Problematic Internet Use and Technology Interactions Among Transgender Adolescents

Abstract

Purpose:

To compare risk of problematic internet use (PIU) and importance of digital media interactions for transgender and cisgender adolescents.

Methods:

A nationally representative group of adolescents took an online survey that included a measure of PIU (Problematic and Risky Internet Use Screening Scale-3 [PRIUSS-3]) and technology interactions (Adolescent Digital Technology Interactions and Importance scale). We compared mean scores for these scales and their subscales and rates of positive screens for PIU for transgender and cisgender adolescents.

Results:

Of 4575 adolescents participating, 53 (1.2%) were transgender, nonbinary, and gender-diverse (TNG) adolescents. TNG adolescents had higher PRIUSS-3 scores and higher mean scores for importance of technology to explore identity/go outside their offline environment.

Conclusions:

TNG adolescents report higher PIU risk, which may relate to differences in technology importance for this group.

Introduction

Problematic internet use (PIU), or “[i]nternet use that is risky, excessive, or impulsive in nature, leading to adverse life consequences”¹ (p. 1885), is associated with depression, attention-deficit/hyperactivity disorder, and increased stress.^2,3 PIU is a broad term that includes internet addictions (such as Internet Gaming Disorder, a Diagnostic and Statistical Manual of Mental Disorders 5 diagnosis). This term also acknowledges negative effects of internet use beyond addictive behaviors, such as interference with social interactions. In demographic assessments, adults who are male⁴ or who identify as lesbian, gay, bisexual, or unsure have a higher likelihood of internet addiction.⁵

Transgender, nonbinary, and gender-diverse (TNG) adolescents and their caregivers describe the internet as an important resource to circumvent barriers to identity exploration, expression, and support⁶ encountered by this group.^7,8 Lesbian, gay, bisexual, transgender, and queer/questioning (LGBTQ+) youth spend more time online than their heterosexual and cisgender peers.⁹ Qualitative research shows that TNG adolescents specifically identify the internet as an important resource¹⁰ and opportunity to build confidence, experience belonging and hope, give back to TNG communities, and escape discrimination and violence.¹¹ In adult TNG populations, online spaces and social media are central to community building, support, and knowledge sharing.¹² While these studies have described the ways that TNG people may engage with the internet, the specifics of technology interactions and importance have not been assessed in this group using validated measures.

Despite the identification of the importance of online space and digital resources in this group, literature related to PIU in TNG people is limited. Studies of TNG adult gamers have suggested lower rates of Internet Gaming Disorder than found in the general population¹³ and a functional role for video games in identity exploration,¹⁴ although these studies are limited by use of nonclinical instruments and very small sample size (four case studies), respectively. Quantitative studies have not specifically examined PIU and technology interactions in TNG adolescents. Our goal in this study was to compare PIU and importance of technology interactions between cisgender and TNG adolescents.

Methods

This secondary analysis used cross-sectional survey data collected through the online survey platform, Qualtrics, between February and March 2019. The University of Wisconsin-Madison Education and Social Behavioral Sciences IRB approved this study. We used a survey panel approach through Qualtrics survey platform to study a representative sample of adolescents in the United States. Qualtrics was chosen because of their ability to provide demographic samples within 10% of their reciprocal U.S. population values for race, sex, and annual household income.¹⁵ Recruitment was initiated by the Qualtrics survey manager who identified adult parents of adolescents who had signed up for survey participation.

Parents and young adults (18 years of age) provided consent; adolescents (13–17 years old) provided assent before initiation of the survey. Participants were given the option to discontinue the survey at any time without loss of benefits. Inclusion criteria for participation in this study were adolescents (13–18 years) who answered a question asking their gender identity and did not select “Prefer not to answer.” Additional description of Qualtrics recruitment as well as consent/assent procedures are detailed elsewhere.¹⁶

Measures

Demographics

Demographic survey questions assessed age, gender, race, and ethnicity. Participants were considered cisgender if they answered “Female” or “Male” and TNG if they answered “Nonbinary gender,” “Female to Male transgender,” “Male to female transgender,” or “Other.” It is important to note that this is not up-to-date terminology and does not follow standard two-step data collection for sex and gender identity.¹⁷ As this was a secondary analysis, we were unable to change the questions asked. With this in mind, and given that the questions asked were unlikely to misidentify cisgender adolescents as TNG, we opted to use the gender categories available to allow for analysis (rather than invisibility of) of TNG adolescents with acknowledgment of the methodological limitations.

Measures of digital media use

PIU was measured using the validated short version of the Problematic and Risky Internet Use Screening Scale-3 (PRIUSS-3; Cronbach's alpha α=0.87).¹⁸ Questions include: “How often do you (1) experience increased social anxiety due to your internet use? (2) feel withdrawal when away from the internet? and (3) lose motivation to do other things that need to get done because of the internet?” with answers ranging from Never (0) to Very Likely (4) on a Likert Scale. A score of three or more on PRIUSS-3 was considered at risk for PIU.

