Exploring sleep quality, obesity, eating jetlag, chronotype, and Internet Gaming Disorder in early adolescents: a mixed-methods study

Abstract

Background

Recent research indicates that video games and screen time are directly related to children’s sleep patterns and lifestyles. Uncontrolled and prolonged gaming may negatively impact sleep quality, alter meal timings, and increase the risk of obesity. The effects are more pronounced in those with an evening-type orientation. This study aims to examine the interaction among sleep quality, obesity, eating jetlag, chronotype, and Internet Gaming Disorder (IGD) in early adolescents.

Methods

A total of 280 students participated in the study, aged 11–14 years, and 42.9% were female. Additionally, 6.1% of the students were classified as “disordered gamers.” The research was conducted using a mixed-methods approach, with quantitative data collected through various scales: Demographic Form, Sleep Quality Scale and Sleep Variables Questionnaire, Morningness‐Eveningness Scale for Children, and Internet Gaming Disorder Scale—Short Form. Qualitative data were obtained through semi-structured interviews.

Results

The analyses reveal that IGD scores are positively associated with eating jetlag and poor sleep quality, while MESC scores are negatively associated with both IGD and sleep quality. Multiple linear regression analysis found that eating jetlag and poor sleep quality were significantly associated with IGD scores, while MESC showed a negative association. Although IGD scores showed a weak association with BMI, BMI was not a significant predictor in the regression analysis. Additionally, the three-way ANCOVA analysis indicated that sleep quality and MESC categories had significant effects on IGD scores. The IGD scores were significantly influenced by the interaction between sleep quality and MESC categories, as well as the interaction between eating jetlag and MESC categories. Qualitative findings also highlighted that evening-type individuals exhibited more irregular eating habits and unhealthy lifestyles. They emphasized the negative effects of online gaming, reported having more irregular sleep patterns, and experienced worse mood and lower energy levels in the mornings. Furthermore, it was found that they engaged in more online gaming.

Conclusions

The findings of this study indicate that IGD may be influenced by chronotype, sleep quality, and irregularities in eating patterns, suggesting a multidimensional nature.

Introduction

Games are a natural part of human life and provide opportunities to contribute to physical, mental, and psychological development. Digital games, especially educational ones, can positively impact children’s development when played within controlled and appropriate time limits. However, in recent years, with the rapid development of the internet and the widespread availability of devices with internet access, more and more adolescents are turning to online games for entertainment, relaxation, or to establish new social connections [1–3]. Online games are electronic games played over computer networks in an online environment, which can be experienced solo or in multiplayer mode, and can operate on various digital devices with an internet connection [4]. Today, online games stand out as one of the most common forms of entertainment on a global scale, with the number of players worldwide expected to reach 1.2 billion by 2030 [5].

Although playing online games does not lead to Internet Gaming Disorder (IGD) in every individual, in some people, problematic digital gaming behavior can lead to serious psychological, physical, and social issues [6]. Research has shown that playing online games may be associated with various positive effects [7–9]. Therefore, moderate gaming is generally considered harmless and may even support individuals’ mental health and cognitive abilities. However, some studies report that excessive gaming behavior, due to its addictive potential, can lead to negative outcomes and the development of IGD [10–14]. Indeed, IGD has become an increasing global issue in recent years. Due to excessive interest in online games, individuals have devoted significant amounts of time and energy away from their personal lives to gaming [15].

The World Health Organization (WHO) has officially defined Gaming Disorder in the 11th Revision of the International Classification of Diseases. This disorder is characterized by the individual’s loss of control over gaming, prioritizing gaming over other activities, and continuing or increasing this behavior despite negative consequences [16]. Similarly, in the most recent revision of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) by the American Psychiatric Association (APA) [17], it is stated that gaming can be considered a disorder if it leads to significant impairment in the individual’s life or psychological distress, and for a diagnosis, at least five symptoms must persist for one year.

IGD can also lead to shorter sleep duration and decreased sleep quality. A systematic review conducted in 2021 found a strong relationship between IGD, short sleep duration, poor sleep quality, and daytime sleepiness [18]. Internet addiction is also linked to delayed circadian rhythms, insomnia, and generally insufficient and poor-quality sleep [18–20]. Key elements for healthy sleep in children include regular sleep times, adequate sleep duration, and good-quality sleep [21, 22].

Sleep is aligned with the circadian rhythm, a roughly 24-h day-night cycle that regulates bodily functions. Individuals’ circadian preferences determine their tendency toward an earlier or later sleep–wake schedule and are generally classified as morning-type (M-type), neither-type (N-type), and evening-type (E-type) [23]. M-type individuals perform better in the early hours of the day, while E-type individuals are more efficient in the evening [24]. In particular, early school start times disrupt children’s sleep patterns due to the misalignment between the biological and social clocks, leading to sleep issues in school-age children, such as early waking on weekdays, late sleeping on weekends, and insufficient sleep duration [25]. Brambilla et al. [26] found that only 50% of children aged 10–14 adhered to recommended sleep durations, and 79.5% of children in this age group used screen devices before bed.

Recent research shows that video games and screen time are directly related to children’s sleep patterns and overall lifestyle [27, 28]. In Turkey, children aged 6–8 years spend an average of 2.7 h daily playing video games [29]. Intense gaming and screen time can lead to reduced physical activity, an increased risk of excess weight, and irregular eating habits [30]. Specifically, irregular sleep, late-night eating, eating jetlag, and low levels of physical activity can result in changes to body mass index (BMI) and pose health risks [31–34]. Eating jetlag refers to significant shifts in meal timing between weekdays and weekends [34]. In this context, it can be stated that the chronotype, sleep patterns, eating habits, and internet gaming create an interrelated cycle that can directly impact the health of school-aged children.

In recent years, the increase in the frequency of internet gaming and the problems it causes necessitate a detailed examination of not only gaming behaviors but also lifestyle factors such as eating habits, physical activity levels, and sleep patterns and quality. Certain characteristics of digital games lead children to spend long hours on them and become addicted to these games. Uncontrolled and prolonged gaming can hinder children’s ability to meet their sleep needs at an adequate level and quality. This can result in changes in eating times, an increased risk of obesity, and particularly more pronounced negative effects in individuals with an E-type orientation. IGD poses serious risks, especially for children and adolescents. It is known that half of mental disorders emerge by the age of 14 and about 75% of them by the age of 24 [35]. Therefore, excessive and uncontrolled gaming behaviors can lead to serious psychosocial and health issues in children and adolescents [36]. Consequently, a comprehensive understanding of this group, including the presence of IGD, and its relationship with sleep patterns (e.g., chronotype and sleep quality), eating habits (e.g., eating jetlag), and BMI, will provide us with a better understanding. Thus, the primary aim of this study is to examine the relationships among sleep quality, obesity, eating jetlag, chronotype, and IGD in middle school students using both quantitative and qualitative data.

Materials and methods

In this study, a mixed-methods approach was used, where quantitative and qualitative data were collected separately, and their results were discussed together. This study is a cross-sectional study, as data were collected from participants within a specific time period.

Participants

The data were collected from students attending a public middle school selected through purposive sampling. The middle school students completed the data collection tools in the classroom environment under the guidance of a School Counselor. The selection of participants was carried out in May of the 2024–2025 academic year. During the participation process, informative presentations regarding the purpose and content of the research were made by the School Counselor and the researcher, and students were encouraged to voluntarily participate in the study. Following the informative presentations, written informed consent was obtained from the parents or legal guardians of the students, and the researcher provided contact information for further communication. The inclusion criteria for participation were: being a middle school student, having no diagnosed physical or psychiatric disorders, no clinically diagnosed sleep disorders, regular gaming habits, and no frequent meal skipping. These criteria were applied to minimize the potential confounding effects of pre-existing health conditions on sleep-related outcomes.

