Video Gaming and Its Effects on Mental Health in Portuguese Higher Education Students: An Exploratory Analysis

Abstract

Background: Depression, anxiety, and stress are increasingly prevalent among university students, raising concerns about the role of video gaming behaviors, social support, and academic factors in mental health. Internet Gaming Disorder (IGD), recognized in international classifications, has been linked to psychological distress but remains underexplored in Portuguese higher education students. Objective: This study aimed to examine the relationships between IGD, social support, academic performance, and mental health outcomes. Methods: A cross-sectional survey was conducted with Portuguese university students, collecting sociodemographic information, gaming habits, academic performance, social support, and mental health indicators. Data analyses included non-parametric tests, Spearman correlations, and multiple linear regression models to explore group differences, associations, and predictors of mental health outcomes. Results: No significant gender or age differences were observed in social support or mental health. Students living with parents, engaging in multiplayer gaming, and exercising regularly reported higher social support. Social support correlated negatively with depression, anxiety, and stress, whereas IGD correlated positively with these symptoms. Regression analyses identified academic performance, IGD, and intimacy as predictors of depression; family satisfaction as a predictor of anxiety; and family satisfaction and academic performance as protective factors against stress. Conclusions: Findings highlight the interplay of behavioral, social, and academic factors influencing students’ mental health. Effective interventions should reinforce family support and students’ academic self-efficacy, alongside fostering responsible gaming practices. Limitations include cross-sectional design and self-reported measures, indicating the need for longitudinal studies.

1. Introduction

University students’ mental health has become an increasing concern, with elevated levels of anxiety, stress, and depression observed in this population [1,2]. Among the associated factors, the intensive use of digital games and related behavioral disorder—Internet Gaming Disorder (IGD) has garnered attention. IGD is officially recognized by the World Health Organization in the 11th edition of the International Classification of Diseases (ICD-11) [3] and is listed as a condition requiring further research in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) [4].

IGD is defined as persistent or recurrent gaming behavior, in which gaming takes priority over other daily activities and continues despite negative consequences [5]. Research has consistently shown strong associations between IGD and mental health issues such as anxiety, stress, and depression in young adults [6,7,8]. A recent meta-analytic review confirmed that problematic internet use and IGD are moderately correlated with depressive and anxiety symptoms, lower well-being, and social difficulties, highlighting the importance of investigating these relationships in diverse student populations [9]. Studies in several cultural contexts, including Spain, Egypt, Jordan, and Macao, have similarly documented significant relationships between IGD and psychological distress among university students [6,10,11]. Cross-sectional international studies further support the link between IGD and poor mental health outcomes, including insomnia, decreased life satisfaction, and overall psychological distress [12,13]. Moreover, longitudinal evidence suggests bidirectional relationships, where pre-existing mental health problems may predict the development of IGD over time [7,14]. These findings indicate that IGD is both a consequence and a potential predictor of mental health difficulties, emphasizing the complexity of these associations.

Video games offer engaging forms of entertainment and can serve as an escape from daily stress, providing short-term relief from anxiety and fostering cognitive abilities such as strategic reasoning and problem-solving [11,15]. However, excessive or inappropriate use may lead to negative outcomes such as dependence, social isolation, and sleep disturbances [6,10,15]. Striking a healthy balance in gaming habits is essential: regular breaks, social interaction, and diverse leisure activities have been recommended to promote mental well-being and prevent problematic use [10,15].

Social support has also been identified as a key protective factor that can mediate or buffer the adverse psychological effects of IGD [10,16]. For example, Malak et al. [10] found that social support significantly reduced the negative psychological symptoms linked to IGD among university students. Additional evidence reveals that offline social support, including family and peer interactions, has a stronger protective effect than online social interactions, reducing anxiety, stress, and depressive symptoms in students with IGD [16].

Contextual and personal factors such as satisfaction with family and peer relationships, engagement in social activities, gender, age, housing conditions, academic status (student vs. working student), type of degree (undergraduate vs. postgraduate), academic year, and perceived academic performance and effort are all variables shown to affect mental health among higher education students [2,17,18,19]. A Recent Portuguese study emphasizes the role of housing conditions, peer relationships, and academic stress in shaping mental health and vulnerability to IGD, reinforcing the need to examine these factors in culturally specific contexts [20].

