Exploring the Links Among Risky Substance Use, Problematic Internet Use, and Academic Outcomes in University Freshmen: The Role of Mediating Factors
Department of Clinical and Experimental Sciences, University of Brescia, Viale Europa 11, 25123 Brescia, Italy; chiara.buizza@unibs.it (C.B.); giuseppe.rainieri@unibs.it (G.R.); alberto.ghilardi@unibs.it (A.G.)
Research and Clinical Trials Unit, Fondazione Poliambulanza Istituto Ospedaliero, Via Bissolati 57, 25124 Brescia, Italy; clarissa.ferrari@poliambulanza.it
Abstract
Background: Alcohol and substance use among young people is a well-documented public health concern, and is particularly prevalent in college populations. Problematic internet use is also an emerging issue, with potential negative effects on academic achievement. University dropout remains a critical challenge, especially among freshmen, with research highlighting the role of academic engagement factors such as motivation, self-efficacy, and university connectedness in students’ academic trajectories. Methods: This study explored the relationships among risky substance use, problematic internet use, academic engagement factors, and academic outcomes, identifying potential mediators. Freshmen from an Italian university were invited to complete an online survey assessing these variables. The study defined two academic outcomes: (i) academic performance (Grade Point Average, GPA) and (ii) dropout intentions. Spearman’s rho coefficients and multiple linear regression models examined the associations among risky substance/internet use, academic engagement factors, and academic outcomes. Mediation analyses assessed whether academic engagement variables mediated the relationship between risky substance/internet use and academic outcomes. Results: The results showed that only problematic internet use was significantly associated with GPA, with self-efficacy and lack of motivation fully mediating this relationship. Regarding dropout intentions, problematic internet use and the risky use of alcohol, cannabis, and sedatives were directly and positively associated with dropout intentions. Several motivation subscales, self-efficacy, and university connectedness mediated these relationships. Conclusions: These findings highlight the role of academic engagement factors in mitigating the impact of risky behaviors on students’ academic trajectories, emphasizing the need for targeted prevention and intervention strategies.
Untitled section
Keywords: freshmen, alcohol, substance use, internet use, motivation, engagement, self-efficacy, academic dropout, academic achievement
Article notes
Untitled section
Received 2025 Mar 28; Revised 2025 May 30; Accepted 2025 Jun 4; Collection date 2025 Jun.
1. Introduction
Alcohol and substance use among young people is a significant and well-documented public health concern, and is particularly prevalent in college students (Arria et al., 2008; White & Hingson, 2013). In Italy, 2020 data indicate that approximately 850,000 young adults (aged 18–25) engage in risky alcohol consumption (Italian Ministry of Health, 2022), while a 2023 national survey reported the past-year use of psychoactive substances (mainly cannabis) to be 31.8% in men and 22.5% in women in this age group (Italian Presidency of the Council of Ministers, 2023). In addition to substance use, young people (especially college students) are also at an increased risk of problematic mobile phone and internet use (Long et al., 2016; Younes et al., 2016). Recent meta-analyses and systematic reviews have demonstrated that college students exhibiting mobile phone or internet addiction are more likely to experience heightened levels of anxiety and depression (Y. Li et al., 2020; Melca et al., 2023). These behaviors have also been linked to poor academic outcomes (Javdi & Mohammadi, 2025; Kassaw & Demareva, 2023; Meng et al., 2025). In particular, excessive internet use has been linked to lower academic self-efficacy and reduced motivation, which can negatively impact academic achievement (Chavez-Yacolca et al., 2025; Şen, 2025).
School dropout is another critical issue among young adults. In higher education, this is particularly true for freshmen, as the first year is a crucial period in determining university persistence or attrition (Murai & Nakayama, 2008). Dropout is a multifaceted phenomenon influenced by various academic engagement factors, including motivation, self-efficacy, and a sense of connectedness with the university (Berhanu & Sewagegn, 2024; Biasi et al., 2017; Díaz Mujica et al., 2019; Fior et al., 2022; Marôco et al., 2020; Robbins et al., 2004).
Given the well-established effects of alcohol and substance use on cognitive functioning (Gould, 2010; Zeigler et al., 2005), recent studies have explored whether these behaviors contribute to academic underperformance and dropout risk. Several studies support this link, showing that substance use is associated with lower academic achievement and an increased likelihood of dropping out (Heradstveit et al., 2024; Hjarnaa et al., 2023; Trusty et al., 2025; Valkov, 2018). However, the evidence is not entirely consistent. For instance, Andersson et al. (2009) found no significant differences in alcohol use between students who dropped out and those who persisted in their studies. More broadly, recent findings confirm that alcohol and substance use, particularly when frequent or severe, is negatively associated with academic performance (Ache Akua & Samek, 2023; Berchtold et al., 2024; Karabacak et al., 2024; Lederer et al., 2024; López-Moreno et al., 2021; Meda et al., 2017; Mota et al., 2010; Rees, 2019). Yet, some discrepancies remain; for example, Reid et al. (2024) reported inconsistent associations between alcohol use and Grade Point Average (GPA).
A similar pattern has emerged regarding problematic internet and smartphone use. These behaviors have been linked to reduced academic self-efficacy and intrinsic motivation (Chavez-Yacolca et al., 2025; Şen, 2025), as well as to poorer academic outcomes overall (Akhter, 2013; C. Li et al., 2023). Nonetheless, research in this area is still relatively limited and warrants further investigation. Importantly, most existing studies have examined substance and internet misuse in isolation, without considering their potential combined impact on students’ academic engagement. Moreover, little is known about the psychological mechanisms underlying these associations, especially in early university stages.