A total of 291 individuals met the inclusion criteria. After excluding participants with a significant amount of missing data, the final analysis included data from 280 participants. The average age of the participants was 12.78 ± 1.02 years, with ages ranging from 11 to 14 years. 42.9% of the participants were female. Studies suggest adopting a cutoff score of 32 points when diagnosing between disordered and nondisordered gamers [37]. According to this criterion, 6.1% (n = 17) of the students who participated in the study were classified as disordered gamers.

Measurement instruments

The quantitative data for this study were collected using the Demographic Form, Sleep Quality Scale and Sleep Variables Questionnaire, Morningness‐Eveningness Scale for Children, and Internet Gaming Disorder Scale—Short Form. The qualitative data were gathered through a semi-structured interview form. These two types of data aim to provide a more comprehensive understanding of student behaviors and health conditions.

Demographic form

The students’ age, gender, meal times on school days and weekends (breakfast, lunch, and dinner), height, and weight were collected using the Demographic Form.

Sleep Quality Scale and Sleep Variables Questionnaire (SQS-SVQ)

The measurement tool developed by Meijer and van den Wittenboer [38] was adapted to Turkish culture by Önder and colleagues [39]. This tool includes a total of 15 items, consisting of seven scale items to measure sleep quality and eight questionnaire items to assess various sleep variables. In this study, the seven scale items used to measure sleep quality were employed. The Cronbach’s Alpha internal consistency coefficient of the scale was determined to be 0.67. Additionally, the model fit of the single-factor structure of the sleep quality scale, consisting of seven items, was tested using Confirmatory Factor Analysis (CFA). The fit indexes obtained after all operations were as follows: χ²/df = 1.52, RMSEA = 0.039, NFI = 0.95, NNFI = 0.97, SRMR = 0.037, GFI = 0.97, AGFI = 0.95, CFI = 0.98. All values indicate a very good fit or are close to a good fit. In this study, the Cronbach’s Alpha coefficient was calculated to be 0.70.

Morningness‐Eveningness Scale for Children (MESC)

The scale was developed by Carskadon, Vieira, and Acebo [40] and adapted to Turkish by Önder and Beşoluk [41]. The scale consists of 10 items, with five or four response options. The scale scores range from 10 (evening-type) to 43 (morning-type). The internal consistency of the scale is reported as 0.69, and the test–retest reliability is 0.78. The second-order three-factor model was tested. CFA was conducted to determine whether the data fit the model. The CFA analysis yielded the following goodness of fit indices: χ² = 42.77 (SD = 13, p = 0.00), χ²/SD = 3.29, RMSEA = 0.068, NFI = 0.95, NNFI = 0.95, SRMR = 0.040, GFI = 0.98, AGFI = 0.95, CFI = 0.97. The second-order three-factor model was a good fit for the data, and all the fit indices were in a good range. In this study, the Cronbach’s Alpha coefficient was calculated to be 0.67.

Internet Gaming Disorder Scale – Short Form (IGDS9-SF)

The Internet Gaming Disorder Scale-Short Form (IGDS9-SF), developed by Pontes and Griffiths [42] and adapted to Turkish by Arıcak, Dinç, Yay, and Griffiths [43], was used to measure IGD. The scale consists of nine items, rated on a 5-point Likert scale from “1 = Strongly disagree” to “5 = Strongly agree.” Higher scores indicate a higher risk of gaming disorder. Due to its quick response time and practicality, this scale is frequently preferred by researchers. The Cronbach’s Alpha internal consistency coefficient of the scale is reported as 0.88. The one-factor structure was tested using CFA. The model fit indices for the CFA analysis were as follows: χ²/SD = 4.79, TLI = 0.87, CFI = 0.90, and RMSEA = 0.09. The results of the CFA using the maximum likelihood technique indicated that the scale retained its unidimensional structure in the Turkish sample. In this study, the Cronbach’s Alpha coefficient was calculated to be 0.76.

Semi-structured interview form

The students’ opinions were gathered through the following questions. These questions were developed by the researchers and reviewed by two experts from the Department of Science and Mathematics Education, with their feedback incorporated to finalize the questions:

• Please share your thoughts on your meal times, eating habits, and exercise routine, explaining the differences between school days and weekends, the decisions you make to maintain a healthy lifestyle, and your experiences in this process.

• How do you evaluate your internet gaming habits, the differences between school days and weekends, and the impact of gaming on your life?

• How do you explain your sleeping and waking habits, the differences in your sleep patterns between school days and weekends, and the reasons for these differences based on your own experiences and perceptions?

Calculations and statistical analyses

Determining the cutoff score for the Sleep Quality Scale (SQS)

Since there is no established cutoff score for the Sleep Quality Scale (SQS) in the literature, the cutoff score in this study was determined using Receiver Operating Characteristic (ROC) analysis. ROC analysis is a reliable statistical method that allows for the identification of the optimal cutoff point by evaluating test performance in terms of sensitivity and specificity [44]. Among the values obtained from the analysis, the cutoff point was determined based on the highest Youden Index (YI) score. The diagnostic validity of the SQS was compared with the Pittsburgh Sleep Quality Index (PSQI), which is frequently used and considered a reliable and valid measurement tool. A total of 89 participants completed both the PSQI and the SQS. Participants were divided into two groups based on the PSQI: good sleep quality (total PSQI score < 5) and poor sleep quality (total PSQI score ≥ 5) [45]. The sensitivity, specificity, and Youden Index values calculated are presented in Table 1. According to the analysis results, the highest Youden Index (YImax = 0.461) was obtained at a score of 12.5. Therefore, the appropriate cutoff score determined by the ROC analysis is 12.5, and scores of 12.5 or higher indicate poor sleep quality. This established cutoff score strengthens the applicability of the SQS in the research context.

Table 1.

Sleep quality ROC analysis results

Cutoff Score	Sensitivity	Specificity	Youden Index
8.0000	0.000	1.000	0.000
9.5000	0.051	1.000	0.051
10.5000	0.102	1.000	0.102
11.5000	0.288	1.000	0.288
12.5000	0.661	0.800	0.461
13.5000	0.932	0.500	0.432
14.5000	1.000	0.300	0.300
16.0000	1.000	0.033	0.033
18.0000	1.000	0.000	0.000

BMI and its categories

BMI values of children were calculated using the formula: BMI = Weight (kg)/Height2(m2). Based on the BMI values, children were categorized into four groups: underweight, normal weight, overweight, and obese. This classification was made by examining the BMI values for each percentile according to gender, using reference data specific to Turkish children [46]. According to the relevant reference data, the categories are defined as follows: Underweight: BMI < 3rd percentile; Normal weight: BMI between 3rd and 90th percentiles; Overweight: BMI between 90 and 97th percentiles; Obese: BMI ≥ 97th percentile.

Eating jetlag and its categories

The participants’ breakfast, lunch, and dinner times on weekdays (school days) and weekends were determined through survey questions: “What time do you typically have breakfast, lunch, and dinner on weekdays/weekends?” Based on the responses, the midpoint of eating times and eating jetlag calculations were carried out according to the formulas defined by Zerón-Rugerio, Hernáez, Porras-Loaiza, Cambras, and Izquierdo-Pulido [34]. The midpoint of eating time was calculated as the midpoint between the first and last meal: Eating Midpoint (local time) = (Last Meal Time − First Meal Time)/2 + First Meal Time. The difference between the weekday and weekend eating midpoints was defined as Eating Jetlag: Eating Jetlag = Weekend Eating Midpoint—Weekday Eating Midpoint.

Eating jetlag values were categorized into two groups to reflect the shift in individuals’ eating rhythms: Group 1: (–1 < x < 1) and Group 2: (x ≤ –1 or x ≥ 1). This classification allows for a comparative analysis of the degree of shift in eating times and the differences among individuals.

Determination of MESC categories

Based on the MESC scores, individuals are classified into M-, N-, and E-types. In this classification, the cut-off points at the 10th and 90th percentiles are typically used [47]. In this study, the corresponding MESC total score for these percentiles is 19/31.