Despite the increasing global research on the relationship between video gaming, IGD, and psychological health, this topic remains underexplored in the Portuguese context. In Portugal, research has shown that problematic internet use correlates with poor mental health outcomes in university students [21], and recent studies highlight the relationship between IGD and impulsivity or inhibitory control deficits in Portuguese young adults [14].

Given this background, the present study aims to investigate the predictors of anxiety, stress, and depression among Portuguese university students. This study aims to provide a deeper understanding of the factors that shape mental health in this population, contributing to international literature from a culturally specific perspective.

2. Materials and Methods

2.1. Participants and Procedure

The target population for this study comprised 91 higher education students in Portugal who play video games. A convenience sampling strategy was employed, recruiting voluntary participants through the dissemination of an online questionnaire via social media platforms, student mailing lists, and Discord channels. This non-probability sampling method selected participants based on their easy accessibility and willingness to participate, rather than random selection, allowing for rapid recruitment but limiting the statistical representativeness of the sample. Moreover, the sample was predominantly male (80.5%), which further constrains the generalizability of the findings to female students and may skew results regarding gender-related differences. The online questionnaire included a brief description of the study’s objectives and collected sociodemographic data: gender, age, marital status, cohabitation during the academic term (with parents or with others) type of institution attended (public/private; polytechnic/university), type of course (bachelor’s, master’s, or doctoral degree), year of study, enrollment status (full-time student or working student), and self-perceived academic commitment and performance rated on a scale from 1 [mediocre] to 5 [excellent]. Additional information was gathered on physical exercise habits, average daily sleep duration, and, concerning gaming-related variables, mode of play (single-player or multiplayer), usual gaming companions, motivations for playing, game genre, frequency of play, interaction with other players, and the average number of days and hours spent gaming per week, among other variables. Data collection took place in December 2023.

2.2. Measures

The Social Support Satisfaction Scale evaluates an individual’s perception regarding both the quality and quantity of social support they receive across various contexts, such as family, friends, and other support networks. It addresses multiple dimensions of social support, including emotional, instrumental, informational, and evaluative aspects, thus providing a comprehensive understanding of how supported individuals feel in their social environments. This assessment is crucial for examining the relationship between social support and well-being, as it enables the identification of specific support needs within populations or individuals and informs psychosocial interventions designed to improve mental health and quality of life. The instrument consists of 15 items, each rated on a five-point Likert scale ranging from “strongly agree” (A), “somewhat agree” (B), “neither agree nor disagree” (C), “somewhat disagree” (D), to “strongly disagree” (E). In addition to an overall score, the scale includes four subscales: Friend Satisfaction (α = 0.849), which measures contentment with friendships; Intimacy (α = 0.756), assessing perception of close social support; Family Satisfaction (α = 0.823), evaluating satisfaction with family support; and Social Activities (α = 0.855), which gauges satisfaction with participation in social engagements [22]. These Cronbach’s alpha coefficients demonstrate good to excellent internal consistency, indicating that the subscales reliably measure their intended constructs and supporting the robustness of the instrument in this study’s context.

DASS-21 is a widely used psychometric instrument designed to assess the levels of depression, anxiety, and stress in individuals, thereby characterizing their psychological state. The Depression subscale (α = 0.86) measures symptoms such as dysphoria, inertia, hopelessness, devaluation of life, lack of interest or engagement, self-criticism, and anhedonia. The Anxiety subscale (α = 0.757) evaluates autonomic nervous system activation, musculoskeletal tension, situational anxiety, and subjective anxiety experiences. The Stress subscale (α = 0.839) assesses difficulties relaxing, nervous arousal, agitation, irritability, exaggerated responses, and impatience. Each of the three subscales consists of seven items that describe negative emotional symptoms experienced “over the past week.” Responses are scored on a four-point Likert scale from 0 (“Did not apply at all”) to 3 (“Applied very much or most of the time”). Scores for each subscale range from 0 to 21, with higher scores indicating more severe negative emotional states [23]. The internal consistency values obtained in this study confirm that the DASS-21 subscales reliably measure their intended constructs, supporting the robustness of the instrument in this population.