The effects of substance and internet misuse extend beyond cognitive impairments, as some studies suggest that these behaviors may also impact motivation and self-efficacy (Andrews & Duncan, 1997; Juthamanee & Gunawan, 2021; Phillips et al., 2015). However, the extent to which academic engagement factors, such as self-efficacy, motivation, and university connectedness, mediate the relationship between substance/internet misuse and academic performance has not been thoroughly explored. Addressing this gap is crucial, as identifying potential targets may inform early intervention strategies aimed at enhancing academic success and reducing dropout rates among university freshmen.
The present study aimed to address this gap by examining the interrelations among risky substance use and problematic internet use, academic engagement factors, and academic achievement, with a particular focus on identifying potential mediators. Based on the available literature, we hypothesized the following:
H1. We hypothesized an association among academic engagement factors and academic achievement.
H2. We hypothesized an association among risky substance use, problematic internet use, and academic achievement.
H3. We hypothesized that academic engagement factors would mediate the association among risky substance/internet use and academic achievement.
The findings of this study may contribute to the development of targeted interventions designed to support freshmen in achieving academic success and reducing their likelihood of dropping out.
2. Materials and Methods
2.1. Study Design
This cross-sectional study was conducted in June 2022 and involved freshmen from a university in Northern Italy. In collaboration with the University Secretariat, all enrolled freshmen were invited via email to participate in a multidimensional online survey created with LimeSurvey (https://www.limesurvey.org, accessed on 3 May 2022), ensuring anonymous data collection. The email included a link to the survey and an informative document detailing the study. Participation was voluntary, and responses were anonymous. Informed consent was obtained online before survey access. The study adhered to the World Medical Association’s Declaration of Helsinki for research involving human subjects and received ethical approval from the Board of Directors of the University of Brescia (provision no. 330, 22 November 2021).
2.2. Participants
Out of the 3,756 freshmen invited to participate in the study, 1618 (43.1%) accessed the survey, and 721 (19.2%) completed it. Most participants were female (63.4%) and Italian (95%). The mean age of the sample was 20.83 years (SD = 3.83). Academic characteristics are reported in Table 1.
| Variable | N | % |
|---|---|---|
| Field of study | ||
| Medicine | 302 | 41.9 |
| Engineering | 157 | 21.8 |
| Economics | 180 | 25 |
| Law | 46 | 6.4 |
| Pharmacy | 23 | 3.2 |
| Agricultural | 13 | 1.8 |
| Employment status | ||
| Student | 473 | 65.6 |
| Working student | 248 | 34.4 |
| Mean | SD | |
| Grade Point Average 1 | 24.80 | 2.80 |
| Percentage of attended lessons | 78.67 | 27.85 |
| Mean number of hours spent studying per day | 3.90 | 2.03 |
2.3. Instruments
The online survey assessed socio-demographic characteristics, academic careers, and physical and mental well-being, including the following standardized instruments:
- World Health Organization’s Alcohol, Smoking, and Substance Involvement Screening Test v3.0 (ASSIST): this is a self-report questionnaire (Barreto et al., 2014) designed to detect and manage substance use. It includes eight questions covering ten substances: tobacco, alcohol, cannabis, cocaine, amphetamine-type stimulants, inhalants, sedatives, hallucinogens, opioids, and “other drugs”. An example item is “In your life, which of the following substances have you ever used?”. Each substance receives a score from 0 to 36, categorized as low, moderate, or high-risk use (with higher scores corresponding to higher risk). In this study, tobacco was excluded from the analyzed substances, primarily due to its distinct social context and less acute potential for immediate academic impairment compared to alcohol and illicit substances. In our sample, 17.9% of participants exhibited a moderate-to-high-risk pattern of use for at least one substance. Specifically, 11.5% reported moderate-to-high-risk alcohol use (cut-off score = 10), 7.1% reported moderate-to-high-risk cannabis use, and 3.2% reported a moderate-to-high-risk sedative use (cut-off score = 3 for both cannabis and sedatives). For all other substances, the proportion of moderate-to-high-risk users was below 0.5%.
- Internet Abusive Use Questionnaire (IAUQ, Calvo-Francés, 2016): this is a 12-item scale rated on a five-point Likert scale from 0 (“Totally disagree”) to 4 (“Totally agree”), assessing problematic internet use. An example item is “You lose sleep in order to stay online”. The total score ranges from 0 to 48, with the established cutoff of 24 or higher indicating problematic internet use. In this sample, the overall Cronbach’s alpha value was 0.91, and 10.7% of students scored above this cutoff.
- An adapted form of the Academic Motivation Scale (AMS, Biasi et al., 2017): this was developed based on the Self-Determination Theory from Vallerand et al. (1992), it assesses motivation through five subscales. Each subscale is composed of four items rated on an 11-point Likert scale ranging from 0 (“Not at all true”) to 10 (“Completely true”). The five subscales are (i) Amotivation (i.e., lack of motivation, example item: “I can’t see why I go to school and, frankly, I couldn’t care less”), (ii) External Regulation (i.e., motivation driven by external rewards, punishments, or demands, example item: “Because someone else expects me to”), (iii) Introjected Regulation (i.e., motivation driven by internal pressure like guilt or shame, example item: “To prove to myself that I am capable of completing my high-school degree”), (iv) Identified Regulation (i.e., motivation based on personal values and conscious choice, example item: “Because eventually it will enable me to enter the job market in a field that I like”), and (v) Intrinsic Regulation (i.e., motivation arising from inherent interest, enjoyment, and satisfaction in the academic experience, example item: “Because my studies allow me to continue to learn about many things that interest me”). The total score ranges from 0 to 40, with a higher score indicating a greater adherence to the construct represented by each subscale. In our sample, the overall Cronbach’s alpha values for the five subscales ranged from 0.78 to 0.94.