Statistical analyses

Normality of all variables was assessed using skewness and kurtosis values, histograms, and Q-Q plots. Continuous variables were defined using mean and standard deviation, while categorical data were presented as percentages (%). Relationships between variables were examined using zero-order Pearson correlations. To explore whether eating jetlag, sleep quality, BMI, and MESC predicted IGD while controlling for age and gender, multiple regression analysis was conducted. In regression analyses, eating jetlag was treated as a continuous variable and entered into the models in minutes. Regression analysis was performed using the Enter method. The models included unstandardized regression coefficients (B), standardized regression coefficients (β), total explained variance (R²), and the variance contribution of each predictor variable (ΔR²). All variables were entered into the models using the Enter method. To investigate differences in the total IGD scores between eating jetlag, sleep quality, and MESC groups while controlling for age and gender, a univariate three-way analysis of covariance (ANCOVA) was applied. LSD Post Hoc analyses were conducted to examine group differences. Eta-squared (η²) was used to determine effect size. An effect size of 0.01 to 0.05 was considered small, 0.06 to 0.13 as medium, and greater than 0.14 as large. All statistical analyses were performed using SPSS software, and statistical tests were considered significant when p < 0.05.

Qualitative methodology and data analysis

Participants for the qualitative component were selected from the initial quantitative sample. Participants were selected through purposive sampling considering their chronotype, sleep quality, and eating jetlag categories. For example, individuals with an eveningness chronotype and poor sleep quality, as well as those with an eveningness chronotype and good sleep quality, were included in the study. A similar selection process was applied for morningness chronotype individuals. Additionally, individuals experiencing eating jetlag were selected based on their chronotype. Recruitment was coordinated with school teachers, and researchers identified students in the classrooms based on names provided during the quantitative phase, arranging interview schedules directly with them. Students were clearly informed that interviews would take place after school hours and that participation was entirely voluntary. Specifically, the qualitative sample consisted of a total of 77 participants, including 40 E-type individuals (33 poor sleep quality, 7 good sleep quality) and 37 M-type individuals (12 poor sleep quality, 25 good sleep quality). Among these participants, 38 individuals were classified in the eating jetlag category 2, including 24 E-type and 14 M-type individuals. The mean age of the participants was 12.82 years (SD = 1.05). The participants’ ages ranged from 11 to 14 years. Regarding gender distribution, 53.2% were female, and 46.8% were male.

The interviews were conducted in a school setting by three trained researchers. Before each interview, students were reminded of their right to withdraw at any time, and their verbal assent was obtained to confirm willing participation. Given that the initial informed consent from parents or legal guardians explicitly covered both study phases, no separate written consent was required. Each researcher interviewed approximately 25 students, and the data collection process was carried out simultaneously. Individual, face-to-face interviews were conducted in a quiet, private area of the school. Each interview lasted approximately 20–25 min. Data collection was completed over a three-week period on three days per week, during after-school hours, with each researcher conducting approximately 3–4 interviews per day, which allowed for a systematic and manageable data collection process.

The qualitative data were analyzed using content analysis. The analysis resulted in the creation of codes, categories, and themes. The emerging codes were interpreted based on their frequencies and percentage values. The percentage values of the codes were calculated based on the total number of codes in the respective category. Since multiple codes could be derived from a single participant’s response in each category, the percentage values were calculated relative to the total number of codes in that category. The qualitative data were independently evaluated by two researchers to ensure inter-rater reliability, and agreement was reached between them.

Results

Descriptive results

Table 2 presents the descriptive statistics for IGD, BMI, MESC, eating jetlag, and sleep quality, categorized by age, gender, eating jetlag, sleep quality, MESC, and BMI categories.

Table 2.

Descriptive results

Variable	N	IGD	BMI	MESC	Eating jetlag	Sleep quality
Total	280	20.31 ± 6.53	20.11 ± 3.97	25.20 ± 4.99	1:12 ± 1:05	13.58 ± 2.53
Age
11	44	19.25 ± 6.07	18.71 ± 3.66	26.02 ± 5.11	1:21 ± 1:14	13.79 ± 2.33
12	47	20.77 ± 6.16	19.75 ± 3.37	25.79 ± 4.35	1:03 ± 1:08	13.40 ± 2.62
13	107	20.08 ± 6.90	20.53 ± 4.20	25.42 ± 4.76	1:20 ± 0:59	13.35 ± 2.37
14	74	20.72 ± 6.39	20.46 ± 3.95	24.22 ± 5.34	1:02 ± 1:04	13.76 ± 2.77
Gender
Female	120	18.68 ± 5.89	19.45 ± 3.67	24.84 ± 5.14	1:08 ± 1:04	13.88 ± 2.53
Male	158	21.42 ± 6.66	20.65 ± 4.13	25.43 ± 4.77	1:15 ± 1:06	13.31 ± 2.51
Eating jetlag categories
Group 1	65	19.38 ± 5.86	20.38 ± 3.90	26.11 ± 5.27	0:18 ± 0:30	13.31 ± 2.55
Group 2	132	21.57 ± 6.70	20.42 ± 4.35	24.53 ± 4.76	1:38 ± 1:01	13.88 ± 2.34
Sleep quality categories
Good	94	18.00 ± 5.67	20.43 ± 4.04	27.74 ± 4.13	1:04 ± 1:08	11.05 ± 1.18
Poor	173	21.46 ± 6.72	19.92 ± 3.88	23.74 ± 4.89	1:15 ± 1:05	14.95 ± 1.95
MESC categories
M-type	43	16.79 ± 6.59	20.30 ± 4.54	32.60 ± 1.81	0:47 ± 0:52	11.95 ± 2.00
N-type	190	20.04 ± 5.50	20.14 ± 3.89	25.37 ± 2.71	1:15 ± 1:05	13.44 ± 2.33
E-type	44	24.73 ± 8.17	19.84 ± 3.68	17.23 ± 2.04	1:20 ± 1:13	15.72 ± 2.41
BMI categories
Underweight	19	18.63 ± 4.79	15.05 ± 0.99	24.28 ± 5.34	1:07 ± 1:06	13.61 ± 1.97
Normal Weight	196	19.92 ± 6.38	19.12 ± 2.42	25.30 ± 4.85	1:10 ± 1:07	13.44 ± 2.39
Overweight	29	20.90 ± 7.01	25.83 ± 1.71	26.32 ± 5.25	1:15 ± 1:07	13.46 ± 3.06
Obese	14	24.64 ± 9.22	29.00 ± 2.32	23.71 ± 5.74	1:34 ± 0:34	14.50 ± 3.28

Group 1 (–1 < x < 1), Group 2 (x ≤ –1 or x ≥ 1)

BMI Body Mass Index, MESC Morningness-Eveningness Scale for Children, IGD Internet Gaming Disorder, M-type Morning type, N-type Neither type, E-type Evening Type

In this study, 76.0% of the participants were classified as normal weight, 11.2% as overweight, and 5.4% as obese. Regarding sleep quality, 55.9% exhibited poor sleep quality. According to the eating jetlag categories, 33.0% were in Group 1 (–1 < x < 1), and 67.0% were in Group 2 (x ≤ –1 or x ≥ 1). When evaluated by age, IGD scores ranged from 19.25 to 20.77, BMI from 18.71 to 20.53, MESC scores from 24.22 to 26.02, eating jetlag from 1:02 to 1:21 h, and sleep quality scores from 13.35 to 13.79. When evaluated by gender, the scores for female and male participants were as follows: IGD 18.68 and 21.42; BMI 19.45 and 20.65; MESC 24.84 and 25.43; eating jetlag 1:08 and 1:15 h; sleep quality 13.88 and 13.31.

