Psychometric Properties of the Greek Version of the Relational Provisions Loneliness Questionnaire (RPLQ) in Early Adolescents

Abstract

Background

Loneliness during adolescence has increased worldwide in recent years and has been consistently associated with a broad range of adverse psychosocial outcomes. The availability of valid and reliable measures is essential for the early identification of loneliness and for the rigorous evaluation of intervention effectiveness. Nevertheless, multidimensional, psychometrically validated instruments for assessing loneliness in Greek adolescents remain limited.

Objective

The aim of the present study was to examine the psychometric properties of the Greek version of the Relational Provisions Loneliness Questionnaire (RPLQ; Hayden-Thomson, 1989), which assesses social and emotional dimensions of loneliness (i.e., integration and intimacy) within family and peer relationships.

Methods

The sample consisted of 503 students aged 13–14 years, recruited from secondary schools through stratified random sampling. The factor structure was tested via Confirmatory Factor Analysis, while internal consistency, validity evidence, and measurement invariance across gender were examined.

Results

The four-factor model demonstrated the best fit (CFI = .997, TLI = .997, RMSEA = .032, SRMR = .050) and yielded high standardized factor loadings (λ = .63–.93). Internal consistency was high across all dimensions (ω = .87–.92), and convergent, discriminant, and concurrent validity were supported. Measurement invariance across gender was supported at the configural, metric, and scalar (threshold) levels.

Conclusions

The Greek version of the RPLQ is a valid and reliable instrument for the multidimensional assessment of adolescent loneliness, supporting its use in research and prevention efforts.

Plain Language Summary

Loneliness during adolescence can take different forms and occur in different relationships. For some young people, it may mean not feeling close to their family. For others, it may mean feeling excluded by friends or not fully accepted in their everyday relationships. A measure designed for adolescents therefore needs to capture these different experiences, rather than treating loneliness as one general feeling. In Greece, however, there are still few validated tools that assess different forms of loneliness in this age group. This study examined the Greek version of the Relational Provisions Loneliness Questionnaire, a self-report questionnaire that adolescents complete on their own. It focuses on two important relational contexts -family and peers- and, within each context, on two aspects of loneliness: lack of emotional closeness and lack of inclusion and connection in everyday relationships. The study included 503 students aged 13-14 years in Greek secondary schools, who responded to a set of statements about their family and peer relationships. The findings supported the Greek version of the questionnaire. It measured four related but distinct kinds of loneliness -emotional and social loneliness- each in the context of family and of peers. The responses within each part were consistent with one another, and the questionnaire reflected loneliness in the way it was expected to. It also performed in a similar way for boys and girls. These results suggest that the Greek version of the questionnaire can be used by researchers and practitioners to better understand how adolescents experience loneliness with family and peers.

Keywords

relational provisions loneliness questionnaire psychometric validation factor structure reliability adolescence

Introduction

Loneliness is a negative and unpleasant experience that emerges when an individual perceives a discrepancy between actual and desired social relationships, in terms of quality and/or quantity. It is therefore conceptualized as a subjective experience that is distinct from objective social isolation, in so far as it reflects the perceived inadequacy of an individual’s intimacy and/or social ties (Cacioppo et al., 2006; Perlman & Peplau, 1981).

Adolescence is a critical period marked by the formation and reconfiguration of social identity; adolescents are called upon to negotiate competing developmental needs. Within this context, heightened and often unrealistic interpersonal expectations, experiences of rejection, challenges in establishing social roles, and restrictive parenting practices have been associated with elevated loneliness (Heinrich & Gullone, 2006; Larson et al., 1996; Qualter et al., 2013, 2015; Van Rode et al., 2015). Twenge et al. (2021), based on Programme for International Student Assessment (PISA) data across 37 countries, documented an increasing trend in students’ reported loneliness within the school setting between 2000 and 2018. In Greece, the proportion of students reporting loneliness increased from 8.96% in 2003 to 24.16% in 2018. Although gender differences have been examined extensively, empirical findings remain inconsistent (Maes et al., 2017). Notably, a meta-analysis of 638 studies indicated no statistically significant gender differences overall (Maes et al., 2019).

Intense and chronic experiences of loneliness render adolescents particularly vulnerable (Hawkley & Cacioppo, 2010; Heinrich & Gullone, 2006), with long-term adverse effects on physical and mental health (Mahon et al., 2006; Qualter et al., 2010), academic performance (Benner, 2011; Eccles et al., 2021; Jefferson et al., 2023), and career outcomes (Maes et al., 2022). Consequently, the accurate identification of adolescents experiencing loneliness and the rigorous evaluation of interventions aimed at reducing its negative effects require valid and reliable measurement of the phenomenon (Goossens & Beyers, 2002; Maes et al., 2022).

Loneliness can be conceptualized as either a unidimensional or a multidimensional construct. The unidimensional approach views loneliness as a relatively homogeneous experience that is assumed to be similar across individuals, regardless of circumstances, causes, or relational domains, and thus is assessed via a single measure. In contrast, the multidimensional framework posits that loneliness can take different forms, vary in intensity, and differ as a function of specific social contexts, relationship types, and antecedent conditions (Goossens et al., 2009; Houghton et al., 2014; Maes et al., 2022; Ribeiro et al., 2019). This multidimensional perspective is grounded in Weiss’s (1973) social provisions theory, which proposes that different types of social relationships fulfil distinct social needs. The degree to which these needs are met determines the specific form(s) of loneliness experienced (Galanaki, 2014; Maes et al., 2017).

A central distinction discussed in the literature is that between emotional and social loneliness. Emotional loneliness can be defined as the absence of positive experiences derived from close affective bonds, such as emotional support, love, and intimacy. On the other hand, social loneliness denotes the perceived lack of a network of social relationships which facilitate social integration and a sense of belonging. Emotional loneliness is typically associated with the abrupt disruption of an attachment relationship, whereas social loneliness usually emerges gradually in response to the absence of a social network. These forms of loneliness are considered distinct yet complementary rather than mutually exclusive, as adequacy in one relation domain does not compensate for deficits in another (Galanaki, 2014; Maes et al., 2022; Weiss, 1973). Moreover, both forms may vary by relational context. Current theoretical perspectives define adolescent loneliness as a multidimensional construct comprising social and emotional loneliness within salient relational contexts, particularly relationships with parents and peers (Goossens et al., 2009; Houghton et al., 2016; Maes et al., 2016).