- Perceived School Self-Efficacy Scale (SASP, Biasi et al., 2017; Pastorelli & Picconi, 2001): this measures students’ perceptions of their ability to regulate and focus on their studies. It includes nine items rated on a five-point Likert scale ranging from 1 (“Not capable at all”) to 5 (“Fully capable”). An example item is “Focus on your studies without getting distracted”. The total score ranges from 9 to 45, with a higher score indicating a greater perceived level of self-efficacy. In this sample, the overall Cronbach’s alpha value was 0.88.
- University Connectedness Scale (UCS, Stallman & Shochet, 2008): this measures students’ perceived support and sense of belonging within their university. It consists of 18 items rated on a seven-point Likert scale, ranging from 1 (“Not at all”) to 7 (“All the time”). An example item is “Class sizes are so large that I feel like a number”. The total score ranges from 18 to 126, and higher scores indicate a stronger sense of connectedness and support. In our sample, the overall Cronbach’s alpha value was confirmed at 0.88.
- Freshmen’s dropout intentions were assessed using a composite score adapted from Hardre and Reeve (2003) for the Italian university context (Biasi et al., 2017). Students were asked how often they (1) Think about dropping out of college and pursuing something else; (2) Feel insecure about continuing their college studies year after year; (3) Consider the idea of discontinuing their college education; and (4) Intend to drop out of college. Each item was rated on a five-point Likert scale ranging from 1 (“Never”) to 5 (“Always”), with higher scores indicating greater dropout intentions. The dropout intention score was calculated as the mean of these four items (De Vincenzo, 2024).
Further details on the means and standard deviations of all assessment scale scores and academic outcomes are provided in Table 2.
| Questionnaire | Mean | SD | Scale Range |
|---|---|---|---|
| ASSIST alcohol score | 4.71 | 5.04 | 0–36 |
| ASSIST cannabis score | 0.84 | 3.33 | 0–36 |
| ASSIST sedative score | 0.43 | 2.65 | 0–36 |
| IAUQ total score | 12.43 | 9.42 | 0–48 |
| AMS—subscales’ scores | 0–40 | ||
| Amotivation | 6.21 | 7.67 | |
| External regulation | 5.99 | 5.69 | |
| Introjected regulation | 20.87 | 10.49 | |
| Identified regulation | 31.20 | 9.60 | |
| Intrinsic regulation | 30.70 | 8.21 | |
| SASP total score | 28.75 | 6.01 | 9–45 |
| UCS total score | 82.93 | 17.07 | 18–126 |
| Dropout Intentions | 2.18 | 1.01 | 1–5 |
2.4. Outcome Definition
This study aimed to evaluate academic engagement factors that could mediate the impact of substance and internet risky use on students’ academic trajectories. The following primary outcome measures were defined:
- Academic performance, measured as GPA (continuous variable).
- Dropout intentions, a continuous score ranging from 1 to 5, where higher values indicate greater dropout intentions.
2.5. Statistical Analysis
The statistical analyses were performed by R-statistical software (version 4.4.1, 2024 The R Foundation for Statistical Computing, Vienna, Austria) and the software Jamovi (the jamovi project version 2.4, 2025, https://www.jamovi.org, accessed on 13 March 2025). Descriptive statistics were computed for socio-demographic and academic variables, as well as the questionnaires’ scores, reporting percentage distributions for categorical variables and means with standard deviations (SD) for continuous variables. Based on the observed prevalence of risky substance use and problematic internet use (as detailed in the Instruments section), the following variables were selected as direct predictors of the two outcomes: risky alcohol use, risky cannabis use, risky sedative use, and problematic internet use. Spearman’s rho coefficients and multiple linear regression models assessed the correlations between continuous outcomes, substance/internet risky use, and academic engagement variables. A stepwise model selection procedure (carried out by the stepAIC function of the R-package MASS) was used to identify the best independent predictors for each outcome. Mediation analyses were conducted to examine whether academic engagement variables mediated the relationship between risky substance use and problematic internet use with academic outcomes, following Baron and Kenny’s (1986) framework (Figure S1, Supplementary Materials). The process of assessing mediation follows three steps:
- Step 1: Assess the association between the independent variable (i.e., risky substance use and problematic internet use) and the dependent variable (i.e., GPA and dropout intentions). This initial assessment establishes the c’ path, representing the direct effect of the independent variable (X) on the dependent variable (Y) before including the mediator (M, i.e., academic engagement factors).
- Step 2: Assess the association between X and M. This step determines the strength and direction of the a path, verifying that X has an effect on the proposed M.
- Step 3: Assess the association between M and Y while simultaneously controlling for X. This step evaluates the b path, confirming whether M significantly influences Y beyond the effect of X.
Based on the results of these assessments, particularly the effect of X on Y when controlling for M, the type of mediation can be determined. Full mediation occurs when, after controlling for M, X is no longer significantly associated with Y, indicating that the entire effect of X on Y is transmitted through M. Partial mediation occurs when the association between X and Y is significantly reduced but still statistically significant after controlling for M. This suggests that M accounts for part, but not all, of the relationship, leaving a residual direct effect of X on Y.