Correlation analysis

The zero-order Pearson correlations between the study variables are presented in Table 3. As shown, the total IGD scores were positively and moderately correlated with eating jetlag (r = 0.231, p < 0.01) and poor sleep quality (r = 0.245, p < 0.01). MESC scores showed a moderate negative relationship with IGD (r = –0.305, p < 0.01), a small negative relationship with eating jetlag (r = –0.184, p < 0.01), and a moderate negative relationship with poor sleep quality (r = –0.468, p < 0.01). BMI was found to have a small positive relationship with IGD (r = 0.148, p < 0.05) but did not show a significant relationship with eating jetlag, poor sleep quality, or MESC (r = 0.061, r = 0.035, r = –0.001, respectively, not significant). The correlation strengths were interpreted according to the criteria proposed by Cohen [48].

Table 3.

Zero-order correlations among the variables

Varibles	IGD	Eating jetlag	Sleep quality	MESC
Eating jetlag	.231**
Sleep quality	.245**	.051(n.s.)
MESC	-.305**	-.184**	-.468**
BMI	.148*	.061 (n.s.)	.035(n.s.)	-.001(n.s.)

IGD Internet Gaming Disorder, Sleep Quality (higher scores indicate poor sleep quality), MESC Morningness-Eveningness Scale for Children

^*p < 0.05; ** p < 0.01; n.s.: non-significant

Regression analysis

Multiple linear regression analysis was conducted to examine whether BMI, eating jetlag, sleep quality, and MESC predict IGD, while controlling for age and gender. Multicollinearity was assessed using the variance inflation factor (VIF), and whether the critical threshold (VIF > 5) was exceeded was checked. Additionally, the Durbin-Watson value was examined for autocorrelation. The results of the multiple linear regression analysis and the model summary are presented in Tables 4 and 5.

Table 4.

Results of multiple linear regression analysis for the variable

Model	Variable	B	Std. Error	Beta	t	p	VIF
Model 1	(Constant)	18.054	6.835	—	2.641	.009	—
	Gender	2.951	1.000	.223	2.951	.004	1.045
	Age	.090	.538	.013	.167	.868	1.045
Model 2	(Constant)	24.125	8.354	—	2.888	.004	—
	Gender	3.538	.919	.267	3.849	.000	1.106
	Age	-.430	.494	-.060	-.872	.384	1.103
	BMI	.134	.107	.084	1.251	.213	1.032
	Eating Jetlag	.014	.007	.138	2.025	.044	1.065
	Sleep Quality	.478	.220	.167	2.178	.031	1.347
	MESC	-.399	.104	-.300	−3.846	.000	1.395

Table 5.

Model summary

Model	R	R2	Adjusted R2	Std. Error	R2 Change	F	p	Durbin Watson
Model 1	.226	.051	.040	6.50	4.671	.011	.226
Model 2	.510	.260	.234	5.81	9.938	.000	.510	1.905

In the multiple linear regression analysis, in Model 2, the associations of the independent variables were examined while controlling for gender and age. In this analysis, BMI was found to have no significant association, whereas eating jetlag (β = 0.138, p = 0.044) and poor sleep quality (β = 0.167, p = 0.031) showed a small-to-moderate positive association with the dependent variable. MESC, on the other hand, showed a moderate negative association (β = –0.300, p < 0.001). The model indicates that the independent variables explained 23.4% of the variance in the dependent variable (adjusted R² = 0.234), and the model was statistically significant with an F value of 9.938 and p < 0.001. The Durbin-Watson value of 1.905 suggests that there is no autocorrelation, and all VIF values being below 1.395 rule out multicollinearity issues. The effect size was calculated using Cohen’s f² formula (f² = R²/(1 – R²)) and was found to be moderate at f² = 0.305 [48].

ANCOVA analysis

To assess the effects of MESCC (M-type, N-type, and E-type), EJC (Group 1 and 2), and SQC (good and poor) variables on IGD, a three-way ANCOVA was conducted, controlling for age and gender. Prior to conducting the ANCOVA analyses, several assumptions were examined. The normality of residuals was assessed using the Shapiro–Wilk test, and the results indicated no significant deviation from normality (p = 0.067). Homogeneity of variances was tested using Levene’s test, which was statistically significant (p = 0.021), suggesting a violation of the assumption. However, given the unequal group sizes across some categorical variables and the relatively large overall sample size, ANCOVA is considered robust to moderate violations of this assumption. The assumption of homogeneity of regression slopes was examined by testing the interaction effects between the covariates (age and gender) and the independent variables [eating jetlag categories (EJC), sleep quality categories (SQC) and MESC categories(MESCC)], and none of the interaction terms were statistically significant (all p > 0.05). Multicollinearity diagnostics indicated no concerns (VIF range = 1.04–1.22). Finally, outliers and influential cases were assessed using standardized residuals, studentized residuals, Cook’s distance, and leverage values, all of which were within acceptable ranges. Preliminary analyses confirmed that the assumptions of ANCOVA were adequately satisfied. The results of the analysis are presented in Table 6.

Table 6.

Three-way analysis of covariance results

Source	SS	df	MS	F	p-value	Effect size
Gender	425.243	1	425.243	13.678	.000	.074
Age	17.232	1	17.232	.554	.458	.003
EJC	75.098	1	75.098	2.415	.122	.014
SQC	630.848	1	630.848	20.291	.000	.106
MESCC	214.631	2	107.316	3.452	.034	.039
EJC * SQC	50.073	1	50.073	1.611	.206	.009
EJC * MESCC	230.169	2	115.084	3.702	.027	.041
SQC * MESCC	342.070	2	171.035	5.501	.005	.060
EJC * SQC * MESCC	5.207	1	5.207	.167	.683	.001
Error	5347.580	172	31.091
Total	87,345.000	185

EJC Eating Jetlag Categories, SQC Sleep Quality Categories, MESCC MESC Categories, SS Sum of Squares, df Degrees of Freedom, MS Mean Squares, F F-statistic

A three-way ANCOVA was conducted to examine the effects of EJC, SQC, and MESCC on IGD scores, controlling for gender and age. The main effect of gender was significant, F(1, 172) = 13.678, p < 0.001, η² = 0.074, whereas the covariate age was not significant, F(1, 172) = 0.554, p = 0.458, η² = 0.003. Among the independent variables, SQC (F(1, 172) = 20.291, p < 0.001, η² = 0.106) and MESCC (F(2, 172) = 3.452, p = 0.034, η² = 0.039) were significant. But EJC was not significiant, F(1, 172) = 2.415, p = 0.122, η² = 0.014. Students with poor sleep quality had higher IGD scores (M ± SE = 21.79 ± 0.78) compared to students with good sleep quality (16.67 ± 0.91). M-type students had lower IGD scores (16.98 ± 1.16) compared to both N-type (20.37 ± 0.58) and E-type (21.59 ± 1.34) students.

Interaction effects showed that EJC × MESCC (F(2, 172) = 3.702, p = 0.027, η² = 0.041) and SQC × MESCC (F(2, 172) = 5.501, p = 0.005, η² = 0.060) were significant. The EJC × SQC interaction (F(1, 172) = 1.611, p = 0.206, η² = 0.009) and the three-way interaction EJC × SQC × MESCC (F(1, 172) = 0.167, p = 0.683, η² = 0.001) were not significant (see Figs. 1 and 2). LSD pairwise comparisons revealed the following:

• EJC × MESCC: Within EJC Group 2, M-type students had lower IGD scores (16.36 ± 1.61) compared to both N-type (20.66 ± 0.64) and E-type (21.95 ± 1.71) students. All other comparisons were not significant (p > 0.05).

• SQC × MESCC: within M-type students, students with poor sleep quality had higher IGD scores compared to students with good sleep quality (20.11 ± 1.89 vs. 13.84 ± 1.30; p = 0.007). Similarly, within E-type students, students with poor sleep quality had higher IGD scores compared to students with good sleep quality (24.14 ± 1.22 vs. 16.48 ± 3.22; p = 0.028). In the poor sleep quality group, N-type students had significantly lower IGD scores than E-types (21.12 ± 0.67 vs. 24.14 ± 1.22; p = 0.031). Among the good sleep quality group, N-type students had higher scores than M-types (19.61 ± 0.95 vs. 13.84 ± 1.30; p < 0.001). All other comparisons were not significant (p > 0.05).

