Digital Technology Use and Well-Being in Young Adolescents

Introduction

Media reports often highlight the negative impacts of digital technology on adolescents, including depression, suicide, sleep deprivation, and poor academic performance.
However, recent research and large-scale analyses suggest only small or negligible associations between digital technology use and well-being.
Traditional research often assumes a direct causal link from digital media use to well-being, but co-construction theories suggest a more complex, bidirectional relationship.
Co-construction theories view adolescents as active participants in online contexts, where online activities reflect offline behaviors and contexts.
Adolescents with offline vulnerabilities may be more prone to negative online experiences, while those with offline support may benefit more from digital technologies.
Family socioeconomic status significantly influences resources, experiences, and opportunities for young people, creating an "opportunity gap."
The digital divide traditionally refers to unequal access to technology based on socioeconomic factors, but this gap is closing in the US.
Digital inequality, a new type of divide, is emerging across socioeconomic strata, with disparities in the quality of online experiences and associations between online activities and well-being.

Methods

Participants: 2104 adolescents (ages 10-15 years) representative of North Carolina Public School students in grades 3-6 during 2011-2012.
- 52% female; 52% White, 23% Black, 15% Latino/a, 10% other.
Data collection: Questionnaires completed in 2015, linked with administrative educational records (with parental consent).
Measures:
- Family economic disadvantage: Assessed from administrative data on eligibility for free or reduced-price lunch (cutoffs around 175% of the federal poverty level), scored as ordinal (not, intermittently, persistently disadvantaged).
  - Not disadvantaged: 41%
  - Intermittently disadvantaged: 22%
  - Persistently disadvantaged: 37%
- Neighborhood income: Estimated median household income from the American Community Survey 5-year estimates (2010-2014), mean-centered and standardized.
- Digital technology use: Measured using questions from Pew Internet & American Life national surveys, assessing ownership and access to devices and social media.
- Frequency of social media use: Scale of 1 (less often than every few weeks) to 6 (several times a day).
- Perceived impairment: 6 items assessing negative effects of technology use on daily life (scale of 0 [never] to 2 [always]; mean = $0.63 \pm 0.42$ ; $\,\alpha = 0.70$ ). Binary marker created for reporting at least 1 item sometimes or always (91%).
- Perceived spillover of online experiences: 6-item checklist of experiences on social media resulting in offline problems (yes/no). Binary measures created for at least 1 spillover experience (29%) and 1 or more serious spillover experiences (15%).
- Academic achievement: End-of-grade standardized test scores for reading and math (2014-2015).
- School belonging: 6-item Psychological Sense of School Membership scale (scale of 0 [not at all true] to 5 [very true]; $\,\alpha = 0.84$ ).
- Conduct problems: 26-item Problem Behavior Frequency Scale of behavioral aggression and violence in the last 30 days (scale of 0 [never] to 5 [$\geq$20 times]). Converted to a count of reported problems.
- Psychological distress: 6-item Kessler Psychological Distress Scale (scale of 0 [never] to 5 [$\geq$20 times]; $\,\alpha = 0.66$ ).
- General physical health: Scale of 0 (poor) to 4 (excellent) from the Add Health General Health and Diet survey.
Statistical analyses:
- Descriptive statistics for technology access and use.
- Regression analyses to show associations between digital technology measures and well-being indicators.
  - Zero-order associations.
  - Associations controlling for demographic and economic covariates (age, race/ethnicity, sex, family and neighborhood-level economic disadvantage).
- Mean comparisons to test for differences across economic disadvantage groups.
- Regression analyses to test for moderation by economic disadvantage in associations with well-being, controlling for demographic and neighborhood income covariates.
- Benjamini-Hochberg Method to correct for false discovery rates.