In our applied mediation analyses, we assessed whether academic engagement variables (potential mediators M) mediated the relationship between risky substance use/problematic internet use (X variables) and academic outcomes (Y variables). Each mediator—self-efficacy (SASP total score), motivation (AMS subscale scores), and university connectedness (UCS total score)—was tested in separate models to identify specific intervention targets aimed at improving academic success and reducing dropout intentions.
The associations of steps 1–3 were assessed by linear models (both univariable and multivariable linear models, with beta regression coefficients as effect sizes of the relations between X, M and Y). The level of statistical significance was set at p = 0.05.
3. Results
3.1. Correlation Analysis Among Primary Outcomes, Substance/Internet Risky Use, and Academic Engagement Variables
Regarding the first outcome, a significant positive correlation was found between problematic internet use and academic performance, as measured by GPA (p < 0.001). Additionally, problematic internet use was correlated with UCS, SASP, and all five AMS subscales, which were, in turn, correlated with GPA (see correlation matrix in Supplementary Materials, Table S1). Interestingly, no significant correlation was found between risky substance use and academic performance. These findings suggest that academic engagement variables (UCS, SASP, and AMS subscales) may act as potential mediators in the direct relationship between problematic internet use and academic performance.
Regarding the second outcome (dropout intentions), it was positively correlated (p < 0.001) with both problematic internet use and risky substance use, although Spearman’s rho coefficients were relatively low for alcohol, cannabis, and sedatives. Moreover, risky substance use was correlated with UCS, SASP and all five AMS subscales, which were, in turn, correlated with dropout intentions (see correlation matrix in Supplementary Materials, Table S2). These correlation analyses further suggest that academic engagement variables may mediate the relationship between substance/internet risky use and dropout intentions.
3.2. Mediation Models for Academic Performance
Univariable linear models were applied to assess the direct effect (Step 1 of the mediation model) of alcohol/substance risky use and problematic internet use on academic performance (GPA). Only problematic internet use was significantly associated with GPA, with a beta regression coefficient equal to −0.040 (p < 0.001). Additionally, the association of problematic internet use with the seven potential mediators (i.e., academic engagement variables) was examined using seven separate regression models (Step 2 of the mediation model). Significant negative correlations were found for UCS, SASP, and Identified regulation and Intrinsic regulation AMS subscales (p < 0.001 for all). Significant positive correlations were found for Amotivation, External regulation and Introjected regulation AMS subscales (p < 0.001 for all). Multivariable regression models (Step 3 of the mediation model) identified SASP and the Amotivation AMS subscale as full mediators of the relationship between problematic internet use and GPA, as the direct effect was no longer significant after adjusting for these mediators. The remaining five academic engagement variables acted as mediators, as the direct effect remained significant but was reduced in magnitude (Figure 1).
3.3. Mediation Models for Dropout Intentions
Similar mediation models were performed for the second outcome, namely dropout intentions. In addition to problematic internet use (Figure 2A), risky alcohol, cannabis, and sedative use was found to be directly and positively correlated to dropout intentions (Figure 2B–D). Regarding the relationships between problematic internet use and dropout intentions, and between risky alcohol use and dropout intentions, all seven mediators acted as partial mediators (Figure 2A,B). A different pattern emerged for the direct positive relationship between risky cannabis use and dropout intentions (Figure 2C): only the UCS, Amotivation AMS subscale, and External regulation AMS subscale acted as full mediators, while the remaining four variables did not satisfy the Step 2 condition of the mediation model. Notably, the UCS score exhibited the strongest mediating effect, significantly reducing the direct association between risky cannabis use and dropout intentions (beta of direct effect: 0.028, p < 0.05; beta after controlling for UCS: 0.013, p > 0.05). Finally, regarding the relationship between risky sedative use and dropout intentions (Figure 2D), five academic engagement variables acted as partial mediators, whereas the Identified regulation and Intrinsic regulation AMS subscales did not satisfy the mediation model criteria. Among the mediators, the Amotivation AMS subscale had the strongest effect in reducing the direct association between risky sedative use and dropout intentions (beta of direct effect: 0.055, p < 0.001; beta after controlling for Amotivation AMS subscale: 0.035, p = 0.007).
4. Discussion
This study aimed to examine the impact of risky substance and internet use on two academic outcomes: academic performance (measured by GPA) and dropout intentions. Additionally, we explored the mediating role of academic engagement factors in a large sample of Italian freshmen.
In line with our first hypothesis (H1), our results confirmed the association between academic engagement factors and academic achievement, highlighting the importance of university connectedness, motivation, and self-efficacy in fostering academic success and student retention (Berhanu & Sewagegn, 2024; Durán-Aponte & Elvira-Valdés, 2015; Sinval et al., 2025). Regarding our second hypothesis (H2), we found that not all anticipated associations between risky substance/internet use and academic outcomes were significant. Specifically, both risky substance use (alcohol, cannabis, and sedatives) and problematic internet use were positively associated with dropout intentions, while GPA was negatively associated with problematic internet use but not with risky substance use. These findings are partly consistent with prior research. Trusty et al. (2025) found that substance use was significantly associated with academic dropout among college students. However, the associations between substance use and academic performance have been more mixed: while Reid et al. (2024) reported no significant relationship between alcohol use and academic performance, Lederer et al. (2024) observed negative associations between both alcohol and cannabis use and GPA. Regarding problematic internet use, an Iranian study found that internet addiction was inversely related to academic performance, consistent with our findings (Javdi & Mohammadi, 2025).