The Internet Gaming Disorder Scale-IGDS9-SF is a nine-item questionnaire designed to measure the severity of video game addiction and its impact, capturing both online and offline gaming behavior over the previous year. Participants respond using a five-point Likert scale ranging from “Never” (1) to “Almost Always” (5). Total scores can range from 9 to 45, with higher values indicating more severe symptoms of gaming disorder. The scale’s developers recommend a cutoff score of 36 to identify individuals with potential gaming dependency, although this threshold lacks definitive empirical validation. Additionally, the presence of five responses marked as “Almost Always” may be indicative of a more accurate diagnosis [24]. In this study, the scale demonstrated good internal consistency (α = 0.849), supporting its reliability for assessing problematic gaming behaviors in the target population.

2.3. Statistical Analysis

Participant sociodemographic, academic, and gaming-related data were summarized using means and standard deviations or percentages, depending on the variable type. Normality of the data was assessed using the Kolmogorov–Smirnov and Shapiro–Wilk tests. When appropriate, non-parametric tests—specifically, the Mann–Whitney U and Kruskal–Wallis tests—were employed to examine group differences. Spearman’s rank correlation coefficients were calculated to explore relationships among variables. Furthermore, multiple linear regression models (stepwise method) were conducted to identify significant predictors of mental health outcomes.

Statistical analysis was performed using SPSS version 29.0, with a significance level set at 0.05.

3. Results

A total of 91 students participated in the study, with ages ranging from 18 to 36 years and an average age of 22.71 years (SD = 3.72). Most respondents were male (80.5%) and single (93.4%), while 57.1% reported living with their parents, and 38.5% exercised 2 to 3 times a week. Regarding the educational context, 60.4% were undergraduate students, and 40% attended public polytechnic institutions. Nearly half (46.6%) enrolled in their second year of study, and 71.4% were full-time students.

Concerning academic engagement, 75.9% rated their level of commitment as good or above. Academic performance was considered above average or better by 73.5% of students. Students averaged 6.8 h of sleep per night (SD = 0.9).

Multiplayer gaming was common, with 63.7% identifying as multiplayer gamers; among these, 47.3% played with real-world friends. The predominant reasons for gaming were entertainment (38.6%) and relaxation (26.4%). On average, students played four to five days per week, spending 2.27 h daily during weekdays (SD = 1.56) and 4.64 h on weekends (SD = 2.64).

3.1. Relationship Between Sociodemographic, Academic Characteristics, and Game Factors with Social Support and Mental Health

This section examines the relationship between sociodemographic and academic characteristics, as well as game factors, and their impact on levels of social support and mental health among higher education students. This is a central topic in educational and social psychology, as multiple studies suggest that social support can serve as a protective factor against the challenges faced during academic journeys.

The results in

Table 1. Comparison of scores between sociodemographic groups with social support and mental health: means, standard deviations, Mann–Whitney U, and Kruskal–Wallis test results.

