Citizenship behaviors yield considerable benefits for both society and individuals. Specifically, such behaviors help individuals expand their social networks and gain social knowledge (Lewicka, 2005). They also promote public participation in democratic activities (Kennedy, 2006; Kennedy, 2018) and enhance overall societal well-being (Wojciszke, 2015). Citizenship behaviors are positively associated with satisfaction with quality of life (Zalewska and Nosek, 2020) and subjective well-being (Lewicka, 2005; Wojciszke, 2015). In terms of mental health, these behaviors play a crucial role in fostering community mental health by combating stigma, supporting the social inclusion and active participation of individuals with mental illnesses, and addressing social inequalities (e.g., racism) that contribute to or exacerbate mental illnesses (Paradies et al., 2015; Ponce and Rowe, 2018). In summary, citizenship behaviors extend beyond their role in promoting social progress by exerting positive effects on individuals’ mental health.

Traditionally, national identity and patriotism were regarded as the core components of citizenship behaviors. Based on their model, Zalewska and Krzywosz-Rynkiewicz (2011) developed Citizenship Behavior Questionnaire for children and youth. Zalewska and Nosek (2020) modified this tool into the Citizenship Behavior Questionnaire-30–General Version (CBQ-30-GV) for measuring seven types of citizenship behaviors within three forms of citizenship, including passive, semiactive and active, in people of different ages. Passive citizenship behaviors include expressions of national identity, such as a sense of belonging to the state and an acceptance of its historical values, and patriotism, which reflects national pride (Kennedy, 2018; Parker, 2009; Zalewska and Krzywosz-Rynkiewicz, 2011). By contrast, active citizenship behaviors involve personal activity (e.g., engaging in self-development, fostering independence, and cultivating creativity), political activity (e.g., joining a political party, representing a local government, and supporting a political party), activity for change (e.g., advocating for government accountability or systemic reform), and social activity (e.g., participating in charitable campaigns and voluntary initiatives for environmental or social causes; Zalewska and Krzywosz-Rynkiewicz, 2011). Semiactive citizenship behaviors include voting (e.g., engaging in community affairs and seeking information to make informed decisions) and loyalty (e.g., adhering to laws and regulations and demonstrating integrity in daily life; Zalewska and Krzywosz-Rynkiewicz, 2011).

Zalewska and Nosek (2020) proposed a third-order factor structure for the CBQ-30-GV, comprising a general factor, a passive–semiactive–active behavior factor, and seven specific factors. However, the adequacy of this third-order factor structure has not been empirically tested. Further, multiple studies have demonstrated a positive association between citizenship behaviors and subjective well-being (Lewicka, 2005; Wojciszke, 2015; Zalewska and Zawadzka, 2016). A study on Polish young adults reported that all dimensions of citizenship behaviors, as measured using the CBQ-30-GV, were positively correlated with subjective well-being (Zalewska and Nosek, 2020), whereas political activity and activity for change were not significantly associated with subjective well-being among Polish adolescents (Zalewska and Zawadzka, 2016). A systematic review including 53 published studies found varying associations between type of civic engagement and well-being in young adults; most of studies confirmed that civic engagement enhances well-being (Fenn et al., 2024).

As university students transition from adolescence to early adulthood, they encounter multiple pressures, including the demands of independent living and societal expectations, which may increase the risk of depression (Chou et al., 2018). However, studies examining the relationship of citizenship behaviors with depression and perceived stress have shown less consistent results. A review for 10 studies suggested that civic engagement programs improve mental health outcomes, including reduced anxiety, reduced sadness, and increased resilience in adolescents or young adults; these programs foster empowerment, sense of belonging, and social connections, while also enhancing interpersonal skills and career aspirations (Oubiña López and Gómez Baya, 2025). Longitudinal studies in Hong Kong and mainland China have found that civic engagement was sequentially associated with better psychological and social well-being, but not emotional well-being in young adults (Chan and Mak, 2020). A study in Canada did not find the prediction of civic engagement, composed of social engagement, social conscience, and civic action, for the optimal trajectory of mental health (Wiium et al., 2023). Citizenship behaviors demonstrate a heterogeneous relationship to depression and perceived stress. Future research should examine the associations of various dimensions of citizenship behaviors with depression and perceived stress in young adults.

Citizenship engagement is essential for improving social conditions for young adults (Swartz et al., 2009). Taiwan currently faces various societal challenges, including a potential military conflict with China (Lin and Yen, 2024; Yen and Hsiao, 2024a; Yen et al., 2024b; Yen et al., 2024c) and increasing housing prices (Chen et al., 2024). Therefore, increased civic engagement among young people is urgently required to contribute to social progress. However, further research is required to determine whether various dimensions of citizenship behaviors have distinct relationships with depression and perceived stress among university students in Taiwan.