Results

Prevalence of digital technology use:
- Internet access: 95%
- Mobile phone ownership: 67% (85% smartphones).
- Social media account: 68%
- Mobile phone ownership and social media access increased with age.
- Perceived moderate levels of technology-related impairment (mean = $0.63 \pm 0.42$ ), with 91% reporting at least one type of impairment.
- Approximately 1 in 3 adolescents reported online-to-offline spillover, with 15% perceiving more serious forms of spillover.
Associations Between Digital Technologies Use and Adolescents’ Well-being
- Mobile phone ownership:
  - Small positive association with higher standardized reading scores ( $\beta = 0.07$ ; $P = 0.003$ ) and more self-reported conduct problems ( $\beta = 0.05$ ; $P = 0.04$ ).
  - No significant associations after controlling for demographic and economic factors.
  - Similar results for smartphone ownership.
- Social media account:
  - Positively associated with every outcome except standardized reading scores in step 1.
  - After controlling for confounding factors, associations with lower standardized math scores ( $\beta = -0.04$ ; $P = 0.046$ ), more conduct problems ( $\beta = 0.17$ ; P < 0.001), and greater psychological distress ( $\beta = 0.06$ ; $P = 0.006$ ) remained.
- Frequency of social media use:
  - Not associated with academic achievement or psychological distress.
  - Associated with more reported conduct problems ( $\beta = 0.08$ ; $P = 0.006$ ).
- Perceived impairment and spillover:
  - More technology-related impairments or spillover reported with more difficulties with all 6 measures of well-being.
  - Associations held over and above controls for economic disadvantage and demographic factors ( $\beta$ s ranged from $-0.05$ to 0.33).
Digital technology use, perceived impairments, and well-being by family economic status
- Adolescents were equally likely to own a mobile phone across economic statuses ( $\chi^2(2) = 2.21$ ; $P = 0.33$ ).
- Persistently disadvantaged youths were more likely to have a social media account ( $\chi^2(2) = 11.69$ ; $P = 0.003$ ).
- Intermittently disadvantaged adolescents reported greater perceived technology impairments (mean = $0.68 \pm 0.44$ ) compared with nondisadvantaged adolescents (mean = $0.60 \pm 0.40$ ), $F(2, 2038) = 5.49$ ; $P = 0.004$ .
- Greater proportion of intermittently (35%) and persistently (33%) disadvantaged adolescents perceived negative online-to-offline spillover of their technology use compared with nondisadvantaged youth (22%), $\chi^2(2) = 36.31$ (P < 0.001).
Moderation by Economic Disadvantage
- No significant interactions between economic status and mobile phone ownership, frequency of social media use, or perceived spillover on well-being.
- Economic disadvantage moderated associations between:
  - Having a social media account and conduct problems ( $\beta_{\text{interaction}} = 0.11$ ; $P = 0.01$ ).
  - Having a social media account and psychological distress ( $\beta_{\text{interaction}} = 0.15$ ; $P = 0.001$ ).
  - Perceived impairment and conduct problems ( $\beta_{\text{interaction}} = 0.13$ ; $P = 0.002$ ).
- Adolescents with a social media account exhibited higher levels of conduct problems across economic groups.
- Persistently disadvantaged adolescents had higher psychological distress when they had a social media account.
- Adolescents with higher perceived impairment exhibited higher levels of conduct problems across economic groups.

Discussion

The study provides estimates of digital technology prevalence and correlates in a large, contemporary, representative sample of young adolescents.
Internet access is nearly universal, and mobile phone/social media use increases rapidly between ages 10-15.
Small bivariate associations were found between digital technology use and well-being, with economic and demographic covariates accounting for initial associations with mobile phone ownership.
Having a social media account was associated with higher psychological distress and more conduct problems after controlling for confounding factors.
Frequency of social media use was consistently associated with greater conduct problems only.
Nearly all adolescents reported technology-related impairments, and approximately one-third reported negative spillover of online experiences.
There were a lack of robust, strong associations between digital technology use and academic, psychological, and physical well-being.
Gaps in access to the Internet, mobile phones, and social media have narrowed or disappeared among adolescents in the US.
A new digital divide may be emerging, with economically disadvantaged youth more likely to report perceived technology-related impairments and spillover of online experiences.

Limitations

The cross-sectional, correlational nature of the study prevents causal interpretations.
The self-reported nature of technology use and measures of well-being.
Binary markers of access and ownership do not allow for classification of heavy users and non-linear associations.

Conclusion

Digital technology use does not seem to be strongly or reliably associated with young adolescents’ well-being at the population level.
Youth with offline vulnerabilities, including those from lower-income households, may be at heightened risk for perceived impairments and stronger negative associations between digital technology use and well-being.
Parents and teachers should monitor youth already struggling with school or health problems to better understand their digital technology use.
Ensuring equity in access, experiences, and opportunities in both online and offline spaces should be prioritized.”