			IGD	Friend Satisfaction	Intimacy	Family Satisfaction	Social Activities	Depression	Anxiety	Stress
		n (%)	M ± SD
Gender	Female	17 (19.5)	14.41 ± 6.26	19.41 ± 4.53	14.53 ± 3.68	11.35 ± 3.43	9.24 ± 3.63	3.82 ± 3.99	3.41 ± 4.56	5.12 ± 4.40
	Male	70 (80.5)	13.67 ± 4.81	20.73 ± 3.90	15.21 ± 3.83	11.76 ± 2.60	9.97 ± 3.71	3.43 ± 3.70	1.61 ± 2.17	3.57 ± 3.74
	p *		0.931	0.254	0.546	0.854	0.397	0.754	0.292	0.176
Age (y)	≤22	55 (60.4)	13.20 ± 4.01	20.80 ± 3.81	14.69 ± 3.66	11.51 ± 2.76	9.91 ±3.62	3.45 ± 3.56	2.22 ± 3.26	4.15 ± 3.91
	>22	36 (39.6)	14.86 ± 6.51	19.56 ± 4.31	15.33 ± 3.95	11.61 ± 2.84	9.33 ±3.83	3.58 ± 4.07	1.56 ± 1.89	3.39 ± 3.83
	p *		0.759	0.182	0.410	0.863	0.470	0.873	0.885	0.345
Coh	WP	52 (57.1)	13.98 ± 5.32	21.67 ± 2.98	15.58 ± 3.68	11.75 ± 2.69	10.65 ± 3.54	3.38 ± 3.75	1.83 ± 2.59	3.40 ± 3.45
	WO	39 (42.9)	13.69 ± 5.04	18.49 ± 4.57	14.10 ± 3.76	11.28 ± 2.91	8.38 ± 3.54	3.67 ± 3.79	2.13 ± 3.11	4.44 ± 4.35
	p *		0.884	0.002	0.046	0.476	0.004	0.903	0.910	0.387
Physical exercise	RN	26 (28.6)	13.73 ± 4.73	19.31 ± 4.33	14.81 ± 3.81	11.38 ± 2.87	8.85 ± 3.60	3.54 ± 2.76	2.42 ± 3.55	4.62 ± 4.09
	OW	14 (15.4)	14.29 ± 5.74	18.21 ± 4.53	14.07 ± 4.18	10.57 ± 3.30	7.79 ± 3.14	2.93 ± 3.56	2.00 ± 2.88	2.93 ± 3.17
	TW	35 (38.5)	13.69 ± 4.50	21.43 ± 3.43	15.17 ± 3.66	11.89 ± 2.76	10.51 ± 3.46	3.77 ± 3.93	1.94 ± 2.50	3.91 ± 4.07
	D	16 (17.6)	14.06 ± 6.99	21.31 ± 3.52	15.44 ± 3.79	11.94 ± 2.11	10.88 ± 4.13	3.38 ± 5.00	1.19 ± 2.01	3.25 ± 3.71
	p **		0.907	0.055	0.795	0.555	0.038	0.589	0.578	0.553

M—Mean; SD—Standard Deviation; *—Mann–Whitney U Test; **—Kruskal–Wallis Test; Coh—Cohabitation; WP—With parents; WO—With others; RN—Rarely/never; OW—Once a week; TW—2 to 3 times a week; D—Daily.

show that no statistically significant differences were found in social support and mental health regarding gender and age. On the other hand, students living with their parents report higher levels of friend satisfaction (21.67 vs. 18.49; p = 0.002), intimacy (15.58 vs. 14.10; p = 0.046), and social activities (10.65 vs. 8.38; p = 0.004). The results also show differences in social activities compared to physical activity (p = 0.038). The results obtained through multiple comparison tests show that students who exercise only once a week have lower levels of satisfaction with social activities compared to those who exercise 2 to 3 times a week (7.79 vs. 10.51; p = 0.024) and those who exercise daily (7.79 vs. 10.88; p = 0.017).

Table 2. Comparison of scores between academic characteristics with social support and mental health: means, standard deviations, Mann–Whitney U, and Kruskal–Wallis test results.