Perceived importance of technology interactions was measured using the 18-item Adolescent Digital Technology Interactions and Importance scale (ADTI; α=0.92),¹⁹ which has three subscales: (1) technology to bridge online and offline experiences and preferences (α=0.87), (2) technology to go outside one's identity or offline environment (α=0.90), and (3) technology for social connection (α=0.82). Responses are summed; higher scores indicate higher perceived importance of technology interactions.

The PRIUSS-3 and ADTI have been validated in general adolescent and young adult populations^18,19; they have not been validated specifically among TNG adolescents.

Analyses

We used a two-tailed t-test to compare age as a continuous variable and Fisher's Exact Test for categorical analysis, including comparing age groups, race, and ethnicity between gender identity groups. Generalized estimating equation (GEE) analysis was conducted to compare the proportions of subjects at risk for PIU (PRIUSS-3≥3). PRIUSS and ADTI scores were compared between gender groups with analysis of covariance. All comparisons were adjusted for age, race, and ethnicity. Statistical analyses were performed using SAS software version 9.4 (SAS Institute, Inc., Cary, NC).

Results

Among 4575 participants, 53 (1.2%) identified as TNG. Age distribution and mean age did not differ between TNG and cisgender adolescents. More TNG adolescents identified as non-white (50.9% vs. 32.7%, p<0.0001) and Hispanic compared with cisgender peers (59.6% vs. 17.6%, p<0.0001) (Table 1).²⁰

Table 1.

Demographics of Cisgender and Transgender, NonBinary, and Gender-Diverse Adolescent Participants

	Cisgender	Transgender/nonbinary/gender diverse	p	Total—all genders (% total)
Gender identity (total study population), n (%)
Female/feminine identity	2130 (46.56)	5 (0.11)
Male/masculine identity	2392 (52.28)	25 (0.55)
Nonbinary identity		23 (0.50)
Total	4522 (98.84)	53 (1.16)		4575
Age (gender category), n (%)
Mean age	14.62 (±1.68) years	14.57 (±1.66) years	0.82
13–14	2160 (47.94)	29 (54.72)	0.3366	2189 (47.85)
15–18	2346 (52.06)	24 (45.28)		2370 (51.80)
Ethnicity (gender category), n (%)
Non-Hispanic	3683 (82.41)	21 (40.38)	<0.0001	3704 (81.96)
Hispanic	786 (17.59)	31 (59.62)		817 (17.86)
Race (gender category), n (%)
White	3041 (67.25)	26 (49.06)	<0.0001	3067 (67.04)
Black	692 (15.30)	5 (9.43)		697 (15.23)
American Indian or Alaskan Native	139 (3.07)	12 (22.64)		151 (3.30)
Asian	227 (5.02)	8 (15.09)		235 (5.14)
Multiracial	219 (4.84)	2 (3.77)		221 (4.83)
Other	204 (4.51)	0 (0.00)

No adolescents selected “Other” in identifying their gender identity. For reference, the percent populations based on race in the 2021 US Census Bureau data are as follows: 76.8% White, 13.6% Black, 1.3% American Indian or Alaskan Native, 6.1% Asian, and 0.3% Native Hawaiian or Pacific Islander. Percent populations based on ethnicity in the 2021 US Census Bureau data are as follows: 81.12% non-Hispanic and 18.88% Hispanic.²⁰

TNG adolescents had significantly higher mean PRIUSS-3 scores than their cisgender peers (5.89 vs. 4.69, p=0.0152; see Table 2). TNG adolescents also showed higher probabilities of scoring at risk for PIU on the PRIUSS-3 compared with cisgender adolescents (0.92 vs. 0.70, p=0.0025), cisgender males (0.70, p=0.0024), and cisgender females (0.71, p=0.0028). Mean total ADTI scores were higher in TNG adolescents compared with cisgender adolescents overall (54.36 vs. 49.29, p=0.0492) (Table 2). Measures of ADTI-1 and ADTI-3 subscales did not differ between groups. TNG adolescents had a higher mean ADTI-2 subscale scores than cisgender adolescents overall (19.90 vs. 15.85, p=0.0004), cisgender females (19.90 vs. 15, p<0.0001), and cisgender males (19.90 vs. 16.64, p=0.0044).

Table 2.

Comparison of Mean Scores of Problematic Internet Use and Technology Interactions and Importance Between Cisgender and Transgender, Nonbinary, and Gender-Diverse Adolescents

Outcome	Female	TNG	p	Male	TNG	p	Male + female	TNG	p
Digital media use measures
PRIUSS-3	4.57	5.89	0.0076	4.81	5.89	0.0291	4.69	5.89	0.0152
ADTI total	48.96	54.36	0.0375	49.59	54.36	0.0656	49.29	54.36	0.0492
ADTI Subscale—Factor 1	17.57	17.86	0.7596	17.07	17.86	0.3985	17.31	17.86	0.5563
ADTI Subscale—Factor 2	15.00	19.91	<0.0001	16.64	19.91	0.0044	15.85	19.90	0.0004
ADTI Subscale—Factor 3	16.39	16.45	0.9408	15.85	16.45	0.4515	16.11	16.45	0.6681

Adjusted for age, race, and ethnicity. Comparisons with p<0.05 are bolded.