This cross-sectional online survey study had two primary objectives. First, this study evaluated the psychometric properties of the Chinese version of the CBQ-30-GV among Taiwanese university students. In addition to assessing the fit of multiple factor structure models and internal consistency, the study investigated the concurrent validity of the CBQ-30-GV by analyzing the association between citizenship behaviors and well-being, as proposed by Zalewska and Nosek (2020). Second, the study explored the relationships between multidimensional citizenship behaviors, as measured by the CBQ-30-GV, and both depression and perceived stress. We hypothesized that all CBQ-30-GV factor scores would be negatively correlated with depression and perceived stress among university students.

Table 1 shows the fit index results of competing models. The results indicated that the one-general-factor model (M0) failed to converge and that both the second-order model (M2) and third-order model (M4) demonstrated poor fit indices (CFI = 0.889, TLI = 0.878, RMSEA = 0.102, and SRMR = 0.112 for M2; FI = 0.889, TLI = 0.878, RMSEA = 0.102, and SRMR = 0.112 for M4) and were statistically equivalent. In addition, the seven-specific-factor model (M1) exhibited a satisfactory model fit (CFI = 0.950, TLI = 0.943, RMSEA = 0.069, and SRMR = 0.075). However, the bifactor model (M3) exhibited the most favorable fit (CFI = 0.987, TLI = 0.984, RMSEA = 0.037, and SRMR = 0.048). Model comparison using the ECVI indicated that M3 had the lowest ECVI value (1.056), supporting its superiority over the three converged models, whose ECVI values ranged from 2.396 to 4.631 (Table 1). The M3 was then used in multigroup CFA to evaluate its measurement invariance. The findings showed that the configural model had acceptable CFI (0.903) and RMSEA (0.072) across gender; and acceptable CFI (0.901) and RMSEA (0.073) across grade. Moreover, both $\mathrm{\Delta }$CFI and $\mathrm{\Delta }$RMSEA were smaller than 0.01 in the comparison between the metric invariance model and the configural model ($\mathrm{\Delta }$CFI = 0.007 and $\mathrm{\Delta }$RMSEA = 0.000 for gender subgroup comparison; $\mathrm{\Delta }$CFI = 0.003 and $\mathrm{\Delta }$RMSEA = 0.001 for grade subgroup comparison); and between the scalar invariance model and the metric invariance model ($\mathrm{\Delta }$CFI = 0.002 and $\mathrm{\Delta }$RMSEA = 0.000 for gender subgroup comparison; $\mathrm{\Delta }$CFI = 0.001 and $\mathrm{\Delta }$RMSEA = 0.001 for grade subgroup comparison).

Table 2 presents item properties, including means, standard deviations, skewness, kurtosis, and factor loadings. Specifically, the distribution of item scores was relatively normal, with skewness values ranging between –2.032 and 0.901 and kurtosis values ranging between –1.250 and 3.976. Among the seven factors, items associated with the political activity factor had the lowest mean scores (1.781–1.894), whereas items associated with the national identity (3.243–3.609), semiactive (3.030–3.680), and personal activity (3.059–3.536) factors had the highest mean scores. In M1, all items demonstrated adequate factor loadings ($>$0.3) on their respective specific factors. Similarly, in M3, the majority of items loaded adequately onto their designated specific factors. However, some items exhibited relatively lower loadings in the bifactor model, particularly Item 1 from the national identity factor (0.148) and Item 8 (0.157) and Item 10 (0.166) from the semiactive factor.

In Table 3 descriptive statistics (Means and SD) are reported for all investigated variables—for all specific dimensions and general score of citizenship behaviors, as well as for wellbeing (PERMA-profiler), depression (CES-D) and stress (PSS-10). Table 4 and Fig. 1 show the results of examining concurrent validity and internal consistency. All seven specific factors in the CBQ-30-GV demonstrated acceptable internal consistency, with McDonald’s $\omega$ values ranging from 0.740 to 0.914. Additionally, the overall CBQ-30-GV exhibited excellent internal consistency ($\omega$ = 0.924). In terms of concurrent validity, both the specific and general CBQ-30-GV factor scores were significantly and positively associated with the PERMA-Profiler scores (r = 0.080 to 0.490; p 0.001).

Fig. 1.

Magnitudes of correlations between studied variables.

Table 5 presents the correlation of citizenship behaviors with depression and perceived stress. All CBQ-30-GV factor scores were significantly and negatively associated with depression (r = –0.426 to –0.135; p 0.001), except for political activity (r = 0.152; p 0.001), social activity (r = –0.009; p $>$ 0.10), and activity for change (r = 0.095; p 0.01). Additionally, all CBQ-30-GV factor scores were significantly and negatively associated with perceived stress (r = –0.346 to –0.165; p 0.001), except for political activity (r = 0.119; p 0.001), social activity (r = –0.087; p 0.01), and activity for change (r = 0.009; p $>$ 0.10).

