Validation study of the Croatian version of the Three-Item Loneliness Scale and its association to social network indices

Abstract

Background

Loneliness is a prevalent and complex phenomenon influenced by various social and demographic factors. The goal of this study was to translate the Three-Item UCLA Loneliness Scale into Croatian and test its validity and reliability in practice. An additional aim was to explore the relationship between social connectedness and loneliness within the Croatian population, with respect to age and gender.

Method

The Three-Item Loneliness Scale was translated into Croatian and validated according to widely accepted guidelines for cross-cultural adaptation and translation of the questionnaire into different languages. Subsequently, loneliness and social connectedness were examined in 309 Croatian participants using an anonymous online survey.

Results

The results indicated a significant association between social connectedness and loneliness. In particular, those with fewer individuals in their social network were associated with higher loneliness. Notably, age had a significant negative association with loneliness, indicating increased loneliness among younger individuals. Meanwhile older individuals with fewer active network domains experienced heightened loneliness compared to other age groups. No effect of gender was observed.

Conclusion

Our findings highlight the importance of social connectedness in relation to loneliness, particularly among younger and older age individuals. This study, through its validation of the instrument for measuring loneliness, offers important insights into the unique Croatian context, emphasizing how cultural and social factors may influence the experience of loneliness and social connectedness.

Introduction

Loneliness and social connectedness represent two fundamental dimensions of human experience, each exerting profound effects on individual health and societal dynamics Wang et al. [1]; Vila [2]. Loneliness is often described as a complex and multifaceted emotional experience arising when an individual perceives a discrepancy between their desired level of social connection and their actual level of social interaction Peplau & Perlman [3]; Cacioppo et al. [4]. It is frequently captured within a variety of related constructs, such as depression, shyness, social anxiety and/or solitude, and more prominently, social isolation Pearce et al. [5]; Veazie et al. [6]; Anyan & Hjemdal [7]. Although these constructs may be interrelated, they are distinct: loneliness specifically refers to the perceived inadequacy of social relationships and the distressing feeling of being alone, even in the presence of others Yanguas et al., [8]. Loneliness has been consistently linked to adverse mental health outcomes such as depression, anxiety, and low self-esteem Nowland et al. [9]; Elovainio et al. [10]; Gizdic et al. [11]. It can also contribute to emotional distress, dysregulation, and a diminished sense of social belonging Yanguas et al. [8]. Moreover, chronic loneliness has been associated with detrimental effects on physical health, such as an increased risk of cardiovascular disease and mortality Paul et al. [12]. However, measuring loneliness remains a persistent challenge in research, as studies have highlighted difficulties in comparing prevalence estimates due to inconsistencies in survey tools and definitions Surkalim et al. [13]. While global research has significantly advanced our understanding of loneliness, it is essential to consider country-specific factors—such as cultural norms, demographic trends, and social structures—to conduct assessments that accurately reflect how loneliness is experienced in different populations. As such, Croatia currently lacks a widely validated, brief, and reliable instrument for assessing loneliness, which limits the ability to systematically measure its prevalence and develop targeted interventions. Given the cultural and linguistic specificities of the Croatian population, as well as pressing public health concerns (e.g., population aging, rural depopulation, and high rates of youth emigration; Nejašmić & Toskić [14]; Grgić et al. [15], a validated adaptation of the UCLA Loneliness Scale is essential to ensure its psychometric reliability and applicability Ćubela & Nekić [16]. While younger individuals may experience loneliness due to lifestyle transitions, education, and employment changes, older adults often face reduced social networks due to retirement, health limitations, or the loss of social connections. Despite these concerns, systematic data collection on loneliness remains scarce. Studies indicate that loneliness is a significant issue among older adults in Croatia, with variations observed based on living arrangements and other sociodemographic factors Neuberg et al. [17].