			IGD	Friend Satisfaction	Intimacy	Family Satisfaction	Social Activities	Depression	Anxiety	Stress
		n (%)	M ± SD
Course type	Degree	55 (60.4)	13.71 ± 5.08	20.89 ± 3.86	14.60 ± 3.93	11.56 ± 2.85	9.84 ± 3.89	3.85 ± 4.24	2.27 ± 3.25	4.22 ± 4.24
	Master	26 (28.6)	14.77 ± 5.74	20.04 ± 3.83	15.58 ± 3.38	11.35 ± 2.87	9.88 ± 3.50	3.31 ± 3.07	1.85 ± 2.07	3.31 ± 3.22
	p *		0.647	0.316	0.283	0.702	0.923	0.810	0.875	0.512
Status	Student	65 (71.4)	13.98 ± 5.35	20.38 ± 4.06	14.82 ± 3.79	11.38 ± 2.93	9.69 ± 3.73	3.66 ± 3.95	2.11 ± 3.02	4.14 ± 3.80
	Working student	26 (28.6)	13.54 ± 4.81	20.12 ± 4.05	15.27 ± 3.77	11.96 ± 2.36	9.65 ± 3.69	3.12 ± 3.22	1.58 ± 2.19	3.12 ± 4.04
	p *		0.700	0.711	0.720	0.509	0.888	0.517	0.582	0.112
Type of institution	Public polytechnic	36 (40.0)	13.31 ± 5.35	20.94 ± 4.06	14.56 ± 4.07	11.61 ± 2.80	10.56 ± 3.57	4.06 ± 4.63	1.64 ± 2.66	3.92 ± 4.24
	Public university	26 (28.9)	14.96 ± 5.16	19.96 ± 3.61	15.73 ± 3.19	11.15 ± 2.88	9.42 ± 3.64	3.04 ± 3.14	1.73 ± 2.01	3.73 ± 3.53
	Private	28 (31.1)	13.64 ± 5.07	19.71 ± 4.44	14.64 ± 3.91	11.75 ± 2.73	8.64 ± 3.72	3.04 ± 2.81	2.61 ± 3.56	3.93 ± 3.89
	p **		0.248	0.352	0.516	0.699	0.134	0.855	0.430	0.970
Curricular year	1st	20 (22.7)	13.85 ± 5.75	21.15 ± 4.06	16.75 ± 2.73	12.10 ± 2.57	10.55 ± 4.08	3.50 ± 4.56	2.15 ± 3.51	3.75 ± 5.37
	2nd	41 (46.6)	14.37 ± 5.61	20.59 ± 3.94	14.80 ± 3.68	11.56 ± 2.76	9.85 ± 3.50	3.78 ± 3.55	2.00 ± 2.56	3.76 ± 3.25
	3rd	27 (30.7)	13.11 ± 3.91	19.96 ± 3.88	14.15 ± 4.13	11.07 ± 3.03	9.26 ± 3.62	3.00 ± 3.63	1.93 ± 2.80	3.89 ± 3.65
	p **		0.144	0.268	0.273	0.434	0.542	0.817	0.804	0.878

M—Mean; SD—Standard Deviation; *—Mann–Whitney U Test; **—Kruskal–Wallis Test.

and

Table 3. Comparison of scores between game factors with social support and mental health: means, standard deviations, Mann–Whitney U, and Kruskal–Wallis test results.

			IGD	Friend Satisfaction	Intimacy	Family Satisfaction	Social Activities	Depression	Anxiety	Stress
		n (%)	M ± SD
Game mode	Single-player	33 (36.3)	14.15 ± 5.78	18.73 ± 4.03	13.76 ± 3.78	11.00 ± 2.89	8.42 ± 3.11	4.18 ± 4.18	2.21 ± 3.03	2.21 ± 3.03
	Multiplayer	58 (63.7)	13.69 ± 4.85	21.21 ± 3.79	15.62 ± 3.62	11.86 ± 2.68	10.40 ± 3.83	3.12 ± 3.46	1.81 ± 2.69	1.81 ± 2.69
	p *		0.940	0.003	0.021	0.163	0.012	0.331	0.759	0.367
Correlation Spearman Coefficient	Days of play per week		0.193	0.205	0.065	0.075	0.124	0.051	−0.069	−0.089
	Average hours of play per week		0.169	0.043	−0.030	0.119	0.033	0.058	0.229 *	0.129
	Average hours of play on weekends		0.129	0.225 *	0.060	0.026	0.147	0.004	0.041	−0.030
	Average consecutive hours of play		0.174	−0.022	0.001	0.106	−0.053	0.041	0.043	0.042

M—Mean; SD—Standard Deviation; *—Mann–Whitney U Test.

show the results of the comparison of scores between academic factors and game factors with social support and mental health, respectively. Only statistically significant differences are found in the game mode. Students who play multiplayer games demonstrate greater friend satisfaction (18.73 vs. 21.21; p = 0.003), intimacy (13.76 vs. 15.62; p = 0.021), and social activities (8.42 vs. 10.40; p = 0.012).

