Page 1 of 5

Advances in Social Sciences Research Journal – Vol. 11, No. 12

Publication Date: December 25, 2024

DOI:10.14738/assrj.1112.18110.

Cavalcanti, B., Hemsley, P., & Moraes, F. (2024). Internet Use and Mental Health: Evidence from a Global Panel of 189 Countries.

Advances in Social Sciences Research Journal, 11(12). 375-379.

Services for Science and Education – United Kingdom

Internet Use and Mental Health: Evidence from a Global Panel of

189 Countries

Bianca Cavalcanti

Federal University of Rio de Janeiro.

Rio de Janeiro, RJ, Brazil

Pedro Hemsley

Federal University of Rio de Janeiro.

Rio de Janeiro, RJ, Brazil

Flavio Moraes

Getulio Vargas Foundation, Brazilian School of Public and

Business Administration, Rio de Janeiro, Brazil

ABSTRACT

This paper examines the relationship between internet use and mental health

outcomes, focusing on anxiety and depression across 189 countries from 1990 to

2017. Using fixed effects panel regressions, we find a positive and statistically

significant association between internet penetration and anxiety prevalence, with

evidence of diminishing marginal effects. For depression, the results are less

consistent, with significance only in specific model specifications. Robustness

checks using lagged internet penetration and alternative model specifications

reinforce the main findings. The results highlight the potential mental health costs

of increased internet use, particularly in low-income countries where the effects are

most pronounced. These findings underscore the need for policy interventions to

promote healthy internet usage and mitigate potential negative mental health

impacts.

Keywords: Internet Use; Mental Health; Anxiety; Depression.

JEL codes: I12, I18, I31

INTRODUCTION

The rapid expansion of the internet has transformed human interactions, social behavior, and

access to information. While these changes have brought undeniable benefits, there is growing

concern about the potential mental health costs associated with internet use. Anxiety and

depression, the two most prevalent mental disorders worldwide, have seen rising incidence

rates in recent decades (Dattani et al [4]). This paper investigates the potential link between

internet use and these mental health disorders using a cross-country panel data from 189

countries between 1990 and 2017.

Research on the relationship between internet use and mental health remains inconclusive.

Some studies suggest that internet usage exacerbates mental health issues by fostering social

comparison, fear of missing out (FoMO), and sleep disturbances (Braghieri et al. [2]; Kim and

Page 2 of 5

376

Advances in Social Sciences Research Journal (ASSRJ) Vol. 11, Issue 12, December-2024

Services for Science and Education – United Kingdom

Lee [6]; Twenge and Campbell [11]; Przybylski et al. [9]). Other studies highlight the potential

of online platforms to offer social support and access to mental health resources (Richards and

Richardson [10]; Andersson and Titov [1]). Given this ambiguity, understanding the

relationship between internet use and mental health outcomes is a critical task, with important

policy implications.

Using data from the Institute for Health Metrics and Evaluation (IHME) for mental health

outcomes and from the World Bank for internet usage, this study applies a fixed effects panel

regression to estimate the relationship between internet use and the incidence of anxiety and

depression. The results indicate a positive correlation between internet usage and the

prevalence of these mental health disorders, with diminishing marginal effects. While the

relationship is statistically significant for anxiety in most model specifications, the evidence for

depression is less robust, emerging in only one of five models tested.

This paper contributes to the literature in two ways. First, it provides a large-scale longitudinal

analysis of the relationship between internet use and mental health, addressing gaps left by

studies that rely on cross-sectional or regional data (Kim and Lee [6]; Twenge and Campbell

[11]; Boers et al. [3]; Exelmans and Bulck [5]). The use of longitudinal and cross-country panel

data is crucial as it captures both temporal and cross-sectional variation, allowing for the

control of country-specific, time-invariant factors such as cultural or institutional

characteristics, and mitigating potential biases from unobserved heterogeneity and

endogeneity. This approach is essential for identifying broader patterns that are not observable

in cross-sectional studies on individual countries or regions (Maurseth [8]; Mars et al. [7]).

Second, by using a quadratic specification of internet use, it captures the nonlinear effects of

exposure, which may explain the mixed findings in previous studies.

The remainder of the paper is structured as follows. Section 2 describes the data sources and

empirical methodology. Section 3 presents the main results and robustness checks. Section 4

briefly concludes.

METHODOLOGY AND DATA

This study relies on panel data from 189 countries over the period 1990 to 2017. The primary

source of data on mental health outcomes (specifically, anxiety and depression) is the Global

Burden of Disease (GBD) database from the Institute for Health Metrics and Evaluation (IHME).

Data on internet usage rates are obtained from the World Bank’s World Development

Indicators (WDI), which includes information on each country's internet penetration rate,

defined as the percentage of the population with internet access, as well as country-level

demographic and macroeconomic control variables.

To estimate the relationship between internet use and mental health disorders, we use the

following fixed effects panel regression model:

Dit = β0 + β1Internetit + β2Internetit

2 + αi + γt + Xit ⋅ δ + εit

Where:

• Dit denotes the prevalence rate of either anxiety or depression in country i at time t.

Page 3 of 5

377

Cavalcanti, B., Hemsley, P., & Moraes, F. (2024). Internet Use and Mental Health: Evidence from a Global Panel of 189 Countries. Advances in Social

Sciences Research Journal, 11(12). 375-379.

URL: http://dx.doi.org/10.14738/assrj.1112.18110

• Internetit is the percentage of the population with access to the internet in country i at

time t.

• Xit is a vector of control variables: GDP per capita, female share of the population,

homicides per capita, and displaced population.

• αi denotes country-specific fixed effects to control for time-invariant heterogeneity

across countries.

• γt captures time-fixed effects to account for global shocks that affect all countries

equally.

• εit is the error term, assumed to be independently and identically distributed.

The inclusion of Internetit

2

allows us to capture potential diminishing or nonlinear effects of

internet use on mental health outcomes. Control variables address potential confounding

factors that could influence mental health and internet use simultaneously. Country fixed

effects account for unobserved characteristics that are constant over time, while time fixed

effects control for global trends that could affect all countries in a similar manner. Lastly, we

use robust standard errors in all specifications. To address endogeneity concerns, robustness

checks include lagged values of the internet penetration variable as instruments and the use of

alternative model specifications.

RESULTS

The main results are summarized in Tables 1 and 2. The fixed effects regression results reveal

a positive and statistically significant relationship between internet use and anxiety prevalence.

The coefficient of Internetit is positive in all model specifications: a higher level of internet

usage is associated with higher rates of anxiety. The coefficient of Internetit

2

is negative and

significant in all specifications, indicating diminishing marginal effects of internet usage on

anxiety rates. The inclusion of quadratic terms suggests that, at lower levels of internet

penetration, the impact on anxiety is stronger, but it diminishes as internet usage increases.