Life Satisfaction and Family Communication as Protective Factors in Problematic Internet Use in University Students

Abstract

Introduction

Problematic Internet use (PIU) is considered a serious condition and a potential factor for various physical and psychological health problems, with higher prevalence rates in the young adult population, a situation aggravated by mandatory exposure to digital platforms and tools of virtual education due to the pandemic, extending the internet connection for academic, social, and leisure activities.

Objective

Examinate if satisfaction with life and family communication constitute protective factors in the problematic use of the Internet in Peruvian university students.

Method

Cross-sectional study of multivariate relationship. And 621 university students participated, with a mean age of 21.63 (SD = 3.04), selected through convenience sampling.

Results

According to the structural regression model, satisfaction with life and family communication has a negative impact with a combined effect of 20% in explaining PIU. Likewise, the logistic regression analysis showed that the predictor variables are associated as protective factors for PIU (odds ratio < 1, p < .01). The descriptive results show that the preferred device to connect to the Internet is the smartphone, only 20% of students use the Internet for academic purposes, around 60% connect more than 30 h a week, a 34% of university students present between serious and clinically significant problematic use.

Conclusion

The existence of favorable life satisfaction and positive family communication considerably reduces the risk of developing PIU.

Introduction/Background

It is undeniable that the use of the internet constitutes, in these times, a very necessary tool for the development of human beings in different areas such as family, social, work, academic, and scientific (Prizant-Passal et al., 2016; Wang et al., 2023).

This research highlights the behavior of two important variables, such as life satisfaction and family communication, to explain the problematic use of the Internet in university students. There is no direct history of multivariate studies relating the three variables in several databases such as Scopus, ScienceDirect, Francis & Tailor, Wiley, and Sage, which is why the antecedents are basically studies that have addressed the prevalence or the relationship between two of the variables identified in this study. For example, the systematic review study by Aznar et al. (2020) with all articles published up to January 2019 in Scopus and Web of Science (WoS) databases identified obsessive-compulsive disorder, alcohol abuse, depression, stress, sleep disorder, eating disorder, and attention deficit hyperactivity disorder as variables related to PIU. In another systematic review study with meta-analysis by Hinojo-Lucena et al. (2021) with articles published up to March 2019 in Scopus and WoS that used only university sample, they identified lack of self-control, anxiety, low self-esteem, low physical activity, and low academic self-efficacy as factors associated with PIU. Finally, in the systematic study by Sánchez-Fernández et al. (2023), using WoS, Scopus and PsycInfo databases in publications from 2013 to 2021, predictors of PIU were found to be time spent online, online gaming, stress, depression, negative affect, poor sleep quality, drug use, and impulsivity.

Currently, approximately 67.9% of the world's population accesses the Internet (Internet World Stats [IWS], 2023). This percentage has increased significantly compared to 2020, when 59% of the world's population were internet users (Statista, 2023). In Asia, South Korea presents the population with the highest number of internet users (97%); in Europe among the countries with the largest population, the percentage of users is as follows: Russia (85.3%), Germany (94%), France (92.2%), Italy (90.8%), the United Kingdom (95%), Spain (93%); similarly in North America, Canada (94%), and the United States (89.9%) present the highest percentage of user population; in South America, Peru is the fourth country with internet user population (87%), after Chile (97.2%), Uruguay (93.2%), and Argentina (91.1%) (IWS, 2023). Also in Peru, in the first quarter of 2022, 72.5% of people accessed the Internet (INEI, 2022).

For university students, internet use is part of their daily lives, and its access allows them to perform a variety of activities such as searching for information, doing homework, watching academic videos, as well as connecting to social networks, chatting, shopping, gambling, watching pornography, etc. (Tsai et al., 2020).

Nevertheless, free access to the internet, flexible schedules, leisure time, lack of control by their parents over the use of the internet inside or outside the home, constitutes a risk for internet addiction (Capa-Luque et al., 2023; Sela et al., 2021). In this same direction, it was observed in Mexican university students that the internet usage is massive and intense and that men are the ones who spend more time on the internet (Morales et al., 2020). This situation has been aggravated by the COVID-19 pandemic, with the consequent aggravation of mental health problems (Capa-Luque et al., 2024; Tasneem-Chowdhury et al., 2022).