These results suggest that, although classic factors such as gender or educational context are not relevant for social support and mental health in this sample, the living situation and involvement in online social activities have a significant impact on students’ relational and social satisfaction. This aligns with evidence highlighting the role of support networks and social context in maintaining mental health and academic well-being.

3.2. Intercorrelation Between Mental Health Components, Social Support, and Internet Gaming Disorder (IGD)

This section examines the relationships between the three core components of mental health—depression, anxiety, and stress—and three key psychosocial dimensions: social support (operationalized through satisfaction with friendships, family, and social activities) and Internet Gaming Disorder (IGD). Descriptive results for the IGDS9-SF revealed a mean score of 13.86 (SD = 5.18), with a median of 12, a minimum of 9, and a maximum of 33, indicating generally low to moderate levels of problematic gaming behaviors within the sample. The inclusion of these measures allows for a nuanced analysis of how emotional well-being interacts with perceived social connectedness and potential problematic gaming behaviors, which are increasingly relevant in contemporary student populations.

Correlation analysis reveal consistent patterns (

Table 4. IGD, Social Support, and Mental Health: descriptive statistics and Spearman coefficient correlations.

	M ± SD	1	2	3	4	5	6	7	8
1. IGD	13.86 ± 5.18	1	−0.219 *	−0.228 *	−0.247 *	−0.292 **	0.457 **	0.272 **	0.407 **
2. Friend Satisfaction	20.31 ± 4.04		1	0.634 **	0.460 **	0.764 **	−0.290 **	−0.209 *	−0.272 **
3. Intimacy	14.95 ±3.77			1	0.389 **	0.543 **	−0.398 **	−0.136	−0.339 **
4. Family Satisfaction	11.55 ± 2.78				1	0.380 **	−0.408 **	−0.188	−0.384 **
5. Social Activities	9.68 ± 3.70					1	−0.268 *	−0.265 *	−0.294 **
6. Depression	3.51 ± 3.75						1	0.486 **	0.730 **
7. Anxiety	1.96 ± 2.81							1	0.592 **
8. Stress	3.85 ± 3.88								1

M: Mean; SD: Standard deviation; * p < 0.05; ** p < 0.01.

): depression, anxiety, and stress are negatively correlated with satisfaction with friendships (r = −0.290; r = −0.200; r = −0.272, respectively). Depression, anxiety, and stress are also negatively correlated with satisfaction with social activities (r = −0.268; r = −0.265; r = −0.294, respectively). In contrast, depression, anxiety, and stress are positively correlated with IGD (r = 0.457; r = 0.272; r = 0.407, respectively).

These findings suggest that higher levels of psychological distress are associated with lower perceived social satisfaction and greater tendencies toward problematic gaming behavior. This pattern aligns with literature indicating that social isolation and maladaptive coping strategies, such as excessive gaming, can co-occur with poorer mental health outcomes. Conversely, stronger social ties appear to be protective, correlating with reduced symptoms of depression, anxiety, and stress.

The results suggest that lower levels of social support and higher problematic involvement in online gaming (IGD) are associated with increased symptoms of depression, anxiety, and stress, highlighting the protective role of support networks and the potential risk of excessive gaming behaviors for mental health.

3.3. Predictors of Mental Health Components: Depression, Anxiety, and Stress

In this subsection, we present the results of the predictive analysis conducted through three multiple linear regression models (stepwise method), each corresponding to one of the assessed mental health components—depression, anxiety, and stress—considered separately as dependent variables. In each model, the independent variables included: gender, age, type of cohabitation during the academic term (living with parents or with others), the three dimensions of social support (friends satisfaction, family satisfaction, social activities, and intimacy), the Internet Gaming Disorder (IGD) score, perceived academic performance, perceived academic commitment, average hours of sleep per day, age at which participants began playing video games, and average daily gaming time during weekdays and weekends. This approach aims to identify which of these factors make a statistically significant contribution to the variation in symptom levels for each mental health dimension in the sample studied.