Various instruments have been developed to assess loneliness, which either focus specifically on children and adolescents or have been adapted for use with these populations. Frequently cited measures include the Children’s Loneliness Scale (Asher et al., 1984), the Loneliness and Aloneness Scale for Children and Adolescents (Marcoen et al., 1987), the Peer Network and Dyadic Loneliness Scale (Hoza et al., 2000), the UCLA Loneliness Scale (Russell et al., 1980), and the Relational Provisions Loneliness Questionnaire (Hayden-Thomson, 1989). Each instrument has its own characteristics, advantages, and limitations (see Maes et al., 2022 for a comprehensive overview). Consequently, instrument selection should be guided by the specific research aims, the loneliness dimension of interest, and the relational context under investigation (Maes et al., 2022).

In the Greek context, the availability of psychometrically robust instruments for assessing loneliness in children and adolescents remains limited, and extant studies often report only basic reliability indices without comprehensive validation. One instrument that has been used in Greek research is the Children’s Loneliness Scale (Asher et al., 1984), adapted by Galanaki and Kalantzi-Azizi (1999). However, this instrument primarily captures the social dimension of loneliness and does not adequately assess emotional loneliness within peer relationships (Maes et al., 2022). Another instrument used with Greek samples is the UCLA Loneliness Scale (Russell et al., 1980), adapted in the Greek context by Anderson and Malikiosi-Loizos (1992). The measure is unidimensional and was originally developed to assess subjective feelings of loneliness and social isolation in adults (Maes et al., 2022).

The Relational Provisions Loneliness Questionnaire (RPLQ), developed by Hayden-Thomson (1989) in Canada, constitutes a multidimensional instrument intended for children and adolescents. The instrument is grounded in Weiss’s (1973) social provisions framework. The items assess loneliness across social and emotional dimensions within two primary relational contexts, namely peers and family. Specifically, the RPLQ comprises four dimensions: Lack of peer-group integration, Lack of peer-personal intimacy, Lack of family-group integration, and Lack of family-personal intimacy. The instrument has been translated into multiple languages and used in various countries, including Belgium, Portugal, Australia, and Greece.

Ribeiro et al. (2019) replicated the four-factor structure in Portuguese adolescents aged 12–15 years. Confirmatory factor analysis results showed excellent fit to the data (χ²/df = 2.69, CFI = .99, RMSEA = .05, MECVI = .26). In Belgian adolescents of comparable age, Maes et al. (2017) reported acceptable fit for the two-factor model (CFI = .911, RMSEA = .081, SRMR = .067) when using peer-related social and emotional loneliness. Similarly, Goossens and Beyers (2002) provided support for a two-factor model of loneliness distinguishing family- and peer-related loneliness, based on confirmatory factor analysis across several loneliness measures; however, their analyses did not explicitly evaluate the four-factor structure of the RPLQ.

With respect to interrelations among the RPLQ dimensions, Ribeiro et al. (2019) observed moderate correlations between Lack of peer-group integration and Lack of family-group integration (r = .44), and between Lack of peer-personal intimacy and Lack of family-personal intimacy (r = .31), suggesting related yet distinct constructs across contexts. Within the peer context, the association between Lack of peer-group integration and Lack of peer-personal intimacy was relatively strong (r = .58), whereas within the family context the correlation between Lack of family-group integration and Lack of family-personal intimacy (r = .79) was notably higher. A similar pattern was reported by Goossens and Beyers (2002), who found low-to-moderate correlations between Lack of peer-group integration and Lack of family-group integration (r = .34) and between Lack of peer-personal intimacy and Lack of family-personal intimacy (r = .27). Likewise, the correlation between Lack of peer-group integration and Lack of peer-personal intimacy (r = .51) was relatively high, whereas the correlation between Lack of family-group integration and Lack of family-personal intimacy (r = .70) was high. Consistent evidence was also provided by Maes et al. (2017), who found moderate-to-relatively high correlations between Lack of peer-group integration and Lack of peer-personal intimacy (r = .59). Overall, these findings suggest that adolescents may differentiate more clearly between integration and intimacy within peer relationships, whereas in the family context these relational provisions may overlap (Hayden-Thomson, 1989; Ribeiro et al., 2019).

Regarding internal consistency, the RPLQ subscales typically yield reliability coefficients ranging from acceptable (>.70) to excellent (>.90) (Galanaki et al., 2012; Goossens & Beyers, 2002; Maes et al., 2017; Ribeiro et al., 2019; Terrell-Deutsch, 1999). Measurement invariance across gender has been supported at the configural, metric, and scalar levels for the four subscales in Portuguese adolescents (Ribeiro et al., 2019), and at the metric and scalar levels for peer-related loneliness dimensions in Belgian adolescents (Maes et al., 2017). These findings indicate that boys and girls interpret the items in equivalent ways, permitting meaningful gender comparisons.

Despite evidence supporting the adequacy of RPLQ’s factor structure, internal consistency, and measurement invariance across gender, the broader empirical base remains limited given the relatively infrequent use of the instrument, highlighting the need for further validation research (Maes et al., 2022). Moreover, Greek literature lacks a psychometrically validated instrument for children and adolescents that adequately captures both social and emotional loneliness across the primary relational contexts in which loneliness is experienced, namely family and peers. From this perspective, the RPLQ offers a substantive methodological advantage by enabling a more comprehensive assessment of loneliness in youth.

The purpose of the study was to evaluate the psychometric properties of the RPLQ in a sample of Greek early adolescents. Specifically, the study examined the structural, convergent, and discriminant validity of the instrument, as well as the internal consistency of its dimensions. The concurrent validity of the instrument was assessed by examining associations between RPLQ dimensions and students’ sense of community in the school, a construct that has been consistently linked to lower loneliness (Cicognani et al., 2014; Pretty et al., 1994, 1996; Prezza et al., 2007). Finally, measurement invariance across gender was tested to determine whether the RPLQ functions equivalently for boys and girls.

Method

Participants

The sample included 503 secondary school students aged 13–14 years, of whom 275 (54.7%) were girls and 228 (45.3%) were boys. The majority of participants (n = 464; 92.2%) were Greek, whereas 39 students (7.8%) reported a different nationality. Stratified random sampling was used to enhance representativeness of the student target population. Data were collected from 20 public schools located within the metropolitan area of Thessaloniki, with an equal distribution across the western (n = 10) and the eastern (n = 10) sectors, thereby ensuring balanced geographical coverage in terms of school placement.