According to the evidence, there is a trend toward a percentage increase in internet consumption and free access to the internet without restrictions inside or outside the university campus (Sánchez-Fernández & Borda-Mas, 2023). In this context, university students are exposed and in a state of vulnerability to fall into PIU (Eladl & Musawi, 2021). This problem is related to the affectation of personal and family life (Copur & Kaya, 2019; Tajalli & Zarnaghash, 2017; Zhou et al., 2024). The only study fairly close to this research is that of Kabasakal (2015), which analyzed the role of life satisfaction and family roles as predictors of PIU in college students. Even after a decade, given the massification of Internet use and the accelerated technological development, it is necessary to continue generating evidence that identifies protective factors such as adequate family communication and life satisfaction in the face of PIU. In public health, the control or reduction of PIU at the population level depends on strengthening variables that fulfill a protective function through specific interventions.

In this sense, there is a large gap to be filled with new studies that replicate the predictive model analyzed, as well as proposing new models that can incorporate variables such as loneliness and stress that have also been identified as predictors of PIU.

Review of Literature

Internet addiction or excessive internet use is defined as a lack of control in its use (Błachnio & Przepiorka, 2016; Tsitsika et al., 2014). In this same direction Arrivillaga et al. (2021) argue that PIU implies uncontrolled, excessive, and impulsive use, in which serious difficulties are evidenced in curbing its inordinate use, intense desire and preoccupation with using it, as well as seeking to use it to achieve regulation of one's negative emotions.

Also, in recent years, the use of the Internet has profoundly affected people's lives in different areas, especially in the personal aspect, such as life satisfaction. Studies indicate that there is negative relationship between life satisfaction and internet use disorder (Blachnio et al., 2019; Lachmann et al., 2016; Lachmann et al., 2018; Longstreet & Brooks, 2017; Shahnaz & Karim, 2014). However, there are studies that point to a positive association between life satisfaction and internet use in a healthy way, but not as excessive use (Lissitsa & Chachashvili-Bolotin, 2016). Along the same lines, Eladl and Musawi (2021) found greater satisfaction with life in nonaddicted students compared to addicts.

According to Diener (2006), life satisfaction represents a report of how a person evaluates his or her life as a whole. This evaluation takes into account different aspects of life and compares them with standards and expectations that the person had previously constructed (Diener et al., 1985). Life satisfaction represents the cognitive dimension of wellbeing that is a balance between expectations and achievements (Diener et al., 2013). Similarly, Inchley et al. (2016) refer that life satisfaction consists of the positive or negative appraisal of the current life experience.

It is worth mentioning that another risk factor related to Internet addiction is family communication. According to Olson (2000) family communication is defined as the interaction between family members in decision making, expression of ideas and feelings. It constitutes the primary means in the relationship between parents and sons, since positive family communication facilitates the ability to cope with the conflicts and challenges to which young college students are exposed (Liu et al., 2024).

As Cudris-Torres et al. (2020) point out, open, and functional family communication favors the positive self-knowledge of sons, constituting a protective factor throughout their development. Thus a positive family relationship would minimize the adverse effects of internet addiction (Akbar et al., 2022; Shi et al., 2023). Conversely, laissez-faire families with low scores in communication and conformity are exposed to addiction (Tajalli & Zarnaghash, 2017).

Likewise, according to Copur and Kaya (2019) family communication is negatively related to PIU, with a greater presence of the problem in boys. In the same direction, Sun and Wilkinson (2020) in the research they conducted with both high school and college students found that problematic parenting styles, strict parental attitudes, severe punishment, and regular withholding of affection are significant predictors of Internet addiction.

In the present research, university students were chosen as the object of study because their main activity is to study and this activity is associated with the daily use of the Internet (exposure) for academic consultations or fulfillment of virtual sessions, making this academic activity (together with the social and recreational use of the Internet) a high-risk factor for PIU.

There are few studies aimed at offering models to explain the role of some variables as protective factors in the problematic use of the Internet in terms of psychological and dispositional factors such as life satisfaction and family communication in the university population.