The assumptions underlying the applicability of the multiple linear regression models (MLRMs) were thoroughly assessed to ensure the validity of the analysis. Specifically, tests examining the linearity of relationships between independent and dependent variables, the normality of residuals, the independence of errors, and the absence of multicollinearity among predictors were conducted. These evaluations confirmed that the fundamental assumptions were met, supporting the robustness and reliability of the regression results presented.

3.3.1. Predictors of Depression

The multiple linear regression analysis identified academic performance, Internet Gaming Disorder (IGD), and intimacy as significant predictors of depressive symptomatology among the participants (

Table 5. Regression Coefficients for Predictors of Depression.

	B	t	p	VIF
Constant	9.169	4.306	<0.001
Academic performance	−1.534	−3.703	<0.001	1.080
IGD	0.213	3.203	0.002	1.065
Intimacy	−0.254	−2.839	0.006	1.069
R2 (adj) = 35.9%, Durbin–Watson = 2.138

VIF = Variance Inflation Factor.

). Specifically, higher self-reported academic performance was associated with lower levels of depression (B = −1.534; p < 0.001), suggesting a protective effect of perceived academic competence on mental health. In contrast, higher IGD scores predicted increased depressive symptoms (B = 0.213; p = 0.002), indicating that problematic gaming behavior may constitute a risk factor for depression in this population. Additionally, greater intimacy within the dimension of social support was linked to reduced depression (B = − 0.254; p = 0.006), highlighting the relevance of close interpersonal connections for psychological well-being. The model accounted for 35.9% of the variance in depression scores.

Overall, these results underscore the multifactorial nature of depression and emphasize the importance of academic achievement, digital behaviors, and intimate social networks in mitigating depressive symptoms among higher education students.

3.3.2. Predictors of Anxiety

The results in

Table 6. Regression Coefficients for Predictors of Anxiety.

	B	t	p	VIF
Constant	4.749	4.012	<0.001
Family Satisfaction	−0.245	−2.484	0.015	1.000
R2 (adj) = 6.1%, Durbin–Watson = 2.155

VIF = Variance Inflation Factor.

revealed that family satisfaction was the sole significant predictor of anxiety symptoms within the sample. Higher family satisfaction relationships were associated with lower anxiety scores (B = − 0.245; p = 0.015), underscoring the protective impact of positive familial support on students’ psychological well-being. The model explained 6.1% of the variance in anxiety levels.

These findings suggest that among the variables tested, the perceived quality of family support stands out as a key buffer against anxiety in this cohort, even though the proportion of variance explained remains modest. This highlights the unique role of family relationships relative to other sociodemographic, academic, and behavioral factors in mitigating anxiety symptoms among higher education students.

3.3.3. Predictors of Stress

The multiple linear regression model for stress identified two significant predictors: family satisfaction and perceived academic performance (

Table 7. Regression Coefficients for Predictors of Stress.

	B	t	p	VIF
Constant	12.503	6.739	<0.001
Family Satisfaction	−0.431	−2.901	0.005	1.180
Academic performance	−1.164	−2.347	0.021	1.180
R2 (adj) = 20.6%, Durbin–Watson = 2.185

VIF = Variance Inflation Factor.

). Greater satisfaction with family relationships was associated with lower levels of stress (B = −0.431; p = 0.005), underscoring the importance of positive familial support in buffering students against stress. Additionally, higher perceived academic performance independently predicted reduced stress symptoms (B = −1.164; p = 0.021), highlighting the protective role of academic self-efficacy in this context. Together, these variables accounted for 20.6% of the variance in stress.

These findings emphasize the combined relevance of supportive family environments and positive academic self-perceptions in mitigating stress among higher education students.

4. Discussion

The present study provides comprehensive insights into the multifaceted relationships among video gaming behaviors, social support, academic factors, and mental health outcomes in Portuguese higher education students. Our findings indicate no statistically significant differences in social support and mental health symptoms concerning gender and age, which concurs with previous research reporting minimal demographic influence on these variables in university populations [1,17]. This suggests that other psychosocial and behavioral factors may play more critical roles in influencing mental health in this context.