Fig. 1.

EJC x MESCC interaction

Fig. 2.

SQC x MESCC interaction

Overall, these findings suggest that IGD scores are significantly influenced by the interaction between sleep quality and MESC levels, as well as by the interaction between eating jetlag and MESC. The three-way interaction was not significant (p > 0.05), indicating that the combined effect of EJC, SQC, and MESCC on IGD was not observed.

Qualitative analysis results

Students’ perspectives on their eating habits, routines, exercise and physical activity, gaming habits, and the impact of gaming on their lives, as well as their views on sleep patterns and habits, are presented in Tables 7, 8 and 9, considering the differences between E-type and M-type individuals, SQC, and EJC Group 2 (x ≤ –1 or x ≥ 1).

Table 7.

Comparison of views on eating habits, routines, and exercise efforts in E-type and M-type individuals: distribution based on SQC and EJC group 2

Themes/subthemes	E-Type Individuals		M-Type Individuals		EJC Group 2 (x ≤ –1 or x ≥ 1)
	Poor Sleep Quality (n = 33)	Good Sleep Quality (n = 7)	Poor Sleep Quality (n = 12)	Good Sleep Quality (n = 25)	E-Type Individuals (n = 24)	M-Type Individuals (n = 14)
Eating Habits and Routines
• Meal Routine
✓ Regular meal eating	15 (51.72%)	7 (58.33%)	11 (78.57%)	25 (96.15%)	14 (60.87%)	9 (69.23%)
✓ Irregular meal eating	14 (48.28%)	5 (41.67%)	3 (21.43%)	1 (3.85%)	9 (39.13%)	4 (30.77%)
• Eating Habits
✓ Healthy eating and efforts	12 (37.50%)	6 (60.00%)	6 (46.15%)	28 (80.00%)	8 (33.33%)	8 (47.06%)
✓ Unhealthy eating	20 (62.50%)	4 (40.00%)	7 (53.85%)	7 (20.00%)	16 (66.67%)	9 (52.94%)
• Weekday and Weekend Meal Time
✓ Same meal times	6 (28.57%)	1 (14.29%)	5 (35.71%)	14 (63.64%)	6 (26.09%)	5 (38.46%)
✓ Different meal times	15 (71.43%)	6 (85.71%)	9 (64.29%)	8 (36.36%)	17 (73.91%)	8 (61.54%)
✓ Reasons for different meal times
◦ I eat earlier on weekdays	5(100%)	5(100%)	1(100%)	5(100%)	6(100%)	3(100%)
Exercise and Physical Activity
• Exercise Participation Status
✓ Efforts to exercise	15 (55.56%)	6 (66.67%)	8 (61.54%)	26 (78.79%)	11 (50.00%)	7 (53.85%)
✓ Lack of exercise	12 (44.44%)	3 (33.33%)	5 (38.46%)	7 (21.21%)	11 (50.00%)	6 (46.15%)

The values in the table represent frequencies, and the values in parentheses indicate percentages

Table 8.

Comparison of gaming habits and effects in E-type and M-type individuals: distribution based on SQC and EJC group 2

Themes/subthemes	E-Type Individuals		M-Type Individuals		EJC Group 2 (x ≤ –1 or x ≥ 1)
	Poor Sleep Quality (n = 33)	Good Sleep Quality (n = 7)	Poor Sleep Quality (n = 12)	Good Sleep Quality (n = 25)	E-Type Individuals (n = 24)	M-Type Individuals (n = 14)
Gaming Habits and Their Effects
• Gaming Time and Reasons
✓ Average online gaming duration
◦ Weekdays, minutes (SD)	126.43 (85.78)	108.75 (42.24)	109.46 (77.39)	68.30 (52.40)	112.92 (80.30)	87.71 (43.63)
◦ Weekends, minutes (SD)	176.18 (111.74)	168.75 (75.11)	181.07 (94.22)	118.02 (96.60)	166.25 (106.07)	133.93 (57.18)
✓ Average offline gaming duration
◦ Weekdays, minutes (SD)	95.15 (84.02)	52.50 (50.07)	92.50 (57.07)	72.70 (96.77)	99.62 (60.36)	61.14 (60.04)
◦ Weekends, minutes (SD)	105.63 (86.39)	71.25 (59.87)	105.83 (72.95)	73.02 (65.60)	113.08 (42.89)	84.09 (69.52)
✓ Reasons for gaming
◦ Social interaction with friends	5 (22.73%)	1 (11.11%)	1 (8.33%)	3 (9.68%)	5 (22.73%)	3 (20.00%)
◦ Boredom	8 (36.36%)	4 (44.45%)	5 (41.67%)	8 (25.81%)	8 (36.36%)	6 (40.00%)
◦ Desire to play	5 (22.73%)	1 (11.11%)	2 (16.67%)	5 (16.12%)	6 (27.27%)	4 (26.66%)
◦ Other	3 (13.64%)	1 (11.11%)	3 (25.00%)	2 (6.45%)	2 (9.09%)	1 (6.67%)
◦ Not playing	1 (4.54%)	2 (22.22%)	1 (8.33%)	13 (41.94%)	1 (4.55%)	1 (6.67%)
• Weekday vs. Weekend Gaming Time
✓ Gaming time is the same	7 (35.00%)	1 (12.50%)	1 (7.69%)	10 (58.82%)	5 (21.74%)	3 (25.00%)
✓ Gaming time is different	13 (65.00%)	7 (87.50%)	12 (92.31%)	7 (41.18%)	18 (78.26%)	9 (75.00%)
✓ Reason for the difference in gaming time between weekdays and weekends
◦ I play more on weekdays	2 (20.00%)	1 (14.29%)	-	-	1 (11.11%)	1 (16.67%)
◦ I play less on weekdays	8 (80.00%)	6 (85.71%)	3 (100%)	11 (100%)	8 (88.89%)	5 (83.33%)
• Impact of Gaming on Life
✓ Positive effects of gaming	17 (56.67%)	8 (66.67%)	6 (50.00%)	6 (42.86%)	11 (50.00%)	8 (53.33%)
✓ Negative effects of gaming	11 (36.67%)	2 (16.67%)	4 (33.33%)	1 (7.14%)	6 (27.27%)	5 (33.33%)
✓ No significant effect of gaming	2 (6.66%)	2 (16.66%)	2 (16.67%)	7 (50.00%)	5 (22.73%)	2 (13.33%)
• Emotional State and Gaming
✓ Mood	9 (56.25%)	3 (60.00%)	1 (50.00%)	4 (40.00%)	7 (63.64%)	3 (60.00%)
✓ Emotional responses	7 (43.75%)	2 (40.00%)	1 (50.00%)	6 (60.00%)	4 (36.36%)	2 (40.00%)

Except for the time values in the table, the other values represent frequencies, and the values in parentheses indicate percentages

Table 9.