From the theoretical perspective of psychology and health of Ribes (1990), the problematic use of the Internet is understood as a state resulting from the interactions of various psychological, organismic, and sociocultural factors with lack of functional adjustment to the demands or situations of life. As a specific model, the present study postulates life satisfaction as a psychological factor and family communication as a sociocultural variable. According to Piña (2008, 2015), Ribes’ theoretical proposal is relevant and viable for healthcare as a way to reduce or eradicate diseases without pretending to replace medical practice, but complementing and covering psychological aspects.

These circumstances and this susceptibility of university students make it necessary to develop the present research and if the hypothesis postulated (favorable life satisfaction and positive family communication are protective factors that considerably reduce the risk of presenting PIU) is corroborated, it will provide empirical evidence to university authorities and agents of social change to adopt prevention and intervention measures with both university students and their families.

The objectives of the research are: (1) to describe the use of the Internet and PIU in university students. (2) To examine whether life satisfaction and family communication are protective factors in the problematic use of the Internet in Peruvian university students. (3) To identify the explanatory magnitude of life satisfaction and family communication on PIU under the methodological framework of latent variables in university students.

Methods

Design

The study assumes a nonexperimental and cross-sectional design, and due to the nature of the relationships, it is multivariate in nature (Hair et al., 2008).

Research Questions

In accordance with the formulated objectives, the study answers the following research questions: (1) What are the characteristics of Internet use and its problematic use among Peruvian university students? (2) Are life satisfaction and family communication protective factors in the problematic use of the Internet in Peruvian university students? (3) In terms of latent variables, to what extent do life satisfaction and family communication explain the problematic use of the Internet in university students?

Participants

The population comprised students from public and private universities residing in metropolitan Lima and Callao (Peru), enrolled during the first academic semester of 2023 (June and September), between 18 and 35 years of age, of both sexes.

The sample of 625 students was estimated with the formula for structural equation modeling (Soper, 2022) for an effect size of 0.20, statistical power of 0.90 and 95% confidence level (α = 0.05). Sampling was nonprobabilistic by convenience.

Inclusion and exclusion criteria: To be included, participants completed the informed consent form, agreed to participate voluntarily, resided in the province of Lima or Callao, and were of Peruvian nationality. Participants with inconsistent or contradictory response patterns in the instruments were excluded from the study.

Data Collection

The instruments were administered using a Google Form, and the URL link was sent to the emails and WhatsApp of the target population. An introductory text was presented at the beginning of the survey with information about the research objectives and a request to read informed consent. Only those who gave consent completed the instruments. The data collection period was between July 8 and September 3, 2023.

Measures

Demographic sheet: An ad hoc sheet was designed to collect information on sociodemographic data and internet use such as: age, sex, marital status, type of university (public or private), degree program, hours of daily and weekly internet connection, preferred device for internet connection, reasons for internet connection.

Satisfaction with Life Scale (Diener et al., 1985), a unidimensional self-report with five items measuring global cognitive judgments of satisfaction with one's life. It is scored on a five-point scale (strongly disagree = 1 to strongly agree = 5). The original version presented concurrent validity, as well as reliability due to internal consistency and temporal stability (Diener et al., 1985). In the various studies with Peruvian adult population, there have been satisfactory evidences of validity and reliability reported (e.g., Bello-Vidal et al., 2019). Higher scores indicate greater life satisfaction. For this study the validity based on the internal structure of the construct estimated with WLSMV was satisfactory (CFI = 0.995, TLI = 0.988; RMSEA = 0.078, SRMR = 0.024). Likewise, for reliability, an ordinal alpha of 0.88 and ordinal omega of 0.90 were obtained.

Family Communication Scale: Designed by Olson et al. (2006). The version adapted for Peruvian university students was used (Copez-Lonzoy et al., 2016), which consists of 10 Likert-type items with five response options ranging from 1 (extremely dissatisfied), to 5 (extremely satisfied). For the present study, the review of the internal structure evidenced satisfactory fit (CFI = 0.988, TLI = 0.984, RMSEA = 0.079, SRMR = 0.029), as for reliability an ordinal alpha of 0.94 and an ordinal omega of 0.95 were obtained.