These findings should also be interpreted within the context of the present study. As freshmen, participants had completed a limited number of exams, meaning that their GPA may not fully reflect their long-term academic performance. Nevertheless, the association between problematic internet use and a lower GPA likely reflects difficulties in time management, concentration, and sleep quality, as suggested by research on social media use and academic performance (Javdi & Mohammadi, 2025; Chandrasena & Ilankoon, 2022). Additionally, the low prevalence of risky cannabis and sedative use may have limited statistical power to detect significant associations. Future research with more representative samples could clarify these effects.
Mediation models were used to assess the role of academic engagement factors in the relationship between problematic internet use and academic performance, as hypothesized in H3. Self-efficacy and lack of motivation fully mediated the negative impact of problematic internet use on GPA, indicating that this behavior influenced academic outcomes primarily through its effects on these psychological variables. Additionally, university connectedness and the other motivation subscales functioned as partial mediators. A strong sense of belonging to the university and motivation based on personal values or driven by interest and enjoyment appeared to buffer, though not eliminate, the negative impact of problematic internet use on academic performance. Conversely, motivation based on the expectation of external rewards or punishments, or driven by feelings such as guilt or shame, appeared to intensify this impact.
These findings align with recent studies showing that problematic internet use is associated with lower academic self-efficacy (Chavez-Yacolca et al., 2025; Şen, 2025). However, the role of different forms of motivation in this association remains unclear in the literature, with mixed evidence. For instance, Saiding et al. (2024) found that external motivation moderated the link between social media addiction and low personal accomplishment, such that students with low external motivation showed a stronger negative association. In contrast, Liu et al. (2020) reported that external motivation hindered academic performance among students with high intrinsic motivation but was beneficial for those with low intrinsic motivation. A recent review (Urhahne & Wijnia, 2023) concluded that both intrinsic and identified motivation are positively associated with academic achievement, while findings for external motivation are inconsistent, ranging from weak negative associations to no clear relationship. These inconsistencies suggest that the motivational profile of students may interact in complex ways with problematic internet use and academic outcomes, highlighting the need for further investigation.
For dropout intentions, multiple mediation models were tested. In the case of internet use and alcohol use, all mediators partially reduced the direct association, with university connectedness and self-efficacy emerging as key protective factors, while lack of motivation contributed to the negative impact of internet and alcohol use on dropout intentions.
For risky cannabis use, university connectedness, lack of motivation, and external motivation fully mediated its impact on dropout intentions, suggesting that a strong sense of belonging mitigates negative effects, while amotivation or externally controlled motivation exacerbates them. These results are consistent with findings from previous research on college students (Biasi et al., 2017; Díaz Mujica et al., 2019).
Finally, for sedative use, university connectedness, self-efficacy, lack of motivation, external and introjected motivation acted as partial mediators. Lack of motivation was the strongest mediator, indicating that sedative use may increase dropout intentions by fostering academic disengagement. The role of external and introjected motivation suggests that sedative use may also be linked to more controlled forms of motivation, potentially jeopardizing academic persistence.
Previous research has shown that high school students who have never used substances (including cigarettes, alcohol, and illicit drugs) report higher levels of academic self-efficacy and emotional engagement compared to their peers who have used substances in the past year (Bugbee et al., 2019). This pattern may carry significant implications for academic persistence. Moreover, both alcohol and substance use have been associated with an increased risk of school dropout in this population (Heradstveit et al., 2024; Hjarnaa et al., 2023), suggesting that these behaviors may undermine students’ long-term academic trajectories.
Our findings extend this evidence to the university context, showing that similar patterns hold among college students. This continuity of potential risk factors across educational transitions is supported by previous studies linking substance use to an increased likelihood of academic dropout in university populations (Valkov, 2018). Moreover, problematic internet use has been associated with lower academic self-efficacy and intrinsic motivation among college students (Chavez-Yacolca et al., 2025; Şen, 2025), both of which are well-established predictors of dropout (Biasi et al., 2017; Díaz Mujica et al., 2019; Marôco et al., 2020).
This underscores the crucial role of motivation, self-efficacy, and university connectedness in counteracting these negative effects, emphasizing the need for targeted interventions to support at-risk students and foster both academic engagement and psychological resources.
Despite its strengths, this study has some limitations. While the large sample size supports robust findings, its generalizability is limited, as participants were drawn from a single university. Additionally, voluntary participation may have introduced self-selection bias. The reliance on self-report measures, while allowing broad data collection, may be subject to biases affecting validity. Finally, the low prevalence of risky substance use, especially cannabis and sedatives, may have reduced our ability to detect certain associations or mediation effects that larger samples with higher substance use rates might reveal.
5. Conclusions
Our findings underscore the critical role of self-efficacy, university connectedness, and motivation, particularly the detrimental impact of low or externally regulated motivation in mediating the academic consequences of risky substance and internet misuse. These results suggest clear, actionable strategies for universities committed to enhancing student success and retention. To strengthen self-efficacy, universities should implement practical courses or workshops focused on essential academic skills such as time management, study strategies, and stress management. Addressing motivational challenges is equally crucial, with robust tutoring and mentoring programs capable of sustaining motivation and supporting students in overcoming academic difficulties. Furthermore, fostering an inclusive academic environment is paramount; this is achievable by promoting extracurricular activities, facilitating study groups, offering accessible counseling services, and organizing engaging social events that significantly enhance students’ sense of belonging. Finally, given the observed impact of problematic internet and substance use, universities must also prioritize educational campaigns to raise awareness of their potential negative effects on academic performance and well-being. These evidence-informed interventions, derived from our findings, can contribute to cultivating a more supportive and conducive learning environment, ultimately promoting student persistence and academic success from the crucial freshman year onward.