Introduction

Media reports often highlight the negative impacts of digital technology on adolescents, including depression, suicide, sleep deprivation, and poor academic performance. However, it is important to note that the relationship is complex and multifaceted.
Recent research and large-scale analyses suggest only small or negligible associations between digital technology use and well-being. This challenges the commonly held belief that digital technology has a significant negative impact on adolescents' well-being.
Traditional research often assumes a direct causal link from digital media use to well-being, but co-construction theories suggest a more complex, bidirectional relationship. Co-construction theories emphasize that adolescents are active participants in online contexts and that their online activities are influenced by their offline behaviors and contexts.
Co-construction theories view adolescents as active participants in online contexts, where online activities reflect offline behaviors and contexts. Adolescents are not simply passive recipients of online content but actively shape their online experiences.
Adolescents with offline vulnerabilities may be more prone to negative online experiences, while those with offline support may benefit more from digital technologies. This highlights the importance of considering the broader social and emotional context in which adolescents use digital technology.
Family socioeconomic status significantly influences resources, experiences, and opportunities for young people, creating an "opportunity gap." Adolescents from lower-income families may have fewer opportunities to participate in extracurricular activities, access quality healthcare, and receive support from their parents.
The digital divide traditionally refers to unequal access to technology based on socioeconomic factors, but this gap is closing in the US. While access to technology has become more widespread, disparities in the quality of online experiences persist.
Digital inequality, a new type of divide, is emerging across socioeconomic strata, with disparities in the quality of online experiences and associations between online activities and well-being. Adolescents from lower-income families may have limited access to high-speed internet, newer devices, and educational resources online.

Methods

Participants: 2104 adolescents (ages 10-15 years) representative of North Carolina Public School students in grades 3-6 during 2011-2012. The sample was diverse, with 52% female; 52% White, 23% Black, 15% Latino/a, 10% other.
Data collection: Questionnaires were completed in 2015, linked with administrative educational records (with parental consent). This allowed for the collection of both self-reported data and objective measures of academic achievement.
Measures:
- Family economic disadvantage: Assessed from administrative data on eligibility for free or reduced-price lunch (cutoffs around 175% of the federal poverty level), scored as ordinal (not, intermittently, persistently disadvantaged).
- Not disadvantaged: 41%
- Intermittently disadvantaged: 22%
- Persistently disadvantaged: 37%
- Neighborhood income: Estimated median household income from the American Community Survey 5-year estimates (2010-2014), mean-centered and standardized. This provided a measure of the socioeconomic context in which adolescents lived.
- Digital technology use: Measured using questions from Pew Internet & American Life national surveys, assessing ownership and access to devices and social media. These questions captured the range of digital technology use among adolescents.
- Frequency of social media use: Scale of 1 (less often than every few weeks) to 6 (several times a day). This provided a measure of how often adolescents were using social media.
- Perceived impairment: 6 items assessing negative effects of technology use on daily life (scale of 0 [never] to 2 [always]; mean = $0.63 \pm 0.42$ ; $\alpha = 0.70$ ). A binary marker was created for reporting at least 1 item sometimes or always (91%). This captured the extent to which adolescents perceived that technology use was negatively affecting their lives.
- Perceived spillover of online experiences: 6-item checklist of experiences on social media resulting in offline problems (yes/no). Binary measures were created for at least 1 spillover experience (29%) and 1 or more serious spillover experiences (15%). This captured the extent to which adolescents were experiencing negative consequences in their offline lives as a result of their online activities.
- Academic achievement: End-of-grade standardized test scores for reading and math (2014-2015). This provided an objective measure of academic performance.
- School belonging: 6-item Psychological Sense of School Membership scale (scale of 0 [not at all true] to 5 [very true]; $\alpha = 0.84$ ). This captured the extent to which adolescents felt connected to their school community.
- Conduct problems: 26-item Problem Behavior Frequency Scale of behavioral aggression and violence in the last 30 days (scale of 0 [never] to 5 [ $\geq$ 20 times]). Converted to a count of reported problems. This provided a measure of adolescents' engagement in delinquent behaviors.
- Psychological distress: 6-item Kessler Psychological Distress Scale (scale of 0 [never] to 5 [ $\geq$ 20 times]; $\alpha = 0.66$ ). This captured the extent to which adolescents were experiencing symptoms of anxiety and depression.
- General physical health: Scale of 0 (poor) to 4 (excellent) from the Add Health General Health and Diet survey. This provided a measure of adolescents' overall physical health.
Statistical analyses:
- Descriptive statistics for technology access and use. This provided an overview of the prevalence of digital technology use among adolescents.
- Regression analyses to show associations between digital technology measures and well-being indicators.
- Zero-order associations. This examined the simple correlations between digital technology use and well-being.
- Associations controlling for demographic and economic covariates (age, race/ethnicity, sex, family and neighborhood-level economic disadvantage). This examined the extent to which the associations between digital technology use and well-being were explained by demographic and economic factors.
- Mean comparisons to test for differences across economic disadvantage groups. This examined whether there were differences in digital technology use and well-being between adolescents from different socioeconomic backgrounds.
- Regression analyses to test for moderation by economic disadvantage in associations with well-being, controlling for demographic and neighborhood income covariates. This examined whether the associations between digital technology use and well-being were different for adolescents from different socioeconomic backgrounds.
- Benjamini-Hochberg Method to correct for false discovery rates. This ensured that the findings were robust to the problem of multiple comparisons.