Measures

Relational Provisions Loneliness Questionnaire (Hayden-Thomson, 1989). The instrument assesses social and emotional loneliness within two central relational contexts, namely family and peers. It consists of 28 items organized into four dimensions of seven items each. More specifically, the dimension Lack of peer-group integration captures perceived inclusion and membership within the peer group (e.g., ‘I feel that I am part of a group of friends who do things together’). The dimension Lack of peer-personal intimacy evaluates perceived availability of a close friend with whom one can share feelings and thoughts (e.g., ‘I have at least one close friend I can talk to about what’s on my mind’). The dimension Lack of family-group integration assesses perceived belongingness and inclusion within the family system (e.g., ‘I feel that I fit in with my family members’). Finally, the dimension Lack of family-personal intimacy evaluates perceived capacity for open emotional expression within the family (e.g., ‘I have at least one person in my family with whom I can talk about what is bothering me’). Items are rated on a 5-point Likert-type scale ranging from 1 (“always”) to 5 (“never”). Dimension scores range from 7 to 35, with higher scores indicating greater loneliness.

The Greek version of the instrument, translated and culturally adapted by Galanaki et al. (2012) for students aged 10–11 years, was administered with permission from the original developers. Prior to the main data collection, a pilot study was conducted with a small group of students aged 13–14 years (N = 22), to test the item comprehension and linguistic appropriateness. No modifications were deemed necessary.

Sense of Community in the School (Prati et al., 2017). The instrument was developed to assess the students’ perceived sense of community at school, and consists of 10 items rated on a 5-point Likert-type scale. The scale operationalizes three dimensions: Membership (3 items), reflecting perceived belongingness, identification with the school, and emotional safety (e.g., ‘I am proud to belong to this school’); Emotional connection (3 items), referring to relational ties and affective bonds with other members of the school community (e.g., ‘In this school, I feel that I can share experiences and interests with other students’); and Opportunities for integration and fulfillment of needs (4 items), capturing perceived opportunities to meet personal needs and exert influence through community membership (e.g., ‘Students are involved in organizing a variety of school events’). Scale scores are calculated as item means for both the total scale and each subscale, with higher mean scores indicating a stronger sense of school community (Prati et al., 2017). Prior research has supported the instrument’s psychometric adequacy (Benavente et al., 2025; Prati & Cicognani, 2021; Prati et al., 2018). The Greek version of the instrument, which demonstrates satisfactory internal consistency for both total and subscale scores (α = .79–.85), was used in the present study (Vagiatis & Zissi, 2026). In the current sample, the measure exhibited high internal consistency (α = .84).

Procedure

Ethical approval was obtained from the Research Ethics Committee of the University of the Aegean, and the study was conducted in accordance with the ethical principles of the Declaration of Helsinki (2025). Subsequently, formal authorization was secured from the relevant Regional Education Authority and participating school administrations.

During the initial school visit, students received sealed envelopes containing an information letter and a parental consent form. In the subsequent phase, students for whom parental consent had been obtained completed the questionnaires via a paper-and-pencil format. Administration occurred collectively within the classroom during regular instructional time and was supervised by a member of the research team. Students were informed about the purpose of the study, the confidentiality of the data, the anonymous and voluntary nature of participation, and their right to withdraw at any stage. Data collection took place between February and April 2025.

Statistical Analysis

Data were analyzed using IBM SPSS Statistics (version 26) and R (version 4.5.1), specifically the lavaan and psych packages.

Missing data were examined prior to analysis. The distribution of missing values showed no evidence of a systematic pattern, and the proportion of missing values was low (<2%); therefore, no missing-data handling procedures were applied. Given that the sample was drawn from multiple schools, the Intraclass Correlation Coefficient (ICC) was calculated to assess potential clustering effects. The ICC value (.017) indicated a negligible clustering effect and did not warrant multilevel modelling. Descriptive statistics were computed for each subscale and for the total scale. Internal consistency was examined using Cronbach’s alpha (α) and McDonald’s omega (ω) reliability coefficients. The latter does not require the assumption of tau-equivalence and is thus considered more appropriate for multidimensional scales (Dunn et al., 2014). Values exceeding .70 indicate acceptable reliability (Awang, 2012).

Confirmatory factor analysis (CFA) was selected as the most appropriate method for examining structural validity, given that the RPLQ is grounded in a well-defined theoretical framework (Weiss, 1973) and has been documented as a four-factor instrument (Hayden-Thomson, 1989), with subsequent support across different sociocultural contexts (Goossens & Beyers, 2002; Ribeiro et al., 2019; Terrell-Deutsch, 1999). CFA models were estimated with the Weighted Least Squares Mean and Variance Adjusted (WLSMV) estimator, which is appropriate for Likert-type ordinal data, does not assume multivariate normality, and yields robust parameter estimates, standard errors, and model fit indices (Brown, 2015; Park, 2023). Three competing models were examined. The first was a unidimensional model, with all items specified to load on a single general loneliness factor. The second was a two-factor model organized by relational context. In this model, the items from the Lack of peer-group integration and Lack of peer-personal intimacy dimensions were specified to load on a Peer-related loneliness factor, whereas the items from the Lack of family-group integration and Lack of family-personal intimacy dimensions were specified to load on a Family-related loneliness factor. The third was the four-factor model, which corresponded to the four RPLQ dimensions described in the Measures section. Model fit was evaluated using the Comparative Fit Index (CFI ≥.90), Tucker-Lewis Index (TLI ≥.90), Root Mean Square Error of Approximation (RMSEA ≤.08, with 90% confidence interval), and Standardized Root Mean Square Residual (SRMR ≤.08) (Hu & Bentler, 1999; Kline, 2016). Convergent validity was assessed via standardized factor loadings and Average Variance Extracted (AVE) as an additional indicator of construct validity. Factor loadings greater than .50 and AVE values greater than .50 were taken as evidence of convergent validity (Brown, 2015; Rönkkö & Cho, 2022). Discriminant validity was examined through latent factor correlations and the Heterotrait-Monotrait ratio (HTMT). Latent correlations lower than .85 between factors, while HTMT values lower than .90 were interpreted as evidence of adequate discriminant validity (Henseler et al., 2015; Kline, 2016; Rönkkö & Cho, 2022).