Sleep patterns and habits in E-type and M-type individuals: distribution according to SQC and EJC group 2

Themes/subthemes	E-Type Individuals		M-Type Individuals		EJC Group 2 (x ≤ –1 or x ≥ 1)
	Poor Sleep Quality (n = 33)	Good Sleep Quality (n = 7)	Poor Sleep Quality (n = 12)	Good Sleep Quality (n = 25)	E-Type Individuals (n = 24)	M-Type Individuals (n = 14)
Sleep Patterns and Habits
• Bedtime and Wake-up Time and Method
✓ Weekday bedtime average, minute (SD)	23:36 (01:31)	23:13 (0:43)	22:29 (0:58)	22:24 (0:50)	23:21 (0:31)	22:41 (0:53)
✓ Weekday wake-up time average, minute (SD)	07:40 (00:28)	7:35 (0:32)	7:00 (0:36)	7:23 (0:28)	7:30 (0:31)	7:04 (0:35)
✓ Weekend bedtime average, minute (SD)	01:41 (01:25)	01:16 (1:05)	0:09 (0:15)	23:16 (0:41)	1:16 (1:12)	23:28 (0:27)
✓ Weekend wake-up time average, minute (SD)	10:58 (01:38)	10:55 (1:20)	9:12 (2:05)	9:17 (1:20)	10:56 (1:36)	9:22 (1:19)
✓ Wake-up and bedtime method
◦ Self	16 (50.00%)	11 (73.33%)	9 (69.23%)	21 (72.41%)	13 (48.15%)	12 (63.16%)
◦ External factors	16 (50.00%)	4 (26.67%)	4 (30.77%)	8 (27.59%)	14 (51.85%)	7 (36.84%)
• Weekday and Weekend Sleep Patterns
✓ Same sleep pattern	6 (17.14%)	1 (6.67%)	4 (28.57%)	11 (55.00%)	7 (28.00%)	3 (21.43%)
✓ Different sleep pattern	29 (82.86%)	14 (93.33%)	10 (71.43%)	9 (45.00%)	18 (72.00%)	11 (78.57%)
✓ Reason for the difference between weekday and weekend sleep patterns
◦ Wake up earlier on weekdays	12 (44.44%)	5 (45.45%)	4 (40.00%)	9 (60.00%)	7 (41.18%)	3 (50.00%)
◦ Go to bed earlier on weekdays	15 (55.56%)	6 (54.55%)	6 (60.00%)	6 (40.00%)	10 (58.82%)	3 (50.00%)
• Morning Emotional State and Energy
✓ Morning mood
◦ Good	4 (33.33%)	1 (50.00%)	2 (40.00%)	7 (100%)	1 (25.00%)	2 (50.00%)
◦ Bad	8 (66.67%)	1 (50.00%)	3 (60.00%)	-	3 (75.00%)	2 (50.00%)
✓ Morning mental and physical energy
◦ Good	1 (8.33%)	1 (25.00%)	2 (50%)	6 (75.00%)	2 (18.18%)	3 (60.00%)
◦ Bad	11 (91.67%)	3 (75.00%)	2 (50%)	2 (25.00%)	9 (81.82%)	2 (40.00%)

Except for the time values in the table, the other values represent frequencies, and the values in parentheses indicate percentages

When examining the frequencies and percentages related to the themes and subthemes derived from student statements in Table 7, it is observed that E-type individuals with poor sleep quality tend to exhibit irregularities in their eating habits and healthy lifestyle practices, are more prone to unhealthy eating, and neglect physical activity. This is reflected in students’ statements such as “I do not pay attention to my diet” and “Hmm… I don’t do any physical exercise”. In contrast, E-type individuals with good sleep quality have reported an increase in efforts to eat regularly, although irregularities in their lifestyles persist. For M-type individuals, even with poor sleep quality, statements emphasize regular eating habits and healthy eating behaviors, as indicated by “I regularly eat my meals every day” and “I take care of my diet”. Additionally, expressions related to exercise were found to be more frequent. However, among M-type individuals with poor sleep quality, there was a noticeable decline in efforts toward exercise. M-type individuals with good sleep quality reported perceiving themselves as significantly more successful in both healthy eating and regular physical activity. These findings suggest that E-type individuals tend to adopt a more irregular lifestyle depending on their sleep quality, while M-type individuals predominantly exhibit more regular and healthier habits. Furthermore, E-type individuals with eating jetlag reported more irregular eating and exercise habits, whereas M-type individuals indicated that they adopted a more regular and healthy lifestyle compared to E-type individuals.

When examining the frequencies and percentages of the themes and sub-themes obtained from student statements in Table 8, differences are observed in the online and offline gaming durations, as well as in the categories of sleep quality and eating jetlag, among E-type and M-type individuals. Firstly, the online gaming duration for E-type individuals is higher than for M-type individuals, both during weekdays and weekends. This is reflected in statements such as “…I play games all day from morning to evening” and “I play games for 5 h on weekends and…”, indicating prolonged gaming durations among E-type individuals. In addition to the longer online gaming time, E-type individuals’ statements indicate a significant difference in gaming time between weekdays and weekends, as expressed in “My gaming time differs between school days and weekends because…”. Both E-type and M-type individuals report that their gaming time is much higher on weekends. This is supported by statements such as “I play more games on weekends” and “I play more games on weekends than on weekdays”. M-type individuals, however, show more balanced gaming durations, as reflected in “My gaming time does not change much between weekdays and weekends; I do not play very much anyway,” indicating more stable gaming habits. Among individuals experiencing eating jetlag, the gaming time on weekdays is shorter, while it increases on weekends, and statements indicating a difference between weekday and weekend gaming times are more dominant. Based on student statements, the difference in gaming time between weekdays and weekends may primarily be influenced by school routines and may vary based on individual habits and social interactions.

When examining the reasons for playing games, it was found that students in all categories most commonly cited boredom as the primary reason for playing, as reflected in “I play games when I am very bored”. Additionally, the statements of E-type individuals were notably concentrated on social interaction with friends. Furthermore, the reason for not playing games was more frequently mentioned by students with good sleep quality, particularly those in the M-type category. These findings suggest that students view gaming as a response to various emotional and social needs, serving functions such as entertainment, socialization, and passing the time. Results also indicate that sleep quality may be related to students’ attitudes toward internet gaming. When evaluating the impact of gaming on life, it was observed that the positive effects of gaming were prominent across all groups. Both M-type and E-type individuals with poor sleep quality more intensively reported the negative effects of gaming, as reflected in “Unfortunately, playing games negatively affects my life” and “I feel like playing games wastes my time”. Additionally, students in the M-type group with good sleep quality were significantly more likely to report that gaming did not affect their lives compared to other groups. In the EJC group, however, the distribution of opinions between M-type and E-type students was similar. In terms of emotional states, the positive effect of gaming on mood was generally observed across all groups. A similar trend was also observed in emotional responses.

Based on the findings from Table 9, E-Type individuals tend to have a more irregular sleep pattern. Their bedtime during weekdays is usually late, and there is a noticeable delay in wake-up times on weekends as well. This is reflected in statements such as “…my sleep pattern differs between school days and weekends” and “I wake up later on weekends and then…”. This pattern is more pronounced in E-Type individuals with poor sleep quality. Their statements regarding weekday and weekend sleep patterns support this observation. On the other hand, M-Type individuals have more regular sleep habits, with less difference in sleep times between weekdays and weekends. Particularly, the students in the M-Type good sleep quality group have more balanced statements about their weekday and weekend sleep patterns. This is illustrated by statements such as “I wake up at 8:00 a.m.” and “Well, there is not much change in my sleep pattern between weekdays and weekends”. Additionally, individuals in the E-Type poor sleep quality group are the most likely to report waking up due to external factors, as indicated by “My mother wakes me up” and “My father forces me to get up”, while those in the E-Type good sleep quality group prefer waking up spontaneously, similar to M-Type individuals. These findings suggest that E-Type individuals have more irregular sleep patterns, while M-Type individuals have more regular sleep schedules. A similar pattern is observed in the EJC group for both E-Type and M-Type individuals. This may indicate that individuals’ sleep patterns are shaped not only by biological factors but also by their lifestyles and habits.

When examining morning emotional states and energy levels, E-Type individuals emphasize having a poor mood and low energy levels more frequently in the morning, as reflected in “hımmm.. I wake up feeling very tired in the morning” and “I wake up feeling irritable in the morning”. In the good sleep quality groups, less emphasis is placed on having a poor mood and low energy levels in the morning. A similar trend is observed in the Eating jetlag group. E-Type individuals report more negative moods and lower energy levels in the morning, while M-Type individuals feel more energetic and positive in morning states such as “I wake up feeling very energetic in the morning” and “I wake up feeling happy in the morning”. These findings suggest that E-Type individuals experience more difficulty and lower energy levels in the morning, which may be linked to their lifestyle, sleep patterns, and other habits. M-Type individuals, on the other hand, tend to start their mornings feeling more refreshed and positive, which could be a reflection of their overall lifestyle and habits.

