]. Frontiers in Psychology, 11, 596961. http://doi.org/10.3389/fpsyg.2020.596961.
Meng, H., Cao, H., Hao, R., Zhou, N., Liang, Y., Wu, L., … Zhang, J. (2020). Smartphone use motivation and problematic smartphone use in a national representative sample of Chinese adolescents: The mediating roles of smartphone use time for various activities. Journal of Behavioral Addictions, 9(1), 163–174. http://doi.org/10.1556/2006.2020.00004.
Meng, S. Q., Cheng, J. L., Li, Y. Y., Yang, X. Q., Zheng, J. W., Chang, X. W., … Shi, J. (2022). Global prevalence of digital addiction in general population: A systematic review and meta-analysis [Journal Article; Meta-Analysis; Research Support, Non-U.S. Gov’t; Review; Systematic Review]. Clinical Psychology Review, 92, 102128. http://doi.org/10.1016/j.cpr.2022.102128.
Meyer, I. H. (2003). Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence [Journal Article; Research Support, U.S. Gov’t, P.H.S.; Review]. Psychological Bulletin, 129(5), 674–697. http://doi.org/10.1037/0033-2909.129.5.674.
Pan, Y. C., Chiu, Y. C., & Lin, Y. H. (2020). Systematic review and meta-analysis of epidemiology of internet addiction [Journal Article; Meta-Analysis; Research Support, Non-U.S. Gov’t; Review; Systematic Review]. Neuroscience and Biobehavioral Reviews, 118, 612–622. http://doi.org/10.1016/j.neubiorev.2020.08.013.
Reisner, S. L., Greytak, E. A., Parsons, J. T., & Ybarra, M. L. (2015). Gender minority social stress in adolescence: Disparities in adolescent bullying and substance use by gender identity. Journal of Sex Research, 52(3), 243–256. http://doi.org/10.1080/00224499.2014.886321.
Ricoy, M. C., Martínez-Carrera, S., & Martínez-Carrera, I. (2022). Social overview of smartphone use by teenagers [Journal Article; Review]. International Journal of Environmental Research and Public Health, 19(22) http://doi.org/10.3390/ijerph192215068.
Shinetsetseg, O., Jung, Y. H., Park, Y. S., Park, E. C., & Jang, S. Y. (2022). Association between smartphone addiction and suicide [Journal Article]. International Journal of Environmental Research and Public Health, 19(18) http://doi.org/10.3390/ijerph191811600.
Ueno, T., Ito, K., Murai, T., & Fujiwara, H. (2020). Mental health problems and their association with internet use in medical residents [Journal Article; Research Support, Non-U.S. Gov’t]. Frontiers in Public Health, 8, 587390. http://doi.org/10.3389/fpubh.2020.587390.
Vessey, J. A., Difazio, R. L., Neil, L. K., & Dorste, A. (2022). Is there a relationship between youth bullying and internet addiction? An integrative review [Journal Article]. International Journal of Mental Health and Addiction, 1–25. http://doi.org/10.1007/s11469-021-00731-8.
Wang, Y. Y., Yu, H., Yang, Y., Drescher, J., & Chen, R. (2020). Mental health status of cisgender and gender-diverse secondary school students in China. Jama Network Open, 3(10). http://doi/10.1001/jamanetworkopen.2020.22796.
Wylie, S. A., Corliss, H. L., Boulanger, V., Prokop, L. A., & Austin, S. B. (2010). Socially assigned gender nonconformity: A brief measure for use in surveillance and investigation of health disparities [Journal Article]. Sex Roles, 63(3–4), 264–276. http://doi.org/10.1007/s11199-010-9798-y.
Young, K. S. (1998). Caught in the net: How to Recognize the Signs of internet addiction-- and a winning Strategy for recovery (1st ed.). John Wiley & Sons, Inc.
Zhang, Y., Chen, C., Pan, X., Guo, J., Li, Y., Franco, O. H., … Pan, A. (2021). Associations of healthy lifestyle and socioeconomic status with mortality and incident cardiovascular disease: Two prospective cohort studies. Bmj-British Medical Journal, n604. http://doi.org/10.1136/bmj.n604.
Zhang, M. X., Chen, J. H., Tong, K. K., Yu, E. W., & Wu, A. (2021). Problematic smartphone use during the COVID-19 pandemic: Its association with pandemic-related and generalized beliefs [Journal Article; Research Support, Non-U.S. Gov’t]. International Journal of Environmental Research and Public Health, 18(11). http://doi.org/10.3390/ijerph18115724.
Zhao, M., Xiao, D., Wang, W., Wu, R., Dewaele, A., Zhang, W., … Lu, C. (2021). Association of sexual minority status, gender nonconformity with childhood victimization and adulthood depressive symptoms: A path analysis [Journal Article; Research Support, Non-U.S. Gov’t]. Child Abuse & Neglect, 111, 104822. http://doi.org/10.1016/j.chiabu.2020.104822.
Zucker, K. J. (2019). Adolescents with gender dysphoria: Reflections on some contemporary clinical and research issues [Journal Article; Review]. Archives of Sexual Behavior, 48(7), 1983–1992. http://doi.org/10.1007/s10508-019-01518-8.