Measurement invariance across gender was tested using multi-group confirmatory factor analysis (MG–CFA) with the WLSMV estimator at the configural, metric, and scalar (threshold) invariance levels to examine measurement equivalence between boys and girls. Invariance decisions were based on changes in fit indices, using ΔCFI ≤.010, ΔRMSEA ≤.015, and ΔSRMR ≤.030 as criteria for metric invariance, and ΔSRMR ≤.010 for scalar invariance (Chen, 2007). Independent samples t-tests were conducted to examine gender differences in subscales and total scores, facilitating comparability with prior research. Concurrent validity was evaluated using Pearson’s r correlations among the RPLQ subscales, the RPLQ total score, and the Sense of Community in the School scale, which is theoretically expected to be inversely associated with loneliness (Cicognani et al., 2014; Pretty et al., 1994).

Results

Descriptive Statistics

As shown in Table 1, descriptive statistics were computed for each RPLQ dimension and for the total scale. Across the four dimensions, mean scores ranged from 11.57 to 13.45, whereas the mean score for the total scale was 49.82. Standard deviations of the dimensions ranged from 4.27 to 5.29, while the standard deviation of the total scale was 15.56, indicating moderate response variability. The positive skewness (1.39–1.68) and elevated kurtosis (2.27–3.49) indicated that responses were concentrated toward the lower end of the scale, which is expected in a non-clinical sample.

Table 1.

Descriptive Statistics of Relational Provisions Loneliness Questionnaire (RPLQ) (N = 503)

Dimension	M	SD	Range	Skewness	Kurtosis
Lack of peer-group integration	13.45	4.36	7–35	1.39	2.31
Lack of peer-personal intimacy	12.80	4.27	7–35	1.40	2.27
Lack of family-group integration	11.57	5.29	7–35	1.68	2.64
Lack of family-personal intimacy	12.01	5.28	7–35	1.64	2.68
Total score	49.82	15.56	28–140	1.64	3.49

Confirmatory Factor Analysis

Confirmatory factor analysis (CFA) was conducted using the Weighted Least Squares Mean and Variance Adjusted (WLSMV) estimator, which yields robust parameter estimates and model-fit indices for ordinal indicators (Brown, 2015). Three alternative models were examined to evaluate the structure of the instrument: a unidimensional model, a two-factor model distinguishing peer- and family-related loneliness, and a four-factor model, which is consistent with the theoretically proposed structure (Table 2). The one-factor model demonstrated inadequate model fit (CFI = .910, TLI = .903, RMSEA = .179, SRMR = .168). The two-factor model showed an improved fit relative to the unidimensional model, yet remained only marginally acceptable (CFI = .981, TLI = .980, RMSEA = .082, SRMR = .082). The four-factor model demonstrated the best fit to the data (CFI = .997, TLI = .997, RMSEA = .032, SRMR = .050) relative to the other two models, supporting the proposed multidimensional structure of the instrument.

Table 2.

Model Fit Indices for Confirmatory Factor Analysis Models

Model	χ²	df	CFI	TLI	RMSEA	RMSEA 90% CI	SRMR
One-factor	5969.29	350	.910	.903	.179	[.175, .183]	.168
Two-factor	1523.21	349	.981	.980	.082	[.078, .086]	.082
Four-factor	518.29	344	.997	.997	.032	[.026, .037]	.050

Note. df = degrees of freedom; CFI = Comparative Fit Index; TLI = Tucker-Lewis Index; RMSEA = Root Mean Square Error of Approximation; CI = confidence interval; SRMR = Standardized Root Mean Squared Residual.

Standardized factor loadings for the four-factor model were uniformly high (λ = .63–.93) and statistically significant (p < .001), indicating that the items adequately captured the respective latent dimensions and supporting convergent validity. Specifically, factor loadings ranged from .63 to .81 for Lack of peer-group integration, .71 to .93 for Lack of peer-personal intimacy, .70 to .90 for Lack of family-group integration, and .79 to .90 for Lack of family-personal intimacy (Table 3).

Table 3.

Standardized Factor Loadings for the Four-Factor Model

Item	Factor loading
Item	1	2	3	4
Factor 1: Lack of peer-group integration
Item 1	.748
Item 3	.756
Item 5	.813
Item 7	.791
Item 9	.803
Item 11	.632
Item 13	.650
Factor 2: Lack of peer-personal intimacy
Item 2		.709
Item 4		.714
Item 6		.931
Item 8		.790
Item 10		.788
Item 12		.732
Item 14		.779
Factor 3: Lack of family-group integration
Item 15			.797
Item 17			.857
Item 19			.896
Item 21			.857
Item 23			.877
Item 25			.704
Item 27			.872
Factor 4: Lack of family-personal intimacy
Item 16				.902
Item 18				.903
Item 20				.886
Item 22				.787
Item 24				.875
Item 26				.850
Item 28				.786

Note. All standardized loadings are statistically significant (p < .001).

Convergent validity was additionally assessed using the Average Variance Extracted (AVE) index. The AVE values for the four latent factors were .56, .61, .71 and .74, all exceeding the conventional criterion of .50 (Brown, 2015; Rönkkö & Cho, 2022). These findings indicate that each factor accounts for a satisfactory proportion of variance in its observed indicators, further supporting the convergent validity of the RPLQ.

Discriminant validity was assessed via latent factor intercorrelations and the Heterotrait-Monotrait ratio (HTMT). As shown in Table 4, latent correlations were predominantly moderate, although the correlation between the two family-related dimensions slightly exceeded the .85 criterion (r = .87; Kline, 2016; Rönkkö & Cho, 2022).

Table 4.

Latent Factor Correlations

Dimension	1	2	3	4
1. Lack of peer-group integration	—
2. Lack of peer-personal intimacy	.60	—
3. Lack of family-group integration	.49	.32	—
4. Lack of family-personal intimacy	.42	.36	.87	—

Note. All correlations are statistically significant (p < .001).

HTMT values were uniformly below the recommended threshold of .90 (Henseler et al., 2015). A higher correlation (r = .87) and a higher HTMT value (.896) were observed between the two family-related dimensions, a pattern that is theoretically consistent with the conceptual overlap between integration and intimacy within the family (Hayden-Thomson, 1989; Ribeiro et al., 2019). Overall, the results support the discriminant validity of the RPLQ (Table 5).

Table 5.