ADTI, Adolescent Digital Technology Interactions and Importance scale; PRIUSS-3, Problematic and Risky Internet Use Screening Scale-3; TNG, transgender, nonbinary, and gender diverse.

Discussion

This study is the first to show that TNG adolescents report higher risk for PIU, greater overall perceived importance of technology interactions, and higher importance in using digital media to explore their identity or go outside their offline environment compared with cisgender adolescents. Given that PIU is associated with depression, attention-deficit/hyperactivity disorder, and increased stress in general adolescent populations,^2,3 and that TNG adolescents have increased risk for depression, self-harm, and other mental health concerns,^21,22 increased PIU could be interpreted as a reflection of mental health challenges. However, differences in how TNG adolescents and cisgender adolescents rate the importance of technology interactions for various purposes raise a different possible explanation.

Divergent PIU and ADTI findings between TNG and cisgender adolescents may relate to the two groups' different engagement with digital media affordances, or perceived functional properties of media, in their adolescent identity development. Affordances offer a framework for how people experience different benefits from digital media. Digital media may offer cognitive (learning), social, identity, and emotional affordances.²³ TNG adolescents' higher ADTI-2 subscale scores suggest they are more likely to identify the identity and social affordances of exploring their identity and going outside their offline environment as important in their technology interactions.

This variation in affordances fits with developmental models specifically positing the importance of digital media as a means of exploration, information, and connection as central to identity development for LGBTQ+ youth²⁴ and specifically facilitating rites of passage during gender transition for TNG adolescents.²⁵ In qualitative descriptions of their use of digital media, transgender adolescents describe validation of their experiences in their online interactions.¹⁰ TNG adolescents also identify online space as a source of emotional and informational support^6,10 and a critical place to build hope and confidence, experience belonging, and escape from stigma and violence experienced offline.¹¹ TNG adolescents describe online spaces as opportunities to give back to online communities that have supported their growth and healing,¹¹ such that online engagement may be seen as paying forward positive experiences that facilitate identity formation and resilience.

Given that identification of reflection of oneself in others is a recognized step in identity formation for TNG adolescents,²⁶ it follows that digital media as a means to access these identity and social affordances may be critical in this group and may lead TNG adolescents to spend significant time and energy in online social interactions compared with cisgender peers. With this frame, PIU as measured with current tools may capture use patterns that are central to identity development in this population, complicating the characterization of this use as problematic. In particular, the PRIUSS-3 questions related to feelings of withdrawal when away from the internet and losing interest in other things that need to get done because of the internet may take on new meaning for TNG adolescents if they find positive connection, information, and engagement in online compared with offline environments.

Limitations

While this study includes a robust national adolescent sample with prevalence of TNG identity similar to general population estimates,²⁷ the absolute number of TNG adolescents in our study is limited (n=53). While these factors may limit generalizability, this study represents an important opportunity to examine experiences of TNG adolescents in a large national sample. Additionally, this survey did not use up-to-date terminology to identify gender identity or ask about both sex and gender identity, which may underestimate the number of TNG participants.

Conclusions

Although PIU risk was higher for TNG than cisgender adolescents in this study based on screening with current instruments, this may be related to technology's role in identity formation for TNG adolescents. Our study suggests that it may be important to specifically validate current PIU measures in this population. Future study of both PIU as well as the functional importance of digital media in identity development will help to optimize counseling around PIU as well as online resources and spaces to best serve and support healthy development for TNG adolescents.

Abbreviations Used

ADTI

Adolescent Digital Technology Interactions and Importance scale

PIU

problematic internet use

PRIUSS-3

Problematic and Risky Internet Use Screening Scale-3

TNG

transgender, nonbinary, and gender diverse

Authors' Contributions

B.J.A.: Conceptualization, methodology, project administration, visualization, writing—original draft, and writing—review and editing. Z.E.S.: Writing—original draft and writing—review and editing. B.R.K.: Data curation, investigation, and writing—review and editing. Q.Z.: Data curation, formal analysis, methodology, resources, writing—original draft, and writing—review and editing. M.A.M.: Conceptualization, data curation, funding—acquisition, investigation, methodology, project administration, supervision, and writing—review and editing.

Author Disclosure Statement

No authors have conflicts of interest to report.

Funding Information

This study was funded by a collaborative research agreement between Dr. Moreno as Principal Investigator (PI) and Facebook, Inc. The first author did not receive funds as part of this research agreement.

Cite this article as: Allen BJ, Stratman ZE, Kerr BR, Zhao Q, Moreno MA (2022) Problematic internet use and technology interactions among transgender adolescents, Transgender Health X:X, 1–5, DOI: 10.1089/trgh.2021.0169.