Abbreviations
The following abbreviations are used in this manuscript:
| GPA | Grade Point Average |
| ASSIST | World Health Organization’s Alcohol, Smoking, and Substance Involvement Screening Test v3.0 |
| IAUQ | Internet Abusive Use Questionnaire |
| AMS | Academic Motivation Scale |
| SASP | Perceived School Self-Efficacy Scale |
| UCS | University Connectedness Scale |
| SD | Standard Deviation |
Supplementary Materials
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ejihpe15060105/s1, Table S1: Correlation matrix (Spearman rho’s coefficients) among the first outcome (academic performance), substance/internet risky use, and academic engagement variables; Table S2: Correlation matrix (Spearman rho’s coefficients) among the second outcome (dropout intention), substance/internet risky use, and academic engagement variables; Figure S1: Baron and Kenny mediation model definition.
Institutional Review Board Statement
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Board of Directors of University of Brescia (approved with provision no. 330 on 22 November 2021).
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study.
Data Availability Statement
The data supporting the findings of this study are not publicly available due to the inclusion of sensitive or confidential information related to students. Access to the data may be granted upon reasonable request to the corresponding author, subject to ethical and legal constraints.
Conflicts of Interest
The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.
Funding Statement
This work received a specific grant from the Italian Ministry of University and Research for tutoring orientation actions, as well as recovery and inclusion actions, also with reference to students with disabilities and specific learning disabilities (D.M. n. 752, 30 June 2021).
Footnotes
Footnote Group
References
Untitled section
References
- Ache Akua B., Samek D. R. The developmental unfolding of substance use disorder symptoms and academic achievement in the transition into and out of college. Addictive Behaviors. 2023;137:107530. doi: 10.1016/j.addbeh.2022.107530.
- Akhter N. Relationship between internet addiction and academic performance among university undergraduates. Educational Research and Reviews. 2013;8(19):1793–1796.
- Andersson C., Johnsson K. O., Berglund M., Ojehagen A. Stress and hazardous alcohol use: Associations with early dropout from university. Scandinavian Journal of Public Health. 2009;37(7):713–719. doi: 10.1177/1403494809344359.
- Andrews J. A., Duncan S. C. Examining the reciprocal relation between academic motivation and substance use. Journal of Behavioral Medicine. 1997;20(6):523–549. doi: 10.1023/A:1025514423975.
- Arria A. M., Caldeira K. M., O’Grady K. E., Vincent K. B., Fitzelle D. B., Johnson E. P., Wish E. D. Drug exposure opportunities and use patterns among college students: Results of a longitudinal prospective cohort study. Substance Abuse. 2008;29(4):19–38. doi: 10.1080/08897070802418451.
- Baron R. M., Kenny D. A. The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology. 1986;51:1173–1182. doi: 10.1037/0022-3514.51.6.1173.
- Barreto H. A., de Oliveira Christoff A., Boerngen-Lacerda R. Development of a self-report format of ASSIST with university students. Addictive Behaviors. 2014;39(7):1152–1158. doi: 10.1016/j.addbeh.2014.03.014.
- Berchtold A., Genre N., Panese F. Uses of psychotropic drugs by university students in Switzerland. PLoS ONE. 2024;19(6):e0305392. doi: 10.1371/journal.pone.0305392.
- Berhanu K. Z., Sewagegn A. A. The role of perceived campus climate in students’ academic achievements as mediated by students’ engagement in higher education institutions. Cogent Education. 2024;11(1) doi: 10.1080/2331186X.2024.2377839.
- Biasi V., De Vincenzo C., Patrizi N. Relationships between self-regulation of learning, motivations and academic success of students: Identifying predictive factors of drop-out risk. Italian Journal of Educational Research. 2017;18:181–198.
- Bugbee B. A., Beck K. H., Fryer C. S., Arria A. M. Substance use, academic performance, and academic engagement among high school seniors. Journal of School Health. 2019;89(2):145–156. doi: 10.1111/josh.12723.
- Calvo-Francés F. Internet abusive use questionnaire: Psychometric properties. Computers in Human Behavior. 2016;59:187–194. doi: 10.1016/j.chb.2016.01.038.
- Chandrasena P. P. C. M., Ilankoon I. M. P. S. The impact of social media on academic performance and interpersonal relations among health sciences undergraduates. Journal of Education and Health Promotion. 2022;11:117. doi: 10.4103/jehp.jehp_603_21.
- Chavez-Yacolca D. R., Castro-Champión R. B., Cisneros-Gonzales N. M., Cunza-Aranzábal D. F., Morales-García M., Abanto-Ramírez C. D. Relationship between academic procrastination and internet addiction in Peruvian university students: The mediating role of academic self-efficacy. Frontiers in Psychology. 2025;15:1454234. doi: 10.3389/fpsyg.2024.1454234.
- De Vincenzo C. Orientare il percorso accademico. Strategie cognitivo-motivazionali nella prevenzione del drop-out. RomaTrE-Press; 2024.
- Díaz Mujica A., Pérez Villalobos M. V., Bernardo Gutiérrez A. B., Cervero Fernández-Castañón A., González-Pienda J. A. Affective and cognitive variables involved in structural prediction of university dropout. Psicothema. 2019;31(4):429–436. doi: 10.7334/psicothema2019.124.
- Durán-Aponte E., Elvira-Valdés M. A. Patrones atribucionales y persistencia académica en estudiantes universitarios: Validez de la Escala Atribucional de Motivación de Logro General (EAML-G) Revista Intercontinental de Psicología y Educación. 2015;17(2):201–222.