Heterotrait-Monotrait Ratio (HTMT)

Dimension	1	2	3	4
1. Lack of peer-group integration	—
2. Lack of peer-personal intimacy	.620	—
3. Lack of family-group integration	.506	.358	—
4. Lack of family-personal intimacy	.437	.385	.896	—

Internal Consistency Reliability

Cronbach’s alpha (α) coefficients for the subscales Lack of peer-group integration, Lack of peer-personal intimacy, Lack of family-group integration, and Lack of family-personal intimacy were .87, .86, .91, and .93, respectively. McDonald’s omega (ω) coefficients for the same dimensions were .87, .87, .91, and .92, respectively. For the total scale, Cronbach’s alpha and McDonald’s omega were .93 and .94, respectively (Table 6). According to Awang (2012) and Dunn et al. (2014), these coefficients indicate high internal consistency for both the subscales and the overall score.

Table 6.

Internal Consistency Reliability Indices

Dimension	Items	Cronbach’s α	McDonald’s ω
Lack of peer-group integration	7	.867	.871
Lack of peer-personal intimacy	7	.862	.868
Lack of family-group integration	7	.910	.914
Lack of family-personal intimacy	7	.919	.921
Total score	28	.934	.936

Measurement Invariance Across Gender

Measurement invariance across gender was examined for the four-factor model using multi-group confirmatory factor analysis (MG–CFA) with the WLSMV estimator (Table 7). First, configural invariance was examined, indicating excellent model fit across gender groups. Next, metric invariance was tested by constraining factor loadings to be equivalent across groups. Changes in the fit indices were within acceptable limits (ΔCFI = −.002, ΔRMSEA = .008, ΔSRMR = .004), supporting metric invariance and allowing comparisons of structural relations between boys and girls. Finally, scalar (threshold) invariance was tested by constraining item thresholds to be equivalent across groups. Changes in fit indices relative to the metric model remained within the proposed limits (ΔCFI = .002, ΔRMSEA = −.009, ΔSRMR = −.004), indicating that the restrictions did not affect the model’s fit. These results suggest that observed scores are related to the latent loneliness dimensions in an equivalent manner for boys and girls.

Table 7.

Measurement Invariance Across Gender

Model	CFI	TLI	RMSEA	SRMR	ΔCFI	ΔRMSEA	ΔSRMR
Configural	.997	.997	.033	.063	—	—	—
Metric	.996	.995	.041	.067	−.002	.008	.004
Scalar	.997	.997	.031	.063	.002	−.009	−.004

Note. CFI = Comparative Fit Index; TLI = Tucker-Lewis Index; RMSEA = Root Mean Square Error of Approximation; SRMR = Standardized Root Mean Squared Residual; ΔCFI = change in Comparative Fit Index; ΔRMSEA = change in Root Mean Square Error of Approximation; ΔSRMR = change in Standardized Root Mean Squared Residual.

Independent-samples t-tests were then conducted to examine gender differences in the observed scores. Girls reported significantly higher levels of loneliness than boys in Lack of peer-group integration (t(501) = −2.99, p = .003) and Lack of family-group integration (t(501) = −3.32, p = .001). No significant gender differences were observed for Lack of peer-personal intimacy (t(501) = 1.65, p = .100) and Lack of family-personal intimacy (t(501) = −1.68, p = .093). At the total-score level, girls also reported significantly higher loneliness than boys (t(501) = −2.08, p = .038).

Concurrent Validity

Evidence for concurrent validity was examined through correlations between RPLQ scores and Sense of Community in the School (SOC-S) scores. As shown in Table 8, correlations between SOC-S and each RPLQ dimension, as well as the total RPLQ score, were negative and statistically significant (p < .01). Coefficients ranged from −.30 to −.39 for the subscales and reached −.44 for the total score. The findings suggest that students reporting higher levels of loneliness tend to report lower perceived sense of school community (Cicognani et al., 2014; Pretty et al., 1994). Overall, these findings provide support for the concurrent validity of the instrument.

Table 8.

Correlations Between Relational Provisions Loneliness Questionnaire (RPLQ), Its Dimensions and Sense of Community in the School (SOC-S)

Dimension	1	2	3	4	5	6
1. Lack of peer-group integration	—
2. Lack of peer-personal intimacy	.53	—
3. Lack of family-group integration	.42	.27	—
4. Lack of family-personal intimacy	.39	.30	.80	—
5. RPLQ	.74	.66	.83	.84	—
6. SOC-S	−.39	−.30	−.35	−.33	−.44	—

Note. All correlations are statistically significant (p < .001).

Discussion

The lack of a psychometrically validated multidimensional instrument for assessing loneliness in Greek adolescents – one that captures both social and emotional dimensions across the family and peer relational contexts – constituted the primary rationale for the present study. Additionally, given the relatively limited use of the Relational Provisions Loneliness Questionnaire (RPLQ) in prior research, further investigation of its psychometric properties was warranted.

Accordingly, the purpose of the study was to evaluate the psychometric properties of the RPLQ in a sample of Greek early adolescents, including its factorial structure, internal consistency, convergent, discriminant, and concurrent validity, as well as its measurement invariance across gender.

Consistent with Weiss’s (1973) theoretical framework and prior empirical findings supporting the four-factor model structure of the instrument (Goossens & Beyers, 2002; Hayden-Thomson, 1989; Ribeiro et al., 2019), confirmatory factor analysis was employed as the most appropriate method for evaluating the latent structure. The results indicated that the four-factor model, comprising Lack of peer-group integration, Lack of peer-personal intimacy, Lack of family-group integration, and Lack of family-personal intimacy, demonstrated the best fit to the data relative to the unidimensional and two-factor models, in line with the above studies. Standardized factor loadings were uniformly high and statistically significant (.63–.93), and Average Variance Extracted values exceeded the recommended threshold of .50 for all latent factors. These findings provide evidence supporting the convergent validity, as the items of each factor represent the latent dimensions. Overall, the CFA fit indices in combination with the high standardized factor loadings and the Average Variance Extracted values indicate that the four-factor model, which differentiates the dimensions of integration and intimacy across peers and family contexts, represents the optimal psychometric structure of the instrument in the examined sample. Importantly, this structure is consistent with Weiss’s (1973) conceptualization of loneliness as a multidimensional phenomenon that varies according to both the nature of relational provisions (integration versus intimacy) and the relational domains in which deficits occur (family versus peers).