Social connectedness, on the other hand, described as a combination of interconnected elements encompassing social networks and engagement, social support, and the lack of perceived social isolation has been considered a predictor of a successful life, standing as the primary motivating force driving social behavior Lamblin et al. [18]; Waldinger & Schulz [19]; Choi et al. [20]. It has been associated with numerous social and health-related benefits and has been shown to promote resilience, enhance positive mood, and protect against trauma and stress Macià et al. [21]; Gizdic et al. [22]. Individuals with strong social connectedness tend to experience greater levels of life satisfaction and overall well-being Ehsan et al. [23]; Wickramaratne et al. [24].

While loneliness and social connectedness are often perceived as opposing concepts, they are intricately linked, as social connectedness serves as a protective factor against loneliness Waldinger & Schulz [19]. Strong social connections have been shown to reduce feelings of loneliness by providing emotional, psychological, and practical support Yelpaze et al. [25]; McLoughlin et al. [26]; Rustamov et al. [27]. Conversely, individuals who lack social connections or feel socially disconnected are more prone to loneliness Vella et al. [28]. The dynamics of loneliness and social connectedness vary across cultural contexts and demographic factors such as age, gender, education, and living situation, though findings have been inconsistent. While some factors like marriage or cohabitation, have received strong empirical support, others remain less clear Waldinger & Schulz [19]. Impacts of age, gender, and culture on loneliness remain ambiguous, with limited exploration in the literature on how these factors intersect with social connectedness to predict loneliness. For example, studies have shown conflicting findings regarding which demographic groups are most vulnerable to loneliness. While some suggest that women, older individuals, and those with lower levels of education are at increased risk Boehlen et al. [29]; Fernández-Carro & Gumà Lao [30], others have shown that younger adults, especially men, are more vulnerable to loneliness Ernst et al. [31]. These disparities may be attributed to cultural influences Barreto et al. [32]. In general, people in countries that emphasize individualism and independence tend to report higher levels of loneliness and fewer social connections compared to those living in cultures that promote collectivism and interdependence Schermer et al. [33]; Barreto et al. [32]. However, while these trends are well-documented, there is a gap in our understanding, particularly concerning the Croatian population, which tends to exhibit a mix of both individualistic and collectivistic traits Rajh et al. [34]; Sheldon et al. [35]. There is limited research on the nuanced relationship between loneliness and social connectedness within this population, leaving gaps in our understanding of how Croatia's unique cultural and social factors may influence these dynamics. Therefore, investigating the relationship between loneliness and social connectedness may provide valuable insights into strategies for mitigating loneliness and enhancing social well-being, both within specific populations , such as Croatians, and on a global scale.

The present study

The primary aim of the present study was to adapt the Three-Item Loneliness Scale into Croatian, ensuring its validity and reliability through translation and validation procedures. The Three-Item Loneliness Scale, developed by Hughes et al. [36], is a widely recognized instrument designed to measure subjective feelings of loneliness. The scale includes three questions that assess the extent to which individuals feel disconnected from others and lack of social companionship. By translating and adapting this scale into Croatian, the study aimed to create a reliable tool for assessing loneliness in the Croatian population.

In addition to ensuring the scale's validity and reliability, the study sought to explore the potential impact of both the quantity and quality of social networks on feelings of loneliness. Specifically, we investigated whether having more social contacts and stronger social relationships was associated with lower levels of loneliness. Furthermore, we examined how demographic factors such as age and gender might influence the relationship between social connectedness and loneliness within the Croatian population. Given the international relevance of loneliness as a psychological and social construct, it is important to place this study within a broader global context. Loneliness and social connectedness are fundamental aspects of human experience, with significant implications for both mental health and societal well-being. The dynamics between loneliness and social connectedness may vary across different cultures, societies, and demographic groups, emphasizing the need to understand these phenomena within the context of local cultural norms and social structures. We hypothesized that individuals with a higher number of social contacts and greater quality of social relationships would experience less loneliness. While we did not set an a priori hypothesis regarding age and gender, we recognize that these demographic factors may play a role in shaping the experience of loneliness. Thus, our findings may offer insights that are applicable across different countries and cultures.