Generalized Problematic Internet Use Scale 2(GPIUS2) [Escala de uso problemático generalizado de Internet 2]. It is a scale revised and updated by Caplan (2010) from the first GPIUS version (Caplan, 2002). Adapted to different languages in different European countries (Stover et al., 2023). The version used in the present study is the one translated into Spanish by Gámez-Guadix et al. (2013), which consists of 15 items organized into four subscales: (a) preference for online social interaction (3 items), (b) mood regulation (3 items), (c) negative outcomes (3 items), and (d) poor self-regulation, which is a second-order factor containing a subscale for cognitive preoccupation (3 items) and a subscale for compulsive Internet use (3 items). The response options of the GPIUS2 are Likert-type of 6 gradations ranging from 1 (strongly disagree) to 6 (strongly agree). Gámez-Guadix et al. (2013) reported evidence of validity based on the internal structure of the construct (CFI = 0.92, NNFI = 0.91, SRMR = 0.04, RMSEA = 0.06). The estimated internal consistency with Cronbach's alpha for the entire scale was 0.91, and for the subscales ranged from 0.78 to 0.90. In the recent adaptation study for the adult population of Argentina, satisfactory evidence of validity based on internal structure through CFA and validity related to other constructs was reported, as well as reliability values above 0.75 with ordinal alpha (Stover et al., 2023).

Given that GPIUS2 has not been used in the Peruvian university population, evidence of validity and reliability was examined. Table 1 presents the fit indices estimated with AFC for two models of the internal structure of the GPUIS2 scale. As can be seen, the two models present satisfactory fits (CFI and TLI > 0.95, SRMR < 0.04, WRMR (Weighted Root Mean Square Residual) < 1, and RSMEA ≤ 0.08). From the measurement point of view the first model presents slightly better fit; however, the second order model (M2) offers the necessary evidence to measure the generalized problematic use of the Internet for both primary factors and the general construct.

Table 1.

GPIUS2 Structural Model Fit Indexes.

	χ2(df) p	CFI	TLI	RSMEA	SRMR	WRMR
M1	316.674(78), 0.000	0.987	0.982	0.078 [.069, .087]	0.030	0.828
M2	349.292(82), 0.000	0.985	0.981	0.080 [.072, .089]	0.037	0.978

Note: M1 = GPIUS2 multidimensional model, M2 = Second Order Model.

CFI = Comparative Fit Index; RMSEA = Root Mean Square Error of Approximation; SRMR = Standardized Root Mean Square Residual; WRMR = Weighted Root Mean Square Residual; TLI = Tucker Lewis Index.

Figure 1 shows that both the multidimensional model (M1) and the second-order factor model (M2) present similar factor loadings (between 0.74 and 0.96).

Figure 1.

Structural models with confirmatory factor analysis of GPIUS2.

Note: CIU = compulsive Internet use; CP = cognitive preoccupation; DSR = deficient self-regulation; GPIUS = Generalized and Problematic Internet use; MR = mood regulation; NO = negative outcomes; POSI = preference for online social interaction.

As shown in Table 2, the GPIUS2 scale presents high reliability for both the overall scores and for each of its factors.

Table 2.

Reliability of the Generalized Problematic Internet Use Scale 2, GPIUS2.

	Items	Ordinal alpha	Ordinal omega
Preference for online social interaction	3	0.94	0.94
Mood regulation	3	0.87	0.88
Negative results	3	0.92	0.92
Cognitive concern	3	0.90	0.90
Compulsive use of the Internet	3	0.93	0.93
Global scale	15	0.95	0.97

Ethical Considerations

The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the Research, Innovation and Entrepreneurship Unit of the Faculty of Medical Technology of the National University Federico Villarreal (Reference: Opinion N° 008, R. Resolution N° 1343-2023-UNFV) on February 23, 2023. Informed consent was obtained from the participants and data were collected anonymously to ensure confidentiality.