Discriminant validity was assessed through latent factor correlations and the Heterotrait-Monotrait ratio (HTMT). Correlations between peer and family dimensions were moderate (.32–.49), indicating that, although loneliness experiences across contexts are associated, they remain conceptually distinct. These correlations document that elevated loneliness in one relational context does not entail comparable loneliness in the other. Within the peer context, the moderate to relatively high correlation (.60) between the dimensions of integration and intimacy suggests that they are related, although they capture distinct aspects of adolescents’ loneliness. The differentiation between peer-group integration and close peer emotional bonds provides support for the multidimensional nature of loneliness. Given that adolescents spend a substantial portion of their daily lives in the school environment – the primary setting for peer interaction – they may experience a sense of belonging without necessarily developing dyadic relationships, and vice versa (Goossens & Beyers, 2002; Maes et al., 2017; Zissi & Chiou, 2017). By contrast, a high correlation (.87) was observed between the dimensions of integration and intimacy within the family. This association does not necessarily indicate conceptual overlap; rather, it is consistent with theoretical expectations regarding the structure of family relationships, wherein integration and intimacy frequently co-occur and reinforce one another, as the family simultaneously functions as a primary context of both attachment and belonging (Hayden-Thomson, 1989; Ribeiro et al., 2019; Weiss, 1973). From a methodological standpoint, this interpretation aligns with the caution articulated by Rönkkö and Cho (2022), who argue that latent correlations should be interpreted as indicators of construct relatedness rather than as definitive evidence of construct non-distinctiveness. Additionally, HTMT values for all factors did not exceed the recommended threshold (<.90) (Henseler et al., 2015; Kline, 2016; Rönkkö & Cho, 2022). Taken together, the findings support the discriminant validity of the measure.

Concurrent validity was supported through statistically significant, negative, and moderate correlations with Sense of Community in the School. Specifically, students who reported higher levels of loneliness also reported lower levels of sense of school community, a finding consistent with previous research (Cicognani et al., 2014; Pretty et al., 1994; Pretty et al., 1996; Prezza et al., 2007). The direction of the correlations is consistent with theoretical expectations that loneliness is associated with social integration in the school context, whereas the moderate magnitude is compatible with the view that loneliness is a subjective and contextually embedded experience shaped by multiple determinants beyond sense of community at school alone.

Cronbach’s alpha and McDonald’s omega indicated high internal consistency reliability for each dimension and for the overall scale (α, ω > .80). This pattern replicates findings reported in prior studies using the RPLQ in various sociocultural contexts (Galanaki et al., 2012; Goossens & Beyers, 2002; Maes et al., 2017; Ribeiro et al., 2019; Terrell-Deutsch, 1999).

Measurement invariance across gender was supported at the configural, metric, and scalar levels, indicating that the measure functions equivalently for boys and girls and enabling meaningful comparisons of latent means and observed scores. In the present study, statistically significant gender differences were found, with girls reporting higher loneliness in the integration dimensions within both peer and family contexts, as well as on the total scale. This pattern is congruent with some prior evidence suggesting higher reported loneliness among girls (Goossens & Marcoen, 1999). A plausible explanation for these differences concerns gender-differentiated relational orientations, whereby girls may place greater emphasis on emotional closeness and relational quality, whereas boys may prioritize broader group-based affiliation (Maes et al., 2017). However, it is important to note that the broader literature on gender differences remains inconsistent, with meta-analytic evidence indicating mixed findings across studies (Maes et al., 2019).

Several limitations of the current study should be acknowledged. First, the sample was drawn from a single geographical region of Greece and from a specific age group, thereby constraining the generalizability of the findings to other sociocultural contexts and developmental periods. Second, concurrent validity was assessed using a single external criterion, and temporal stability of the measure was not examined. Future research should therefore test the instrument across different age groups as well as more diverse geographical and sociocultural contexts. Third, the validation evidence could be strengthened by incorporating additional external criteria, including alternative loneliness measures and theoretically relevant constructs. Finally, the assessment of test-retest reliability would provide evidence of measurement stability over time.

In conclusion, the Greek version of the RPLQ appears to constitute a valid and reliable instrument for assessing loneliness in early adolescence. Its multidimensional structure permits the differentiation between integration and intimacy dimensions of loneliness across family and peer contexts, thereby providing a more nuanced understanding of adolescents’ loneliness experiences. In this regard, the instrument addresses a substantive methodological gap in the Greek literature and may also contribute to the broader international evidence base, given the instrument’s relatively limited use. The measure may facilitate the early identification of adolescents experiencing loneliness and inform the design and evaluation of interventions aimed at reducing its adverse effects.

Footnotes

Acknowledgements

The authors are grateful to the participating schools, students, and their families for their contribution to this research.

ORCID iD

Konstantinos Vagiatis

Ethical Considerations

All procedures performed in this study were approved by the Research Ethics Committee of the University of the Aegean (Approval No. 28599/31-08-2023) and were conducted in accordance with the ethical principles of the Declaration of Helsinki. The necessary permissions were also obtained from the relevant Regional Education Authority of the Greek Ministry of Education, Religious Affairs and Sports (Approval No. 26382/29-12-2023).

Consent to Participate

Written informed consent was obtained from the parents or legal guardians of all participating students prior to data collection. Students also provided their informed assent.

Author Contributions

K.V. conceived the study, conducted the statistical analyses, and drafted the manuscript.

A.Z. supervised the study and contributed to the conceptualization and revision of the manuscript.

E.G. contributed to the interpretation of the findings and critically revises the manuscript.

All authors read and approved the final version of the manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on request.*

Author Biographies

Konstantinos Vagiatis is a PhD Candidate at the Department of Sociology, School of Social Sciences, University of the Aegean, Greece. His research interests include educational inequalities, cultural logics of parenting, and psychosocial processes related to academic achievement.

Anastasia Zissi is a Professor of Psychology at the Department of Sociology, School of Social Sciences, University of the Aegean, Greece. Her research interests include community and mental health, psychosocial intervention design and organization, needs assessment, quality-of-life assessment, and socially marginalized populations.

Evangelia Galanaki is a Professor of Developmental Psychology and Director of the Psychology Laboratory at the Department of Pedagogy and Primary Education, School of Education, National and Kapodistrian University of Athens, Greece. Her research interests include emerging adulthood, transition to adulthood, solitude, and loneliness.