Significantly, students living with their parents demonstrated higher levels of friend satisfaction, intimacy, and participation in social activities compared to those living with others. These results emphasize the protective benefits of parental cohabitation on social connectedness and are consistent with studies underscoring the importance of familial environments for psychological resilience and social support networks [10,16]. The positive association between frequent physical activity and increased social engagement corroborates established evidence about exercise’s beneficial effects on social well-being and mood enhancement [11].

Furthermore, the analysis revealed that students engaging in multiplayer gaming exhibited greater friend satisfaction, intimacy, and exposure to social activities, supporting the perspective that online multiplayer games can foster social relationships and provide meaningful opportunities for interaction [11]. These findings align with the emerging recognition of video games as potential facilitators of social capital and emotional support within digital contexts.

Correlation analysis revealed robust negative associations between depressive, anxious, and stress symptoms with friendships and social activities, reaffirming the vital buffering role of social support against psychological distress [10,18]. In contrast, positive correlations between Internet Gaming Disorder (IGD) scores and symptoms of depression, anxiety, and stress align with a growing body of literature identifying problematic gaming as a significant risk factor for mental health problems [5,7,8]. These findings further highlight the clinical relevance of IGD, as classified by the World Health Organization (WHO, 2018) [3] and the American Psychiatric Association (APA, 2013) [4], in associating excessive gaming behaviors with adverse psychological outcomes.

The multiple linear regression results delineated differential predictors for each mental health dimension. For depression, academic performance, IGD, and intimacy were significant predictors, emphasizing that lower perceived academic success and problematic gaming increase vulnerability to depressive symptoms, whereas intimacy serves as a protective factor. These results resonate with prior studies linking academic difficulties and social connectedness to depression in student populations [20,21]. Family satisfaction emerged as the sole predictor of anxiety, highlighting the unique role of familial support in alleviating anxious symptomatology [1]. Regarding stress, both family satisfaction and perceived academic performance were identified as significant protective factors, supporting existing evidence regarding the dual importance of close family relationships and academic self-efficacy in stress mitigation [1,2].

4.1. Limitations

The cross-sectional design limits causal inference, and reliance on self-reported data introduces potential biases related to social desirability and accuracy, particularly concerning sensitive mental health and gaming behaviors. Additionally, the sample, which was predominantly male and restricted to Portuguese higher education students, may limit the generalizability of findings to female students and to populations in different cultural or educational contexts. Future studies would benefit from objective behavioral assessments and diverse, longitudinal cohorts.

4.2. Future Research Directions

Further research should prioritize longitudinal designs to clarify causality and mechanisms linking problematic gaming, social support, academic factors, and mental health. Investigating moderate and mediating variables—such as personality traits, coping strategies, and digital literacy—can deepen insight into vulnerability and resilience. Evaluating the effectiveness of interventions, including digital and family-based approaches, will be essential. Expanding research across cultural groups and education levels will also strengthen the field’s applicability.

5. Conclusions

These findings highlight the complex and multifactorial nature of mental health among university students, where social, academic, and behavioral factors interact intricately. Importantly, the sample was predominantly male, which likely influenced the results, given that Internet Gaming Disorder (IGD) is more prevalent among men, and gaming patterns differ by sex. Therefore, these conclusions are primarily applicable to male students who engage in gaming.

The fact that only family satisfaction predicted anxiety may reflect the unique and consistent emotional support family provides, offering stability that other social supports might lack. Literature suggests that family cohesion serves as a primary buffer against anxiety in young adults, whereas peer or online social support can be more variable and less influential for anxiety regulation.

The association between multiplayer gaming and better social support is an intriguing finding. Multiplayer games often function as social platforms that facilitate communication, cooperation, and community-building, thus enhancing perceived social connectedness beyond in-person interactions. This suggests that gaming, when social in nature, can foster meaningful social support networks.

In terms of practical recommendations, these results emphasize the need for interdisciplinary interventions combining academic support, family involvement, and strategies to manage gaming behaviors. Programs promoting mental health should leverage family engagement and the positive social aspects of multiplayer gaming to enhance well-being while addressing risks linked to problematic gaming. Future longitudinal studies are essential to clarify causal relationships and temporal dynamics among these factors.