Results

Prevalence of digital technology use:
- Internet access: 95%
- Mobile phone ownership: 67% (85% smartphones). This indicates that mobile phones were the primary way that adolescents accessed the internet.
- Social media account: 68%
- Mobile phone ownership and social media access increased with age. This suggests that digital technology use becomes more prevalent as adolescents get older.
- Perceived moderate levels of technology-related impairment (mean = $0.63 \pm 0.42$ ), with 91% reporting at least one type of impairment. This suggests that a significant proportion of adolescents were experiencing negative consequences as a result of their technology use.
- Approximately 1 in 3 adolescents reported online-to-offline spillover, with 15% perceiving more serious forms of spillover. This indicates that a significant proportion of adolescents were experiencing negative consequences in their offline lives as a result of their online activities.
Associations Between Digital Technologies Use and Adolescents’ Well-being
- Mobile phone ownership:
- Small positive association with higher standardized reading scores ( $\beta = 0.07$ ; $P = 0.003$ ) and more self-reported conduct problems ( $\beta = 0.05$ ; $P = 0.04$ ).
- No significant associations after controlling for demographic and economic factors. This suggests that the initial associations between mobile phone ownership and well-being were explained by demographic and economic factors.
- Similar results for smartphone ownership.
- Social media account:
- Positively associated with every outcome except standardized reading scores in step 1.
- After controlling for confounding factors, associations with lower standardized math scores ( $\beta = -0.04$ ; $P = 0.046$ ), more conduct problems ( $\beta = 0.17$ ; P < 0.001), and greater psychological distress ( $\beta = 0.06$ ; $P = 0.006$ ) remained. This suggests that having a social media account was independently associated with negative outcomes, even after accounting for other factors.
- Frequency of social media use:
- Not associated with academic achievement or psychological distress. This suggests that the frequency of social media use was not as important as whether or not adolescents had a social media account.
- Associated with more reported conduct problems ( $\beta = 0.08$ ; $P = 0.006$ ).
- Perceived impairment and spillover:
- More technology-related impairments or spillover reported with more difficulties with all 6 measures of well-being. This suggests that adolescents who were experiencing negative consequences as a result of their technology use were also more likely to be experiencing difficulties in other areas of their lives.
- Associations held over and above controls for economic disadvantage and demographic factors ( $\beta$ s ranged from $-0.05$ to 0.33).
Digital technology use, perceived impairments, and well-being by family economic status
- Adolescents were equally likely to own a mobile phone across economic statuses ( $\chi^2(2) = 2.21$ ; $P = 0.33$ ).
- Persistently disadvantaged youths were more likely to have a social media account ( $\chi^2(2) = 11.69$ ; $P = 0.003$ ). This suggests that social media may be particularly important for adolescents from lower-income families.
- Intermittently disadvantaged adolescents reported greater perceived technology impairments (mean = $0.68 \pm 0.44$ ) compared with nondisadvantaged adolescents (mean = $0.60 \pm 0.40$ ), $F(2, 2038) = 5.49$ ; $P = 0.004$ . This suggests that adolescents from lower-income families may be more vulnerable to the negative consequences of technology use.
- Greater proportion of intermittently (35%) and persistently (33%) disadvantaged adolescents perceived negative online-to-offline spillover of their technology use compared with nondisadvantaged youth (22%), $\chi^2(2) = 36.31$ (P < 0.001).
Moderation by Economic Disadvantage
- No significant interactions between economic status and mobile phone ownership, frequency of social media use, or perceived spillover on well-being.
- Economic disadvantage moderated associations between:
- Having a social media account and conduct problems ( $\beta_{\text{interaction}} = 0.11$ ; $P = 0.01$ ).
- Having a social media account and psychological distress ( $\beta_{\text{interaction}} = 0.15$ ; $P = 0.001$ ).
- Perceived impairment and conduct problems ( $\beta_{\text{interaction}} = 0.13$ ; $P = 0.002$ ).
- Adolescents with a social media account exhibited higher levels of conduct problems across economic groups.
- Persistently disadvantaged adolescents had higher psychological distress when they had a social media account. This suggests that social media may be particularly harmful for adolescents from lower-income families who are already experiencing psychological distress.
- Adolescents with higher perceived impairment exhibited higher levels of conduct problems across economic groups.