References

Anderson

L. R.

Malikiosi-Loizos

(1992). Reliability data for a Greek translation of the revised UCLA loneliness scale: Comparisons with data from the USA. Psychological Reports, 71(2), 665–666. https://doi.org/10.2466/pr0.1992.71.2.665

Asher

S. R.

Hymel

Renshaw

P. D.

(1984). Loneliness in children. Child Development, 55(4), 1456–1464. https://doi.org/10.2307/1130015

Awang

(2012). Research methodology and data analysis (2nd ed.): UiTM Press.

Benavente

Chuecas

M. J.

Alfaro

Galdames

Ditzel

A. L.

(2025). Psychometric properties of the school sense of community scale (SOC-S) for a sample of Chilean adolescents. Journal of Psychoeducational Assessment, 43(6), 665–673. https://doi.org/10.1177/07342829251345975

Benner

A. D.

(2011). Latino adolescents’ loneliness, academic performance, and the buffering nature of friendships. Journal of Youth and Adolescence, 40(5), 556–567. https://doi.org/10.1007/s10964-010-9561-2

Brown

T. A.

(2015). Confirmatory factor analysis for applied research (2nd ed.): Guilford Press.

Cacioppo

J. T.

Hawkley

L. C.

Ernst

J. M.

Burleson

Berntson

G. G.

Nouriani

Spiegel

(2006). Loneliness within a nomological net: An evolutionary perspective. Journal of Research in Personality, 40(6), 1054–1085. https://doi.org/10.1016/j.jrp.2005.11.007

Chen

F. F.

(2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14(3), 464–504. https://doi.org/10.1080/10705510701301834

Cicognani

Klimstra

Goossens

(2014). Sense of community, identity statuses, and loneliness in adolescence: A cross‐national study on Italian and Belgian youth. Journal of Community Psychology, 42(4), 414–432. https://doi.org/10.1002/jcop.21618

10.

Dunn

T. J.

Baguley

Brunsden

(2014). From alpha to omega: A practical solution to the pervasive problem of internal consistency estimation. British Journal of Psychology, 105(3), 399–412. https://doi.org/10.1111/bjop.12046

11.

Eccles

A. M.

Qualter

Madsen

K. R.

Holstein

B. E.

(2021). Loneliness and scholastic self-beliefs among adolescents: A population-based survey. Scandinavian Journal of Educational Research, 67(1), 97–112. https://doi.org/10.1080/00313831.2021.1983865

12.

Galanaki

(2014). Loneliness: The paradox of human nature: Gutenberg.

13.

Galanaki

Amanaki

Xiki

(2012). Bullying/Victimization and children’s loneliness in school and family. Educational Review: The Journal of the Hellenic Educational Society, 54(3), 37–56. https://doi.org/10.26266/jpevol54pp37-56

14.

Galanaki

Kalantzi-Azizi

(1999). Loneliness and social dissatisfaction: Its relation with children's self-efficacy for peer interaction. Child Study Journal, 29(1), 1–22.

15.

Goossens

Beyers

(2002). Comparing measures of childhood loneliness: Internal consistency and confirmatory factor analysis. Journal of Clinical Child and Adolescent Psychology, 31(2), 252–262. https://doi.org/10.1207/S15374424JCCP3102_10

16.

Goossens

Lasgaard

Luyckx

Vanhalst

Mathias

Masy

(2009). Loneliness and solitude in adolescence: A confirmatory factor analysis of alternative models. Personality and Individual Differences, 47(8), 890–894. https://doi.org/10.1016/j.paid.2009.07.011

17.

Goossens

Marcoen

(1999). Relationships during adolescence: Constructive vs. negative themes and relational dissatisfaction. Journal of Adolescence, 22(1), 65–79. https://doi.org/10.1006/jado.1998.0201

18.

Hawkley

L. C.

Cacioppo

J. T.

(2010). Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Annals of Behavioral Medicine, 40(2), 218–227. https://doi.org/10.1007/s12160-010-9210-8

19.

Hayden-Thomson

L. K.

(1989). The development of the relational provisions loneliness questionnaire for children: University of Waterloo. PhD Thesis.

20.

Heinrich

L. M.

Gullone

(2006). The clinical significance of loneliness: A literature review. Clinical Psychology Review, 26(6), 695–718. https://doi.org/10.1016/j.cpr.2006.04.002

21.

Henseler

Ringle

C. M.

Sarstedt

(2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science, 43(1), 115–135. https://doi.org/10.1007/s11747-014-0403-8

22.

Houghton

Hattie

Carroll

Wood

Baffour

(2016). It hurts to be lonely! loneliness and positive mental wellbeing in Australian rural and urban adolescents. Journal of Psychologists and Counsellors in Schools, 26(1), 52–67. https://doi.org/10.1017/jgc.2016.1

23.

Houghton

Hattie

Wood

Carroll

Martin

Tan

(2014). Conceptualising loneliness in adolescents: Development and validation of a self-report instrument. Child Psychiatry and Human Development, 45(5), 604–616. https://doi.org/10.1007/s10578-013-0429-z

24.

Hoza

Bukowski

W. M.

Beery

(2000). Assessing peer network and dyadic loneliness. Journal of Clinical Child Psychology, 29(1), 119–128. https://doi.org/10.1207/S15374424jccp2901_12

25.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

26.

Jefferson

Barreto

Jones

Conway

Chohan

Madsen

K. R.

Verity

Petersen

K. J.

Qualter

(2023). Adolescent loneliness across the world and its relation to school climate, national culture and academic performance. British Journal of Educational Psychology, 93(4), 997–1016. https://doi.org/10.1111/bjep.12616

27.

Kline

R. B.

(2016). Principles and practice of structural equation modeling (4th ed.): Guilford Press.

28.

Larson

R. W.

Richards

M. H.

Moneta

Holmbeck

Duckett

(1996). Changes in adolescents’ daily interactions with their families from ages 10 to 18: Disengagement and transformation. Developmental Psychology, 32(4), 744–754. https://doi.org/10.1037/0012-1649.32.4.744

29.

Maes

Qualter

Lodder

G. M.

Mund

(2022). How (not) to measure loneliness: A review of the eight most commonly used scales. International Journal of Environmental Research and Public Health, 19(17), 10816. https://doi.org/10.3390/ijerph191710816

30.

Maes

Qualter

Vanhalst

Van den Noortgate

Goossens

(2019). Gender differences in loneliness across the lifespan: A meta–analysis. European Journal of Personality, 33(6), 642–654. https://doi.org/10.1002/per.2220

31.