Mixed-methods results

Integration of quantitative and qualitative data

The integration of quantitative and qualitative findings provides a more comprehensive understanding of the relationships among chronotype, obesity, sleep quality, eating jetlag, and IGD. Quantitative results indicated that IGD scores were positively associated with poor sleep quality and eating jetlag, whereas chronotype showed a negative association with IGD. ANCOVA results further revealed that students with an eveningness chronotype and poor sleep quality had significantly higher IGD scores.

The qualitative findings largely converged with these results. E-type individuals were found to have more irregular sleep and eating patterns, more pronounced differences between weekday and weekend routines, and longer gaming durations, particularly during weekends. In contrast, M-type students demonstrated more regular sleep, eating, and gaming habits. Sleep quality also demonstrated a similar pattern of integration. While quantitative analyses showed that poor sleep quality was associated with IGD, qualitative data indicated that students with poor sleep quality experienced more negative mood states, lower energy levels, and a greater tendency to engage in gaming behavior. This suggests that sleep quality may influence gaming behavior both directly and through emotional and functional mechanisms.

Eating jetlag showed a weak but significant association with IGD in quantitative analyses, although its main effect was not significant in the ANCOVA model. However, qualitative findings revealed that E-type students with eating jetlag exhibited more irregular eating and daily routines and reported increased gaming durations, particularly during weekends. These results suggest that eating jetlag may be indirectly associated with gaming behavior as an indicator of broader lifestyle irregularity, especially among E-type students.

BMI was not a strong predictor in the quantitative analyses and showed weak or non-significant associations with most variables; however, the qualitative findings provided some insights to interpret this pattern. Student statements revealed clear lifestyle differences across chronotype groups in terms of eating habits, physical activity, and gaming behaviors, with more sedentary lifestyles and irregular eating patterns particularly evident among E-types. These patterns suggest that BMI may reflect a broader lifestyle profile rather than directly influencing IGD.

Overall, the findings suggest that chronotype-related lifestyle patterns (including sleep, eating, and daily routines) may be linked to higher gaming engagement and an increased risk of IGD, particularly among E-type students, whereas M-type students appear to display lifestyle patterns that could be more protective.

Meta-inferences

Chronotype and IGD risk

E-type students exhibited higher levels of IGD, while qualitative findings indicated that this group also had more irregular daily routines and increased gaming durations during weekends. This supports the notion that chronotype may influence IGD through lifestyle patterns.

Sleep quality and gaming behavior

Poor sleep quality was associated with IGD in quantitative analyses, and qualitative findings showed that poor sleep quality was linked to negative mood states and increased gaming behavior. This suggests that sleep quality may influence behavior through both emotional and behavioral mechanisms.

Protective role of lifestyle regularity

The lower IGD observed among M-type students was consistent with qualitative findings indicating more regular eating and lifestyle habits. This structured lifestyle may function as a protective factor against IGD.

Indirect role of eating jetlag

Although eating jetlag showed a weak association with IGD, qualitative findings indicated that it is related to daily routine disruption and increased weekend gaming behavior. This suggests that eating jetlag may act as an indirect risk indicator.

Weekend routine changes

Increased gaming during weekends was particularly evident among groups with lifestyle irregularities. This finding, together with quantitative results, suggests that weekday-weekend routine differences may contribute to higher gaming engagement.

Discussion and conclusion

This study aims to examine the relationship between IGD and health and lifestyle habits, specifically investigating how factors such as sleep quality, obesity, eating jetlag, and chronotype interact with this disorder. Both quantitative and qualitative findings were integrated to generate meta-inferences regarding how chronotype, sleep quality, obesity, and eating-related behaviors jointly shape IGD risk. Descriptive findings show that 6.1% of the students in the study were identified as disordered gamers, indicating that internet gaming disorder may be a significant issue among middle school students. This rate highlights the effects of gaming addiction in younger populations and suggests that it may negatively impact students’ academic and psychological well-being. Indeed, in his meta-analysis, Fam [49] found the prevalence of IGD among adolescents to be 4.6%, with a wide variation ranging from 0.6% to 19.9%. Studies have noted that this rate varies based on age, sample characteristics, and geography [50, 51]. Therefore, differences in prevalence rates may arise from variations in the characteristics of the sample studied, cultural factors, and the diversity in diagnostic criteria or assessment tools used. However, the high prevalence rates in middle school students underscore the importance of families and schools addressing gaming behavior together, as emphasized by Imataka, Sakuta, Maehashi, and Yoshihara [50]. Moreover, social regulations and preventive measures should be implemented to mitigate and control the negative effects of this behavior. Gender was also found to be a significant predictor in the regression analyses. Previous research has consistently reported gender differences in gaming behaviors, showing that males tend to engage in gaming more frequently and are more likely to develop problematic gaming patterns. For example, studies indicate that IGD is more prevalent among boys than girls, highlighting gender as an important risk factor in gaming-related behaviors [13, 43].

When examining the findings regarding BMI, the 16.6% prevalence of overweight and obesity (11.2% overweight and 5.4% obese) suggests that excessive weight is a significant health issue at a young age. In a recent study similar to this one, the obesity rate among 11–14-year-old students in Turkey was reported to be 6% for boys and 2–6% for girls [52]. Obesity affects children in all age groups, with 8% of children under the age of 5 and one in every four children aged 10–19 in the European Region being overweight or obese [53]. According to the WHO, the proportion of obese children and adolescents was 2% in 1990, and by 2022, this figure had risen to 8%. As of 2022, 390 million children aged 5–19 were reported to be overweight, with 160 million of them being obese [54]. Childhood obesity carries a risk of persisting into adulthood [55], and this situation not only immediately impacts physical and mental health, academic success, and quality of life, but also has negative economic effects, making it an increasingly important issue. However, in this study, BMI did not predict IGD. As noted in a recent systematic review on children and adolescents [56], the relationship between gaming addiction and BMI shows that some studies have found higher levels of gaming addiction in overweight or obese individuals [57], while another study did not find a statistically significant relationship between gaming behaviors and being overweight [58].

The fact that 55.9% of the students reported poor sleep quality indicates the negative impact of sleep patterns on their overall health. In a study conducted by Tuncay and Göger [59], when examining the sleep quality of adolescents (ages 13–18), it was found that 65.1% had good sleep quality, while 34.9% had poor sleep quality. This study highlighted the increasing prevalence of sleep problems among adolescents in Turkey. Similarly, Cavalcanti, Lima, Barros, and Soares [60] observed that 53% of adolescents reported poor sleep quality. Poor sleep quality during childhood and adolescence has been shown to affect cognitive, behavioral, and emotional aspects [61]. Overall, the integration of data suggests that sleep quality may be an important factor linking circadian disruption and gaming behavior.

Additionally, when looking at the distribution in the EJC, it is observed that most of the students (67%) fall into the higher eating jetlag category. In parallel, Gill and Panda [62] note that eating habits during weekdays and weekends significantly differ in adults. In a study by Güzeldere, Akarsu, Durankuş, and Devrim-Lanpir [52], it was found that most school-aged children have unhealthy eating habits, including skipping meals (15.3%), preferring sweets (33%), and not having regular meals during weekdays (47.3%). This indicates that irregularities in students’ eating patterns are common, which may have an impact on metabolism, eating habits, and overall health.

In the correlation analysis, positive relationships were found between poor sleep quality and IGD. This is consistent with previous studies, which suggest that individuals with IGD may be more prone to sleep deprivation and excessive daytime sleepiness [18]. This indicates that individuals with IGD are more likely to delay their bedtime and shorten the amount of time spent in bed. Prolonged interaction with screens can delay their transition to sleep [27]. Therefore, extended interaction with internet games may lead to a deterioration in their sleep quality. In this study, it was found that students with poor sleep quality had higher IGD scores compared to those with good sleep quality. The results suggest a potential link between sleep disorders and higher IGD risk.