Discussion

The study provides estimates of digital technology prevalence and correlates in a large, contemporary, representative sample of young adolescents. This makes the findings more generalizable to other populations of young adolescents.
Internet access is nearly universal, and mobile phone/social media use increases rapidly between ages 10-15. This highlights the importance of understanding the impact of digital technology on this age group.
Small bivariate associations were found between digital technology use and well-being, with economic and demographic covariates accounting for initial associations with mobile phone ownership. This suggests that the relationship between digital technology use and well-being is complex and influenced by a variety of factors.
Having a social media account was associated with higher psychological distress and more conduct problems after controlling for confounding factors. This is a concerning finding, as it suggests that social media may be having a negative impact on the mental health and behavior of young adolescents.
Frequency of social media use was consistently associated with greater conduct problems only.
Nearly all adolescents reported technology-related impairments, and approximately one-third reported negative spillover of online experiences. This suggests that a significant proportion of adolescents are experiencing negative consequences as a result of their technology use.
There were a lack of robust, strong associations between digital technology use and academic, psychological, and physical well-being. This challenges the commonly held belief that digital technology has a significant negative impact on adolescents' well-being.
Gaps in access to the Internet, mobile phones, and social media have narrowed or disappeared among adolescents in the US. This suggests that the digital divide is closing in terms of access to technology.
A new digital divide may be emerging, with economically disadvantaged youth more likely to report perceived technology-related impairments and spillover of online experiences. This suggests that disparities in the quality of online experiences may be emerging as a new form of digital divide.

Limitations

The cross-sectional, correlational nature of the study prevents causal interpretations. This means that it is not possible to determine whether digital technology use is causing the negative outcomes or whether the negative outcomes are leading to digital technology use.
The self-reported nature of technology use and measures of well-being. This means that the data may be subject to bias.
Binary markers of access and ownership do not allow for classification of heavy users and non-linear associations. This means that the study may not have captured the full range of digital technology use among adolescents.

Conclusion

Digital technology use does not seem to be strongly or reliably associated with young adolescents’ well-being at the population level. This challenges the commonly held belief that digital technology has a significant negative impact on adolescents' well-being.
Youth with offline vulnerabilities, including those from lower-income households, may be at heightened risk for perceived impairments and stronger negative associations between digital technology use and well-being. This highlights the importance of considering the broader social and emotional context in which adolescents use digital technology.
Parents and teachers should monitor youth already struggling with school or health problems to better understand their digital technology use. This suggests that a targeted approach may be more effective than a blanket approach to addressing the potential negative consequences of digital technology use.
Ensuring equity in access, experiences, and opportunities in both online and offline spaces should be prioritized.”