Methodology

Description of the scale

The most commonly used scale for assessing loneliness is the University of California Los Angeles (UCLA) Loneliness Scale Russell et al. [37]. It captures a general sense of loneliness without specifically delineating particular aspects or sources of loneliness, emphasizing subjective feelings of loneliness and social isolation. This scale has evolved through three versions, with the third being UCLA Loneliness Scale Version 3 Russell [38]. The Three-Item Loneliness Scale, adapted and refined by Hughes et al. [36], is particularly noteworthy for its simplicity and ease of use. Although the Loneliness Scale has been translated into multiple languages for cross-cultural research Igarashi [39]; Mauri et al. [40], to our knowledge there is no validated translation in the Croatian language.

The Three-Item Loneliness Scale is a brief and widely used self-report measure consisting of three items rated on a 3-point Likert scale Hughes et al. [36]. Individuals rate the frequency of their experiences regarding different aspects of their life, and the responses are coded as 1 ="hardly ever", 2 ="some of the time", and 3 ="often". The individual responses are then summed up with higher score indicating greater loneliness. It has been validated in various populations, demonstrating good reliability and validity and has been found to correlate strongly with longer versions of the UCLA Loneliness Scale, indicating its effectiveness in capturing feelings of loneliness. While the Cronbach's alpha reliability coefficient (0.72) for the Three-Item version is slightly below the typically reported alpha values for full scales Akerlind & Hornquist [41]; Cuffel & Akamatsu [42], the internal consistency remains strong, suggesting that the items effectively measure loneliness.

Procedures for translation the Three-Item Loneliness Scale

Permission for the translation and use of the instrument has been obtained from the author Hughes et al. [36] The authors of the original scale agreed that the Croatian-translated version of the scale can be freely used for research and educational purposes.

To ensure that assessment methods and testing concepts can be adapted or generalized across borders, the Three-item Loneliness Scale was translated according to the guidelines and recommendations for cross-cultural adaptation of questionnaires Epstein et al. [43] and ITC Guidelines for Translating and Adapting Tests International Test Commission [44]. This approach not only enhances the validity of the adapted scale but also ensures that it meets international standards, making it a useful tool for researchers and practitioners worldwide. Figure 1 outlines the detailed procedure for developing the translated version and its validation.

Fig. 1

Procedures for developing the translated version of the Three-Item Loneliness Scale and its validation

Forward translation

The Three-Item Loneliness Scale version underwent an initial evaluation by two bilingual content experts to ensure content equivalence in terms of relevance and sensitivity to the target culture. Subsequently, two proficient native Croatian researchers, well-versed in English and possessing academic and teaching experience, independently forward-translated the items into Croatian. The translation approach was conceptual, prioritizing the conveying of the intended meaning rather than a word-for-word rendition, utilizing the most contextually appropriate Croatian expressions. The Croatian language used in the translation avoided unfamiliar technical terms, or any language that could be deemed sensitive or offensive to the target population. Consistency between the two translated versions was cross-checked, with any variances addressed through consensus discussions between the translators.

Following the initial translation by bilingual researchers, the procedure continued with the review of translated versions by native Croatian speakers. A focus group Morgan [45] of six native Croatian speakers participated in telephone interview to review and provide feedback on the translated questionnaire. This group, familiar with the language and culture, offered valuable insights through in-depth interviews to assess the clarity and cultural appropriateness of the translated questionnaire. Their feedback allowed for further refinement of the translation to enhance linguistic accuracy and cultural suitability, ensuring it aligns well with the target population.

Backward translation

To ensure translation accuracy from Croatian to English, two proficient native Croatian professors/scientists conducted a backward translation, with the initial Croatian translation being blinded. The back-translation was subsequently compared with the original English items to identify any discrepancies and ensure conceptual consistency, prioritizing conceptual equivalence over linguistic accuracy. No inconsistencies were found.

Participants and procedures

Participants

The sample consisted of a total of 309 adults (aged 18 and above) residing in Croatia. Before starting the survey, participants acknowledged that their participation was voluntary and provided consent for their anonymous data to be used for analysis. Participants were informed about the study goals and were advised they could withdraw at any time if they wished. All procedures followed were conducted in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the revised Helsinki Declaration of 1975 (latest revision: 2000) World Medical Association [46]. Ethical approval for the study was obtained from the Ethics Committee of the University of Zagreb, Faculty of Croatian Studies, Croatia. Informed consent was obtained from all participants.