Data Analysis

In the first phase, the psychometric properties of the research instruments were reviewed. For the evaluation of validity based on the internal structure of the construct of the GPIUS2 scale, confirmatory factor analysis (CFA) was used, with a robust WLSMV (Weighted Least Squares Means and Variance Adjusted) estimator for categorical items, and internal consistency reliability was estimated with ordinal alpha and McDonald's omega for categorical items.

In the second phase, in addition to running descriptive analyses of the variables, the main model of the study was analyzed using the binomial logistic regression analysis technique with the SPSS package version 26 for Windows and, as a complement, a structural regression model of latent variables was examined with structural equation modeling, using the R package version 4.0.2 and the Rstudio interface version 4.3.1.

The categorization of the latent variables (life satisfaction and family communication) was performed with the multivariate clustering technique using the k-Means algorithm. The determination of the correct number of clusters was supported by the majority rule and the stability of the clusters expressed in Jaccard's coefficient, a value ≥ 0.70 indicates the existence of stability. In this sense, for the life satisfaction construct the Jaccard coefficients were 0.98 and 0.99, for the family communication construct the coefficients were 0.98 and 0.98. These analyses were performed using the R package version 4.0.2 and the libraries used were factoextra, NbClust, fpc, cluster, tidyverse, rgl.

As for the establishment of the categories of problematic use of the Internet, this was done with decatypes, so the category of nonproblematic use corresponds to decatypes 1 to 5 and problematic use to decatypes 6 to 10.

For the evaluation of the structural regression model, the recommended fit indexes were taken into account, such as CFI (Comparative Fit Index), TLI (Tucker Lewis Index), RMSEA (Root Mean Square Error of Approximation) and SRMR (Standardized Root Mean Square Residual). Values of CFI and TLI ≥ 0.90 evidence adequate fit and good fits ≥ 0.95, for RMSEA a value ≤ 0.08 indicates adequate fit and good fit if ≤ 0.05, for SRMR an index ≤ 0.08 denotes adequate fit and ≤ 0.06 good fit (Hu & Bentler, 1999).

Results

Table 3 shows that the highest percentage of students are between 18 and 25 years of age, with a greater presence of female (58.8%), single (96.8%), national university students (70.5%), and from professional careers in the health area (50.4%).

Table 3.

Sociodemographic Characteristics of Interns in Health Sciences.

	n	%			n	%
Age (years)				Marital status
18–20	267	43.0		Single	601	96.8
21–25	276	44.4		Married	8	1.3
26–30	73	11.8		Cohabitant	12	1.9
31–35	5	0.8
Sex				Professional area
Female (21.65 ± 3.0)	365	58.8		Healthcare	313	50.4
Male (21.70 ± 2.9)	256	41.2		Humanities	71	11.4
University				Engineering	100	16.1
National	438	70.5		Administration y	137	22.1
Private	183	29.5		Finance

Descriptive Analysis of Internet Use Among University Students

Table 4 shows the use of devices by university students; the majority use Smartphones to connect to the Internet (44.4%), approximately 2 out of 10 students connect using multiple devices (multi preference) and the least used devices were Tablet (1%) and Notebook (1%).

Table 4.

Analysis of Descriptive Data on Internet Use Among University Students (n = 621).

	Frequency	Percentage
Preferred device for internet connection
Smartphone	276	44.4
Tablet	6	1.0
Laptop	27	4.3
Notebook	6	1.0
Computer	26	4.2
Smartphone + Laptop	70	11.3
Smartphone + Computer	36	5.8
Laptop + Computer	2	0.3
Multi preference	111	17.9
Tablet + Notebook	37	6.0
Other	24	3.9
Reason for internet connection
Entertainment (video games)	71	11.4
Chat	55	8.9
Information	35	5.6
Academic	98	15.8
Shopping	39	6.3
Chat (e-mail)	13	2.1
Information + Academic	137	22.1
Entertainment + Shopping	22	3.5
Multi motives	151	24.3
Weekly internet connection time
Less than 8 h	23	3.7
8–14 h	45	7.2
15–20 h	75	12.1
21–30 h	126	20.3
31–40 h	113	18.2
41 to more hours	239	38.5

With respect to the reasons for connecting to the Internet, it can be observed (Table 4) 24.3% of university students have different motives, 22.1% for information search and academic reasons, 15.8% for academic matters only, 11.4% for video games, and in a smaller proportion for cyber chat (2.1%).