- Fior C. A., Polydoro S. A. J., Pelissoni A. M. S., Dantas M. A., Martins M. J., Almeida L. S. Impact of self-efficacy and academic performance in the dropout of higher education students. Psicologia: Teoria e Prática. 2022;26:e235218. doi: 10.1590/2175-35392022235218t.
- Gould T. J. Addiction and cognition. Addiction Science & Clinical Practice. 2010;5(2):4–14.
- Hardre P. L., Reeve J. A motivational model of rural students’ intentions to persist in, versus drop out of high school. Journal of Educational Psychology. 2003;95(2):347–356. doi: 10.1037/0022-0663.95.2.347.
- Heradstveit O., Hysing M., Bøe T., Nilsen S. A., Sivertsen B., Bretteville-Jensen A. L., Askeland K. G. Prospective associations between adolescent risky substance use and school dropout and the role of externalising and internalising problems. Nordisk Alkohol- & Narkotikatidskrift: NAT. 2024;41(1):24–38. doi: 10.1177/14550725231188568.
- Hjarnaa L., Møller S. P., Curtis A. B., Becker U., Andersen O., Torvik F. A., Tolstrup J. S. Alcohol intake and academic performance and dropout in high school: A prospective cohort study in 65,233 adolescents. The Journal of Adolescent Health. 2023;73(6):1083–1092. doi: 10.1016/j.jadohealth.2023.07.008.
- Italian Ministry of Health [(accessed on 20 March 2025)];Libro bianco “Informare, educare, curare: Verso un modello partecipativo e integrato dell’alcologia italiana”. 2022 Available online: https://www.salute.gov.it/imgs/C_17_pubblicazioni_3274_allegato.pdf.
- Italian Presidency of the Council of Ministers [(accessed on 20 March 2025)];Relazione annuale al Parlamento sul fenomeno delle tossicodipendenze 2023. 2023 Available online: https://www.politicheantidroga.gov.it/media/ix0b0esf/relazione-al-parlamento-2023.pdf.
- Javdi A., Mohammadi Y. Investigating the relationship between academic procrastination, internet addiction, and academic performance among medical sciences students, Birjand, Iran. Future Medical Education Journal. 2025;15(1):25–29. doi: 10.22038/fmej.2025.84028.1613.
- Juthamanee S., Gunawan J. Factors related to internet and game addiction among adolescents: A scoping review. Belitung Nursing Journal. 2021;7(2):62–71. doi: 10.33546/bnj.1192.
- Karabacak M., Kose E. B., Bahadir Z., Demirel Y., Mutlu E. B., Ozcan Z., Ozkara B. B. Factors associated with substance use among preclinical medical students in Turkey: A cross-sectional study. Canadian Medical Education Journal. 2024;15(3):37–44. doi: 10.36834/cmej.77088.
- Kassaw C., Demareva V. Determinants of academic achievement among higher education students found in low-resource settings: A systematic review. PLoS ONE. 2023;18(11):e0294585. doi: 10.1371/journal.pone.0294585.
- Lederer A. M., Oswalt S. B., Hoban M. T., Rosenthal M. N. Health-related behaviors and academic achievement among college students. American Journal of Health Promotion. 2024;38(8):1129–1139. doi: 10.1177/08901171241255768.
- Li C., Kang L., Miles T. P., Khan M. M. Factors affecting academic performance of college students in China during COVID-19 pandemic: A cross-sectional analysis. Frontiers in Psychology. 2023;14:1268480. doi: 10.3389/fpsyg.2023.1268480.
- Li Y., Li G., Liu L., Wu H. Correlations between mobile phone addiction and anxiety, depression, impulsivity, and poor sleep quality among college students: A systematic review and meta-analysis. Journal of Behavioral Addictions. 2020;9(3):551–571. doi: 10.1556/2006.2020.00057.
- Liu Y., Hau K. T., Liu H., Wu J., Wang X., Zheng X. Multiplicative effect of intrinsic and extrinsic motivation on academic performance: A longitudinal study of Chinese students. Journal of Personality. 2020;88(3):584–595. doi: 10.1111/jopy.12512.
- Long J., Liu T. Q., Liao Y. H., Qi C., He H. Y., Chen S. B., Billieux J. Prevalence and correlates of problematic smartphone use in a large random sample of Chinese undergraduates. BMC Psychiatry. 2016;16:408. doi: 10.1186/s12888-016-1083-3.
- López-Moreno M., Garcés-Rimón M., Miguel M., Iglesias-López M. T. Influence of eating habits and alcohol consumption on the academic performance among a university population in the community of Madrid: A pilot study. Heliyon. 2021;7(6):e07186. doi: 10.1016/j.heliyon.2021.e07186.
- Marôco J., Assunção H., Harju-Luukkainen H., Lin S. W., Sit P. S., Cheung K. C., Maloa B., Ilic I. S., Smith T. J., Campos J. A. D. B. Predictors of academic efficacy and dropout intention in university students: Can engagement suppress burnout? PLoS ONE. 2020;15(10):e0239816. doi: 10.1371/journal.pone.0239816.
- Meda S. A., Gueorguieva R. V., Pittman B., Rosen R. R., Aslanzadeh F., Tennen H., Leen S., Hawkins K., Raskin S., Wood R. M., Austad C. S., Dager A., Fallahi C., Pearlson G. D. Longitudinal influence of alcohol and marijuana use on academic performance in college students. PLoS ONE. 2017;12(3):e0172213. doi: 10.1371/journal.pone.0172213.