Measures

In addition to collecting demographic information, including age, gender, education etc., the survey included two instruments: one measuring loneliness and the other assessing social connectedness.

Three-Item loneliness Scale

The final translated version of the Three-item Loneliness Scale was administered through an online, anonymous survey conducted in Croatian. The survey was created via SurveyMonkey, and the link was distributed via online channels and platforms, including university email lists, social media platforms, and word-of-mouth.

The Social Network Index

The SNI is a 12-item measure designed to assess social network diversity and the number of high-contact social roles. Developed by Cohen [47], it evaluates social network quality, size, and diversity through key variables, including the number of people with whom a participant has regular contact, network diversity (the number of social roles they engage with), and embedded network (the number of network domains in which they are active). The SNI includes scales that measure participation in 12 different social roles, such as spouse, parent, friend, work colleagues, religious groups, and community involvement, as well as the frequency of interactions within these networks. The SNI scale was used to evaluate participants' level of social connectedness.

Data analysis

We calculated descriptive statistics (e.g., means, standard deviations) for each variable to understand the characteristics of the sample. Pearson correlations were computed to examine bivariate associations between the quantity and quality of social networks (network diversity, number of people in social network, and embedded network) and loneliness. Next, linear regression analyses were computed to examine the quality and quantity of social networks as predictors of the loneliness while controlling for demographic factors (age and gender). Additional regression analyses were conducted to explore potential differences across demographic groups (e.g., age differences in the relationship between social network and loneliness). Age was categorized according to Peng et al.'s [48] recommendation as follows: young adults (18–25), adults (26–44), middle-aged adults (45–59), and older adults (60 +). Effect sizes are noted in the tables following Cohen [49]. Bootstrap procedures with 2,000 samples were applied to the regression models.

Results

Following the pre-test phase of the Three-Item Loneliness Scale, the translation and adaptation process confirmed that the instrument is appropriate for use in the Croatian cultural context. To ensure that the Croatian version retained the same factor structure as the original scale Hughes et al. [36], Exploratory Factor Analysis (EFA) was conducted. The Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy was 0.683, and Bartlett’s test of sphericity was significant, indicating that the data were suitable for factor analysis. As expected, EFA revealed a single-factor structure, with strong factor loadings for all three items: Item 1 = 0.919, Item 2 = 0.775, and Item 3 = 0.661. These results support the validity of the Croatian version of the UCLA Three-Item Loneliness Scale as a unidimensional measure of loneliness. The alpha reliability coefficient of 0.82 for the Croatian version of Three-Item Loneliness Scale indicated good internal consistency. Supplementary material (Table S1) contains original English version of the items translated into Croatian. The final sample of 309 participants completed the study (mean age = 29.95, SD = 12.33, range: 18–75; 76% women). Descriptive statistics for all study variables are displayed in Tables 1 and 2.

Table 1 Descriptive statistics of the study sample

Table 2 Descriptive statistics of loneliness and social connectedness

Bivariate correlations indicated a significant association between social network variables and loneliness, as shown in Supplementary material (Table S2). The results of the bivariate correlation analysis, presented in Table S2, show that the relationships between loneliness and social connectedness ranges from low to medium effect sizes, with significant negative correlations between loneliness and network diversity (r = − 0.233), number of people in the social network (r = − 0.306), and embedded network (r = − 0.248). The correlations between different aspects of social connectedness were moderate to strong (r > 0.70).

In the regression analysis examining the quality and quantity of social networks as predictors, the three variables–network diversity, number of people in social network, and embedded social network, were entered simultaneously, alongside age and gender as control variables, to examine their unique contributions over-and-above the other predictor variables. The results indicated that having fewer people in one's social network, with whom they had regular contact was associated with higher levels of loneliness, beyond the effects of other social network indices, age and gender. Network diveristy and embedded social networks did not show significant associations with loneliness levels.