Likewise, Table 4 shows the time spent by university students connecting to the Internet, most of them (38.5%) connect from 41 to more hours, 2 out of 10 students connect between 21 to 30 h and a lower percentage (3.7%) of students connect less than 8 h per week to the Internet.

Table 5 shows the level of PIU by university students. Although only a small group (5%) showed clinically significant problematic use, 6 out of 10 students showed moderate to severe problematic use. It can also be seen that about one-third of students do not have problems with Internet use.

Table 5.

Problematic Internet Use Among University Students.

	Percentiles	Decatypes	Frequency	Percentage
Not problematic	1 a 5	1 y 2	31	5.0
Mild (occasional)	6 a 30	3, 4,	175	28.2
Moderate (regular)	31 a 65	5, 6	203	32.7
Severe (frequent)	66 a 94	7, 8	181	29.1
Problem of clinical significance	95 a 99	9, 10	31	5.0
Total			621	100.0

Life Satisfaction and Family Communication as Protective Factors in Problematic Internet Use in College Students

After performing a multivariate logistic regression analysis, adjusted for sex, age of the participants and type of university (public vs. private), the possibility of confounding factors attributable to these sociodemographic factors was ruled out. These covariates were used because there are studies that reported that PIU varied according to sex (Kabasakal, 2015; Liu et al., 2024), age and socioeconomic level (Sánchez-Fernández & Borda-Mas, 2023; Sela et al., 2021), which in our case is linked to the type of university. But it is also possible the presence of other confounding variables that we have not taken into account such as living alone or family type (Liu et al., 2024). Table 6 indicates that the adjusted logistic model shows almost similar relationships between the predictor variables and the dependent variable (PIU) with respect to the crude model.

Table 6.

Binomial Logistic Regression Models Associated With Problematic Internet Use in College Students.

	Raw modela		Adjusted modelb
	B	OR [95% IC]	B	OR [95% IC]
Satisfaction with life(1a)	−0.593*	0.552 [0.365, 0.835]	−0.601*	0.548 [0.362, 0.831]
Family communication(1b)	−0.753**	0.471 [0.313, 0.710]	−0.739**	0.478 [0.316, 0.721]

Note: * p < .01, **p < .001.

^aCrude multivariate regression model.

^bMultivariate model adjusted for sex, age, and type of university.

CI = confidence interval; OR = odds ratio; (1a) Favorable, (1b) Positive.

According to the odds ratio (OR) values and their confidence intervals the predictor variables are related to the presence of PIU as protective factors (OR < 1, p < .01).

The fitted logistic model is statistically valid because the −2 log likelihood value decreased with respect to the base model (Δ −2LLL = 44.042), likewise the Hosmer and Lemeshow test indicates good model fit (χ² = 9.535, gl = 8, p = .299), and according to Nagelkerke's R² the model allows a 10% explanation for PIU from a parsimonious two-factor model.

Latent Relationships of Life Satisfaction and Family Communication With Problematic Internet Use in College Students

The structural regression model of latent variables (Figure 2) shows that both life satisfaction and family communication are factors with negative effect that allow to explain the generalized and problematic internet usage in 20%. The model also shows that the exogenous variables have a positive and significant covariance (large effect size).

Figure 2.

Indirect effects of life satisfaction and family communication on problematic Internet use in college students.

Note: ** p < .001, CIU = compulsive Internet use; CP = cognitive preoccupation; DSR = deficient self-regulation; GPIUS = Generalized and Problematic Internet use; MR = mood regulation; NO = negative outcomes; POSI = preference for online social interaction.

The model examined presents adequate support from the data because the robust goodness-of-fit indices turned out to be very good: χ² (396) = 1109.603, p < .001; CFI = 0.972; TLI = 0.970, RMSEA = 0.054[0.050, 0.58], SRMR = 0.051.

Discussion

PIU is considered a major public health problem (WHO, 2018) and potential factor for various physical and psychological health problems (Huang et al., 2020).