Maes

Vanhalst

Spithoven

A. W. M.

Van den Noortgate

Goossens

(2016). Loneliness and attitudes toward aloneness in adolescence: A person-centered approach. Journal of Youth and Adolescence, 45(3), 547–567. https://doi.org/10.1007/s10964-015-0354-5

32.

Maes

Vanhalst

Van den Noortgate

Goossens

(2017). Intimate and relational loneliness in adolescence. Journal of Child and Family Studies, 26(8), 2059–2069. https://doi.org/10.1007/s10826-017-0722-8

33.

Mahon

N. E.

Yarcheski

T. J.

Cannella

B. L.

Hanks

M. M.

(2006). A meta-analytic study of predictors for loneliness during adolescence. Nursing Research, 55(5), 308–315. https://doi.org/10.1097/00006199-200609000-00003

34.

Marcoen

Goossens

Caes

(1987). Loneliness in pre-through late adolescence: Exploring the contributions of a multidimensional approach. Journal of Youth and Adolescence, 16(6), 561–577. https://doi.org/10.1007/BF02138821

35.

Park

C. G.

(2023). Implementing alternative estimation methods to test the construct validity of Likert-scale instruments. Korean Journal of Women Health Nursing, 29(2), 85–90. https://doi.org/10.4069/kjwhn.2023.06.14.2

36.

Perlman

Peplau

L. A.

(1981). Towards a social psychology of loneliness. In Duck

Gilmour

(Eds.), Personal relationships in disorder (pp. 31–56): Academic Press.

37.

Prati

Cicognani

(2021). School sense of community as a predictor of well-being among students: A longitudinal study. Current Psychology, 40(2), 939–943. https://doi.org/10.1007/s12144-018-0017-2

38.

Prati

Cicognani

Albanesi

(2017). Psychometric properties of a multidimensional scale of sense of community in the school. Frontiers in Psychology, 8, 1466. https://doi.org/10.3389/fpsyg.2017.01466

39.

Prati

Cicognani

Albanesi

(2018). The influence of school sense of community on students’ well‐being: A multilevel analysis. Journal of Community Psychology, 46(7), 917–924. https://doi.org/10.1002/jcop.21982

40.

Pretty

G. M.

Conroy

Dugay

Fowler

Williams

(1996). Sense of community and its relevance to adolescents of all ages. Journal of Community Psychology, 24(4), 365–379. https://doi.org/10.1002/(SICI)1520-6629(199610)24:4%3C365::AID-JCOP6%3E3.0.CO;2-T

41.

Pretty

G. M. H.

Andrewes

Collett

(1994). Exploring adolescents' sense of community and its relationship to loneliness. Journal of Community Psychology, 22(4), 346–358. https://doi.org/10.1002/1520-6629(199410)22:4%3C346::AID-JCOP2290220407%3E3.0.CO;2-J

42.

Prezza

Pacilli

M. G.

(2007). Current fear of crime, sense of community, and loneliness in Italian adolescents: The role of autonomous mobility and play during childhood. Journal of Community Psychology, 35(2), 151–170. https://doi.org/10.1002/jcop.20140

43.

Qualter

Brown

S. L.

Munn

Rotenberg

K. J.

(2010). Childhood loneliness as a predictor of adolescent depressive symptoms: An 8-year longitudinal study. European Child & Adolescent Psychiatry, 19(6), 493–501. https://doi.org/10.1007/s00787-009-0059-y

44.

Qualter

Brown

S. L.

Rotenberg

K. J.

Vanhalst

Harris

Goossens

Bangee

Munn

(2013). Trajectories of loneliness during childhood and adolescence: Predictors and health outcomes. Journal of Adolescence, 36(6), 1283–1293. https://doi.org/10.1016/j.adolescence.2013.01.005

45.

Qualter

Vanhalst

Harris

Van Roekel

Lodder

Bangee

Maes

Verhagen

(2015). Loneliness across the life span. Perspectives on Psychological Science, 10(2), 250–264. https://doi.org/10.1177/1745691615568999

46.

Ribeiro

Santos

A. J.

Freitas

Rosado

Rubin

K. H.

(2019). Loneliness in adolescence: Confirmatory factor analysis of the relational provisions loneliness questionnaire (RPLQ) in a Portuguese sample. International Journal of Behavioral Development, 43(5), 457–465. https://doi.org/10.1177/0165025419850893

47.

Rönkkö

Cho

(2022). An updated guideline for assessing discriminant validity. Organizational Research Methods, 25(1), 6–14. https://doi.org/10.1177/1094428120968614

48.

Russell

Peplau

L. A.

Cutrona

C. E.

(1980). The revised UCLA loneliness scale: Concurrent and discriminant validity evidence. Journal of Personality and Social Psychology, 39(3), 472–480. https://doi.org/10.1037/0022-3514.39.3.472

49.

Terrell-Deutsch

(1999). The conceptualization and measurement of childhood loneliness. In Rotenberg

K. J.

Hymel

(Eds.), Loneliness in childhood and adolescence (pp. 11–33): University Press.

50.

Twenge

J. M.

Haidt

Blake

A. B.

McAllister

Lemon

Le Roy

(2021). Worldwide increases in adolescent loneliness. Journal of Adolescence, 93(1), 257–269. https://doi.org/10.1016/j.adolescence.2021.06.006

51.

Vagiatis

Zissi

(2026). Greek adaptation and psychometric validation of the scale of sense of community in the school. Social Work. Review of Social Sciences, 40(1), 94–112. https://doi.org/10.12681/socialwork-rss.43556

52.

Van Rode

Rotsaert

Delhaye

(2015). Loneliness and adolescence: Clinical implications and outlook: Literature review. Revue Medicale de Bruxelles, 36(5), 415–420.

53.

Weiss

R. S.

(1973). Loneliness: The experience of emotional and social isolation: MIT Press.

54.

World Medical Association . (2025). World Medical Association Declaration of Helsinki: Ethical principles for medical research involving human participants. JAMA, 333(1), 71–74. https://doi.org/10.1001/jama.2024.21972

55.

Zissi

Chiou

(2017). Material conditions, social capital and mental health of children and adolescents: A brief review. Psychology: The Journal of the Hellenic Psychological Society, 22(1), 30–43. https://doi.org/10.12681/psy_hps.23265