Age was significanlty and negatively associated with levels of loneliness levels (Table 3). When running conducting separate lieaner regressions to explore potential age differences in the relationship between social network and loneliness, the results indicated that specifically, young adults (18–25) with fewer social connections reported reported significantly higher levels of loneliness, while for older adults (60 +), having fewer active network domains (embedded network) was more strongly associated with loneliness compared to the other age groups (Table 4).

Table 3 Linear regressions examining prediction of loneliness by social connectedness controlling for age and gender

Table 4 Linear regressions examining prediction of loneliness by social connectedness across age group

Discussion

The primary objective of this study was to translate the Three-Item Loneliness Scale into Croatian while ensuring its validity and reliability. Our aim was to develop a culturally appropriate version of the scale suitable for use within the Croatian population. Additionally, we sought to explore the relationship between social connectedness and loneliness, while also examining the effects of demographic factors such as age and gender.

Following the adaptation process, a Croatian version of the Three-Item Loneliness Scale was successfully developed, maintaining semantic and conceptual equivalence with the original English version. Our findings from the pre-test phase confirmed the suitability of the adapted instrument for use within the Croatian cultural context.

Subsequent analyses revealed that having fewer individuals within one's social network with whom regular contact was maintained correlated with higher levels of loneliness, even after accounting for other social network indices, age, and gender. Interestingly, network diversity and embedded social networks did not show significant associations with levels of loneliness. Notably, younger people reported greater loneliness, while older adults with fewer active social connections also experienced increased loneliness. These findings suggest that in Croatia, the quantity of social connections (i.e., the number of people in one's social network) plays a more significant role in reducing loneliness than their diversity or embeddedness (i.e., the quality of social networks). This aligns with findings from a study on social connectedness and successful aging in Croatia, which demonstarted that higher levels of social connectedness were associated with a more positive self-perception of successful aging. That study, which included 824 older respondents, found that individuals with higher educational levels, who were married, living in their own homes, using modern technologies, and actively engaged in the community, reported greater social connectedness and a more positive perception of successful aging Plužarić et al. [50].

For younger individuals, maintaining a larger number of regular social contacts appears to be crucial in reducing loneliness, as this age group reported higher levels of loneliness. The significant negative association between age and loneliness highlights the greater vulnerability of younger individuals to feelings of isolation. This susceptibility can be attributed to various factors, including developmental challenges and changes in social dynamics. Adolescence and young adulthood are critical life stages characterized by identity formation and the navigation of complex social relationships de Moor et al. [51]. During these phases, individuals may encounter difficulties in establishing meaningful connections, especially when facing the pressures of forging new friendships, navigating romantic relationships, or transitioning into new environments such as college or the workforce. Additionally, while digital communication and social media offer unprecedented connectivity, they may paradoxically contribute to heightened loneliness by failing to provide genuine human interaction Di Vincenzo et al. [52]. The widespread use of technology for socializing can sometimes replace in-person connections, leading to a disconnect between online engagement and real-world emotional fulfillment. Moreover, significant life transitions, societal expectations, and technological influences, further exacerbate the risk of loneliness among younger to individuals Fardghassemi & Joffe [53]. Addressing these multifaceted challenges requires targeted interventions that foster authentic social connections, strengthen social support networks, and promote resilience in younger populations.

Conversely, the loneliness experienced by older adults with fewer active network domains compared to other age groups may steam from several factors. As individuals age, they often encounter a myriad of life changes that can impact their social connections and contribute to feelings of isolation. For instance, retirement, marks a major transition that may reduce daily social interactions, as individuals lose the workplace environment as a source of social engagement. Additionally, older adults are more likely to experience the loss of friends and family members, leading to a shrinking social circle and fewer opportunities for companionship. Age-related health issues and physical limitations can further restrict social participation, making it more difficult to engage in community activities and maintain meaningful relationships. The consequences of social isolation among older adults are profound, including risks of depression, cognitive decline, and even mortality Rokach [54]. Given these challenges, it is crucial to address the unique social needs of older adults through targeted interventions. This includes facilitating access to community resources, promoting intergenerational connections, and providing support for caregivers, which is essential in mitigating loneliness and improving overall well-being in later life. However, it is important to note that the older age group constituted only 3% of our sample, which prevents us from drawing firm conclusion on these findings. Future research should aim to include a more representative sample of older adults to better understand the specific factors contributing to loneliness in this population.