According to the findings, the device most frequently used to connect to the Internet is the Smartphone (44.4%), while 17.9% prefer to link through various devices. This finding aligns with the study by Ruiz-Palmero et al. (2021) where the majority of students (40.13%) after COVID-19 confinement relied on cell phones. In the same route, Siste et al. (2020) indicated that increased internet use is also associated with the pandemic. Likewise, in the research by Miri et al. (2020) the prevalence of cell phone addiction was high in students (75%), with higher prevalence observed in younger and unmarried students. In Chinese university students (36.6%), excessive use of cell phones was also prevalent (Mei et al., 2022).

Regarding the reasons for connecting to the Internet, a majority group (47.2%) stated that they connect for different social reasons and to look for general information, only 15.8% would link exclusively to research academic topics, while 11.4% would link exclusively for online games. Concordant with the findings of the current research, Alsayed et al. (2020) reported that female nursing students in Saudi Arabia went online mostly for communication (social motives) and information seeking. The comprehensive use of the internet, has significantly modified the interaction of people through connectivity and instant access to information (Eg et al., 2023). However, some students present difficulties in managing the habit of use, experiencing negative effects in different aspects of their daily life (Elhai et al., 2021).

In terms of internet connection time, it was found that a significant number of students (38.5%) connect 41 h or more per week. In view of these results, it is important to bear in mind what IWS (2023) pointed out that Peru is the fourth country in South America with the highest number of internet users (87%); within this figure, university students constitute a significant group of users, since the very condition of being students leads to the use of this tool for their academic activities; however, the high number of hours that these students spend connected to the internet is a warning sign, which could indicate an excessive or problematic use of the internet with serious difficulties in limiting its use (Arrivillaga et al., 2021). On the other hand, in the study conducted by Dogliotti et al. (2020) in Buenos Aires, it was found that 9 out of 10 young people connect to the Internet daily and that 80% connect around 10 h a day, which is a fairly high amount of time spent connected to the Internet per week, which, as in Peru, would indicate a use outside healthy parameters.

In this study, the findings also show an important data regarding PIU at the level of clinical significance, that is, 5% of students are located above the 95% percentile. This result is consistent with what Prizant-Passal et al. (2016) expressed, in that the use of the Internet is profoundly affecting people's lives; and as Arrivillaga et al. (2021) point out, due to an unstoppable desire to use it or to try to alleviate their negative emotions, students end up immersing themselves in a use that is ultimately harmful to themselves. Among some studies that reaffirm the evidence that part of the population of university students fall into problematic use of the Internet, eventually developing addiction is the study by Tsai et al. (2020), who report such prevalence in some Asian countries such as Japan (3.7%), Taiwan (8%–13%), and China (13%). On the other hand, in the results of the present study, a higher percentage of PIU was found in the moderate (32.7%) and severe (29.1%) levels. A study in university students whose results are along the same lines is that reported by Mohamad Ashari et al. (2022), who found mostly moderate levels of Internet addiction.

The main finding corroborates that life satisfaction and family communication are protective factors for decreasing PIU. According to the estimated odds ratios, the presence of favorable life satisfaction decreases the risk of PIU by approximately 45% compared to unfavorable life satisfaction. Result concordant with that reported by Kabasakal (2015) who observed that “Life satistaction decreased as problematic Internet use increased” (p. 301). Otherwise, the existence of positive family communication reduces the risk of PIU (1–0.478) by approximately 52% with respect to university students who present negative family communication. In the same direction Kabasakal (2015) and Zhou et al. (2024) report that good parent–son interaction is much more inhibitory to internet addiction, as well as high levels of life satisfaction and meaning in life would constitute protective factors to reduce the risk of internet addiction. The result is also consistent with the explanation of Liu et al. (2024) who argue that positive relationships with parents lead to meaningful actions thereby decreasing the risk of Internet addiction. Similarly, Shi et al. (2023) posit that family environment is a significant predictor of Internet addiction.