The relationship between chronotype and gaming addiction is another important topic frequently emphasized in the literature [63]. In this study, a negative relationship was found between MESC scores and IGD. Additionally, the fact that M-type students have lower IGD scores suggests that chronotype and sleep habits may be associated with IGD levels. In contrast, E-type students demonstrated higher IGD scores, indicating that circadian preference may be related to increased vulnerability to gaming addiction. Previous studies have also pointed out that E-type individuals are more prone to addiction and more vulnerable to negative effects [64, 65]. Individuals with a nocturnal chronotype spend more time in front of screens [66], which may be linked to shorter and irregular sleep durations. The qualitative findings of this study support this situation. E-type students reported that their bedtime was generally late during both weekdays and weekends, and there was a significant delay in their wake-up time on weekends. Additionally, their sleep patterns were more irregular. Indeed, blue light emitted from mobile phones and computer screens interferes with melatonin production, delaying the stages of the circadian clock and causing disruptions in circadian rhythm [67]. Higuchi, Motohashi, Liu, and Maeda [68] found that exciting games had an impact on sleep onset latency and rapid eye movement (REM) sleep. Similarly, Rigó, Tóth-Király, Magi, Eisinger, Griffiths, and Demetrovics [63] stated that nocturnality is not only a risk factor for problematic online behaviors because of the excessive time spent on these activities, but also because of the poorer mood associated with being a night person. In this study, E-type students emphasized having poor mood and low energy levels more in the morning, especially in the groups with poor sleep quality.

The interaction analyses conducted in this study revealed that the relationships between sleep quality, chronotype, and IGD are more complex and multidimensional. Specifically, the observed differences between M-type and E-type students across sleep quality groups suggest a potential combined influence of circadian preference and sleep patterns on IGD. In fact, the qualitative findings show that E-type students had longer online gaming durations during both weekdays and weekends compared to M-type students, with a significant difference between weekdays and weekends. Furthermore, students with good sleep quality, both E-type and M-type, had lower online and offline gaming durations compared to students with poor sleep quality. Students with good sleep quality more frequently expressed reasons for not playing games, whereas those with poor sleep quality reported feeling the negative effects of gaming more intensely. These findings suggest that sleep quality may influence gaming habits, and students with good sleep quality may be more likely to prefer a more balanced gaming time to avoid negative effects. Therefore, the qualitative and quantitative findings suggest that both chronotype and sleep quality may be associated with IGD.

Both regression and correlation analyses regarding eating jetlag show a positive relationship between eating jetlag and IGD. However, the results of this study suggest that there is no direct significant effect of irregularities in eating patterns on IGD. Taken together, these findings suggest that eating jetlag does not directly predict IGD, but may be associated with IGD through its relationship with chronotype and sleep pattern irregularity. Nevertheless, the significant interactions of EJC × MESCC indicate that eating jetlag, in relation to chronotype, may be associated with IGD. Specifically, in the group with many irregularities in their eating patterns, M-type students had lower IGD scores compared to both E-type and N-type students, which emphasizes this interaction effect. This suggests that irregularities in students’ eating patterns caused by circadian rhythms may negatively impact their eating routines and indirectly contribute to the development of IGD. In fact, Zerón-Rugerio and colleagues [34] found that chronotype is associated with meal timing and eating jetlag, with E-type individuals exhibiting more variability in their meal timing between weekends and weekdays. The researchers suggested that eveningness is characterized by late sleep/wake preferences, and thus, the delay in meal timing seen on weekends aligns with the circadian rhythms of these individuals. Qualitative findings also show that E-type individuals tend to adopt more irregular and unhealthy eating habits depending on their sleep quality. Similarly, E-type individuals in the eating jetlag group emphasized more irregular eating patterns and unhealthy lifestyles. Furthermore, studies have indicated that nocturnal individuals are more likely to use electronic devices before going to bed [20], and are associated with smartphone addiction, which can prolong wakefulness throughout the night [65]. The qualitative findings of this study reveal that E-type individuals have longer online gaming durations compared to M-type individuals on both weekdays and weekends, with noticeable differences between weekday and weekend gaming times. This trend is more pronounced in the eating jetlag group and shows similar tendencies in offline gaming. When these findings are considered together, it can be concluded that nocturnality and associated timing irregularities, especially in interaction with eating jetlag, may create a higher risk profile for addictive behaviors.

Overall, the integrated evidence indicates that IGD may be influenced by chronotype, sleep quality, and irregularities in eating patterns, suggesting a multidimensional nature. In the development of intervention strategies for IGD, interventions aimed at promoting healthy sleep habits and lifestyles aligned with individual chronotypes may be considered. In this context, it can be hypothesized that preventing IGD in young individuals might benefit from the establishment of healthy sleep patterns, encouragement of chronotype-aligned daily routines, and the mitigation of irregularities in eating habits. These observations underscore the potential value of a multidisciplinary approach to IGD. Future research could further inform the development of effective and sustainable intervention strategies by examining in greater detail the relationships and interactions among IGD, sleep, circadian rhythms, and other lifestyle factors.

Limitations

The limitations of this study include several key factors that may affect the generalizability and validity of the findings. First, the use of a cross-sectional design limits the ability to establish cause-and-effect relationships, as data were collected at a single point in time, and it was not possible to examine changes over time. Additionally, given that the surveys and scales used in the study are based on subjective data, responses based on participants’ personal perceptions may introduce bias in terms of accuracy and honesty. Behavioral measurements, such as IGD, may be influenced by social pressures and a lack of individual awareness. The study’s sample consists only of students aged 11–14, so the generalizability of the findings is limited, and it is unclear whether similar results would be obtained in different age groups or cultural contexts. Concepts like eating jetlag and sleep quality were evaluated using subjective measurement tools, and it is clear that these factors should be validated with more objective measurement methods. Furthermore, the study focused only on sleep quality, eating jetlag, and IGD, ignoring other psychological, socio-cultural, or biological factors that may affect individuals’ lifestyles. Additionally, since the findings were derived from Turkey, they may differ in other cultural and geographical contexts, thus limiting the external validity of the results for different populations. Moreover, excluding students with diagnosed sleep disorders or mental health problems may have reduced the variability of the sample and may limit the generalizability of the findings. Future studies including more heterogeneous samples are recommended. Finally, the internal consistency of the MESC scale in the present sample was relatively low (α = 0.67), which may have implications for the reliability of the related findings and should be taken into account when interpreting the results. Future research should address these limitations by using larger and more diverse samples and by incorporating more objective data collection methods, which will enrich the knowledge in this field.

Abbreviations

IGD

Internet Gaming Disorder

BMI

Body Mass Index

M-type

Morning type

N-type

Neither type

E-type

Evening type

MESCC

Morningness‐Eveningness Scale for Children Categories

EJC

Eating Jetlag Categories

SQC

Sleep Quality Categories

SQS-SVQ

Sleep Quality Scale and Sleep Variables Questionnaire

MESC

Morningness‐Eveningness Scale for Children

IGDS9-SF

Internet Gaming Disorder Scale – Short Fo

PSQI

Pittsburgh Sleep Quality Index

WHO

World Health Organization

CFA

Confirmatory Factor Analysis

Authors’ contributions

Conceptualisation, methodology, draft preparation, statistical analyses, and writing of the original manuscript İÖ, ANÖ, and EGY. Content analysis was carried out by İÖ and ANÖ. All the authors contributed to all the other processes in the study. All the authors read and approved the final manuscript.

Funding

The authors declare no funding.

Data availability

Data have been deposited in the Open Science Framework (OSF) under the link https://osf.io/kdc3g/files/gcn2p and will be made available upon reasonable request for academic use by the corresponding author.

Declarations

Ethics approval and consent to participate

All study procedures were conducted in accordance with the Helsinki Declaration and approved by the Sakarya University Educational Research and Publication Ethics Committee (Approval No: 08.05.2024, Decision No: 45 from the 31 st meeting). Written informed consent to participate in the study was obtained from the parents or legal guardians of the students involved.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.