This study offers several strengths that enhance its validity and contribute to our understanding of loneliness and social connectedness within the Croatian population. First, this is the first study to adapt and translate the Three-Item Loneliness Scale into Croatian, ensuring the measurement tool's reliability and validity for assessing loneliness in this cultural context. Second, by analyzing both the quantity and quality of social connectedness, the study provides a nuanced understanding of the relationship between social networks and loneliness in a specific cultural setting, enabling cross-cultural comparisons. Third, by controlling for potential confounding variables such as age and gender, the study minimized the potential sources of bias and provided a more precise assessment of how social networks influence loneliness within the target population. However, certain limitations should be noted. The cross-sectional design of the study is a limitation as it is not possible to infer causality. In addition, the sample was skewed towards younger participants, limiting generalizability. Despite these limitations, the study highlights the importance of different aspects of social connectedness (social quantity, network diversity and embeddedness) in alleviating loneliness, particularly among younger and older age groups. While these patterns may be unique to the Croatian context, further research with more diverse and representative sample is necessary to confirm and expand upon these findings.

A key component of this study was the validation of the Three-Item Loneliness Scale in the Croatian context. We provided a detailed information of the pre-test phase and the steps taken to ensure the cultural appropriateness of the translated version. This process included multiple translation phases, forward and backward translation, and focus group feedback, ensuring that the scale retained both conceptual and linguistic equivalence. The validity of the Croatian version was confirmed through Exploratory Factor Analysis (EFA), revealing a strong single-factor structure and good internal consistency, supporting its reliability for future use. In comparison to other loneliness measures, the Three-Item UCLA Loneliness Scale offers a brief, reliable, and easy-to-administer screening tool Hughes et al. [36]; Igarashi [39]; Trucharte et al. [55], particularly in healthcare settings Sirois et al. [56]. Its efficiency allows for wider applicability beyond the general population, including in clinical contexts and intervention targeting loneliness. While the results of this study were not as strong as expected, with modest correlations and regression results, these findings should be interpreted with caution due to the study’s cross-sectional design and sample limitations. Nevertheless, despite these constrains, the scale’s ability to provide a quick and reliable measure of loneliness represents an important contribution to loneliness research and its practical applications.

Our study of loneliness and social connectedness within the Croatian population provides valuable insights with potential clinical implications. Through the adaptation and translation of the Three-Item Loneliness Scale into Croatian, we have established a reliable instrument for assessing loneliness in this specific cultural context. The brief and dependable nature of the scale will be an effectivel tool for healthcare professionals in identifying and addressing loneliness among patients, thereby aiding in the management of care. Moreover, by examining both the construct of loneliness and the role of social connectedness, we underscore the universal applicability of our findings. This comprehensive approach allows us to draw attention to the broader implications of our research, highlighting its potential relevance across diverse populations. The emphasis on both the quantity and quality of social connectedness offers a valuable insights for customizing interventions aimed at effectively addressing loneliness. Additional research is also necessary to investigate the clinical implications of these findings and their relevance in addressing loneliness within various cultural contexts.

Conclusion

This study successfully adapted the Three-Item Loneliness Scale into Croatian and examined the association of social network indices and loneliness. Our findings emphasized the importance of social networks in shaping feelings of loneliness, particularly among both young and old age groups. The study was conducted within the cultural context of Croatia, which exhibits a blend of individualistic and collectivistic traits. Future research should directly investigate how cultural and social factors influance the relationship between loneliness and social network variables. Cultural and future studies should directly examine cultural and social factors that may influence of the relationship between loneliness and social network variables.