Another important finding that broadens the perspectives of analysis of the variables under study is the approach in terms of latent relationships through the strategy of structural regression analysis. In this sense, the results show that life satisfaction and family communication have negative relationships with PIU. From this, it can be inferred that experiencing greater life satisfaction and greater family communication has an impact on less PIU. There are previous studies whose results are in line with those found here, for example, Masaeli and Billieux (2022) found a negative correlation between PIU and quality of life, recalling that Shin and Johnson (1978) define life satisfaction as the overall judgment of quality of life. Similarly, Blachnio et al. (2019) in the United States found that internet addiction is associated with lower life satisfaction. Regarding family communication and PIU, the findings of this research are supported by Copur and Kaya (2019) who argue that family communication has a negative relationship with PIU; in this same line Sun and Wilkinson (2020) refer that inadequate parenting styles, in which a strict attitude predominates, severe punishment and withholding of affection by parents constitutes a predictor of internet addiction. In light of this evidences and in accordance with what is postulated by Cudris-Torres et al. (2020) it can be argued that open and functional family communication is a protective factor throughout the life cycle and as indicated by Akbar et al. (2022) a positive family relationship minimizes the adverse effects of PIU.

Having corroborated, in the present study, that when a person perceives that he/she has positive family communication and feels satisfaction with life, these are protective factors for not incurring in PIU; this result, seen from the clinical perspective, orients psychologists to develop prevention work with university students, mainly at two levels, primary prevention and secondary prevention. For example, Plaza et al. (2024) have identified that the application of preventive strategies with vulnerable students brings good results such as significant reduction of addiction symptoms, improvement of academic performance, reduction of procrastination, and increased awareness towards the proper use of the Internet; they also highlight the importance of including parents in preventive activities to promote positive interrelationships within the family environment.

Strengths and Limitations

One of the limitations of the study corresponds to the nonprobabilistic sampling used, which implies that the external validity of the research is affected, therefore, any possibility of generalization should be made with caution. Another limitation corresponds to the majority presence of health sciences students in the study sample. In cross-sectional studies, although structural equation modeling through structural regression analysis of latent variables from the statistical point of view assumes causal relationships where direct and indirect effects are analyzed (Byrne, 2010; Keith, 2019), nevertheless, from the methodological point of view it presents limitations to predict strong causal relationships that is typical of experimental designs. However, in spite of these limitations, the present study is considered important given the scarcity of research and its strength lies in seeking explanations with logistic models to identify protective factors or models with latent variables on PIU based on predictors such as life satisfaction and family communication. Therefore, it is recommended that other similar studies be carried out, with larger samples and above all with randomized samples in order to reaffirm the findings.

Implications for Practice

The study is further evidence of how young population segments show a greater tendency of PIU at this time, which generates a risk of compromising their mental health, academic and social performance. A practical implication is that the community in general should become aware of the impact of the use of networks on physical, mental, and social health; that the university system should promote face-to-face academic and recreational interactions and organize support groups for those affected.

According to the results found and the accumulated evidence, it is suggested to researchers, health professionals, and especially to university authorities that in addition to promoting or implementing intervention programs aimed at reducing PIU, it is necessary along with the development of soft skills (related to the perception of life and communication in the family environment), raise awareness with timely information and exposure to preventive programs so that students use the Internet to benefit from its responsible use and that its access becomes a means of social and productive recreation and not a risk factor for their health and wellbeing.

Conclusions

Peruvian university students who participated in the study (38.5%) show a high number of hours per week of internet connection, which makes them prone to excessive and problematic use of the internet.

In the university population, the most used device is the Smartphone, which would be associated with frequent problematic use. Only two out of 10 students in professional training connect to the internet strictly for academic purposes.

In view of the increase in PIU and that variables such as life satisfaction and family communication can be protective factors to reduce problematic use or risk of addiction; it is an important warning sign for the university system to implement policies in line with the third Sustainable Development Goal of Well-being and Health.

Life satisfaction and family communication explain in an important way the problematic use of the internet in Peruvian university students, being these latent relationships of a negative character, so if the student presents favorable (positive appraisal) life satisfaction and also has positive family communication, he/she will probably have a lower risk of PIU, thus avoiding the development of internet addiction.