- Melca I. A., Teixeira E. K., Nardi A. E., Spear A. L. Association of internet addiction and mental disorders in medical students: A systematic review. Primary Care Companion for CNS Disorders. 2023;25(3):22r03384. doi: 10.4088/PCC.22r03384.
- Meng S., Qi K., Shen P., Zhang M., Zhang Y., Onyebuchi N., Zhan G., Wei F., Tong W., Han Y., Ge X. The effects of mobile phone addiction on learning engagement of Chinese college students: The mediating role of physical activity and academic self-efficacy. BMC Public Health. 2025;25(1):110. doi: 10.1186/s12889-024-21250-w.
- Mota N., Alvarez-Gil R., Corral M., Rodríguez Holguín S., Parada M., Crego A., Caamaño-Isorna F., Cadaveira F. Risky alcohol use and heavy episodic drinking among Spanish university students: A two-year follow-up. Gaceta Sanitaria. 2010;24(5):372–377. doi: 10.1016/j.gaceta.2010.02.013.
- Murai H., Nakayama T. A one-year follow-up study on predictors of temporary leaves and drop-outs among students at a women’s junior college. Journal of Epidemiology. 2008;18(1):26–36. doi: 10.2188/jea.18.26.
- Pastorelli C., Picconi L. Scala di autoefficacia scolastica percepita. In: Caprara G. V., editor. La valutazione dell’autoefficacia. Erickson; 2001. pp. 87–104.
- Phillips K. T., Phillips M. M., Lalonde T. L., Tormohlen K. N. Marijuana use, craving, and academic motivation and performance among college students: An in-the-moment study. Addictive Behaviors. 2015;47:42–47. doi: 10.1016/j.addbeh.2015.03.020.
- Rees D. I. Does substance use affect academic performance? IZA World of Labor. 2019:66. doi: 10.15185/izawol.66.v2.
- Reid A. E., Cho G. Y., Carey K. B., Witkiewitz K. World Health Organization risk drinking levels among mandated college students: Associations with alcohol use, mental health, and academic performance. Journal of Studies on Alcohol and Drugs. 2024;85(4):522–527. doi: 10.15288/jsad.23-00074.
- Robbins S. B., Lauver K., Le H., Davis D., Langley R., Carlstrom A. Do psychosocial and study skill factors predict college outcomes? A meta-analysis. Psychological Bulletin. 2004;130(2):261–288. doi: 10.1037/0033-2909.130.2.261.
- Saiding A., Quirk M., Xie Y., Huang Q., Chen C. Extrinsic vs. intrinsic motivation: Navigating the moderating role of academic motivation among social media addiction and academic burnout among university students. SSRN; 2024. [(accessed on 30 May 2025)]. Available online: https://ssrn.com/abstract=4976604.
- Sinval J., Oliveira P., Novais F., Almeida C. M., Telles-Correia D. Exploring the impact of depression, anxiety, stress, academic engagement, and dropout intention on medical students’ academic performance: A prospective study. Journal of Affective Disorders. 2025;368:665–673. doi: 10.1016/j.jad.2024.09.116.
- Stallman H. M., Shochet I. M. University connectedness scale. Queensland University of Technology; 2008.
- Şen Ş. Smartphone addiction as a barrier to intrinsic motivation: The role of task value and self-efficacy. Psihologija. 2025;1:1. doi: 10.2298/PSI240313001S.
- Trusty W. T., Scofield B. E., Christensen A. E., White T. D., Murphy Y. E., Janis R. A., Tan H., Hernandez N. M., Hochstedt K. S. Psychological symptoms and academic dropout in higher education: A six-year cohort study. Journal of College Student Mental Health. 2025:1–21. doi: 10.1080/28367138.2024.2444883.
- Urhahne D., Wijnia L. Theories of motivation in education: An integrative framework. Educational Psychology Review. 2023;35:45. doi: 10.1007/s10648-023-09767-9.
- Valkov P. School dropout and substance use: Consequence or predictor. Trakia Journal of Sciences. 2018;16(2):95–101. doi: 10.15547/tjs.2018.02.004.
- Vallerand R. J., Pelletier L. G., Blais M. R., Brière N. M., Senécal C., Vallières E. F. The academic motivation scale: A measure of intrinsic, extrinsic, and amotivation in education. Educational and Psychological Measurement. 1992;52(4):1003–1017. doi: 10.1177/0013164492052004025.
- White A., Hingson R. The burden of alcohol use: Excessive alcohol consumption and related consequences among college students. Alcohol Research. 2013;35(2):201–218. doi: 10.35946/arcr.v35.2.11.
- Younes F., Halawi G., Jabbour H., El Osta N., Karam L., Hajj A., Rabbaa Khabbaz L. Internet addiction and relationships with insomnia, anxiety, depression, stress, and self-esteem in university students: A cross-sectional designed study. PLoS ONE. 2016;11(9):e0161126. doi: 10.1371/journal.pone.0161126.
- Zeigler D. W., Wang C. C., Yoast R. A., Dickinson B. D., McCaffree M. A., Robinowitz C. B., Sterling M. L., Council on Scientific Affairs, American Medical Association The neurocognitive effects of alcohol on adolescents and college students. Preventive Medicine. 2005;40(1):23–32. doi: 10.1016/j.ypmed.2004.04.044.
Associated Data
Supplementary Materials
Data Availability Statement
The data supporting the findings of this study are not publicly available due to the inclusion of sensitive or confidential information related to students. Access to the data may be granted upon reasonable request to the corresponding author, subject to ethical and legal constraints.