In this study, we discovered that the bifactor structure of the CBQ-30-GV, which comprises seven specific factors and a general factor, exhibited the most favorable and most satisfactory model fit. By contrast, the second-order and third-order factor structures exhibited a poor model fit, thereby not supporting the concept of three higher-order factors (i.e., passive, semiactive, and active). Therefore, each CBQ-30-GV item was identified to contribute to both a specific behavioral dimension and a general citizenship behavior factor. Additionally, the CBQ-30-GV demonstrated adequate internal consistency, with all seven specific and general factor scores exhibiting significant and positive correlations with the PERMA-Profiler scores.

The results of a previous study supported full invariance across gender groups (female vs. male) but not across country groups (Poland vs. Taiwan) (Hung et al., 2025); partial invariance was achieved by relaxing certain item loadings, indicating that Polish and Taiwanese students may interpret some items differently (Hung et al., 2025). The present study further examined the psychometric properties, including the fit of multiple factor structure models, internal consistency and concurrent validity of the CBQ-30-GV among the same Taiwanese young adults in the study of Hung et al. (2025). The present study is the first one to examine multiple psychometric properties of the CBQ-30-GV used for assessing the specific and general dimensions of citizenship behaviors and their associations with depression and perceived stress among Asian people.

Our findings regarding the bifactor structure of the CBQ-30-GV suggest that citizenship behaviors are inherently complex. Specifically, each citizenship behavior can be categorized into distinct dimensions, such as feelings of belonging, national pride, personal activity, political engagement, activity for change, social relationships, and the continuum between passive and active participation (Kennedy, 2018; Parker, 2009; Zalewska and Krzywosz-Rynkiewicz, 2011). These distinct citizenship behaviors can be aggregated into a general citizenship behavior. Therefore, the CBQ-30-GV represents a valuable tool for policymakers and health-care providers, enabling them to assess both individuals’ specific dimensions of citizenship behavior and overall citizenship behaviors. On the other hand, some items (Items 1, 8, and 10) exhibited low loadings (0.2) on the specific factor but adequate loadings ($>$0.4) on the general factor. Item 1 asks participants if they feel a citizen of the Republic of China. Although the Republic of China is the official name, some young adults may identify themselves as Taiwanese; the loading of Item 1 for the national identity factor might be reduced. Item 8 asks participants if one should respect the state institutions (e.g., government, parliament, state offices). Item 10 asks participants if they try to act in accordance with the rules (e.g., traffic regulations, school regulations). Item 8 and Item 10 belong to the semiactive factor assessing behaviors related to loyalty and voting. The results indicated that university study in Taiwan may have low agreement with respecting the state institutions and acting in accordance with the rules in terms of representing loyalty.

Our findings revealed notable variations in participant engagement across different dimensions of citizenship behaviors. Specifically, national identity, semiactive, and personal activity factors received high scores, whereas political activity factors received low scores. National identity refers to an individual’s sense of belonging to a nation and of accepting its historical values (Zalewska and Nosek, 2020). The high scores for the national identity factor observed in this study indicated that the participants demonstrated high levels of national identification and national consciousness. The semiactive factor of the CBQ-30-GV assesses behaviors related to voting and loyalty. Zalewska and Nosek (2020) proposed that voting behavior reflects civic engagement, although it requires less active participation compared with political or social activities. Loyalty refers to everyday behaviors reflecting respect for state institutions, compliance with laws and regulations, and ethical daily conduct (Zalewska and Nosek, 2020). The high scores observed for this factor in this study suggested that the participants exhibited institutional trust and acceptance of social order.

The personal activity factor of the CBQ-30-GV assesses participants’ behaviors related to self-development and learning engagement (Zalewska and Nosek, 2020). The high scores observed in this study indicated that the participants demonstrated strong self-regulation and individualistic values. By contrast, political activity evaluates participants’ intentions of joining a political party, participating in governance, representing a local government, and supporting a political party (Zalewska and Nosek, 2020). The low scores observed for this factor in the present study suggested a lack of interest among the participants in political engagement. Their low political activity and relatively high level of semiactive citizenship may indicate that Taiwanese students—like adolescents in high-HDI (Human Development Index) countries—accept the existing social order and demonstrate a relatively high level of loyalty (respect for state institutions, law, and rules), and do not feel the need to engage in socio-political activities. However, given that youth disengagement from political activities can hinder a nation’s sociopolitical progress, determining the underlying reasons for Taiwanese university students’ low political participation is crucial.

Our findings indicated that both the general citizenship behavior factor and the majority of specific factor scores of the CBQ-30-GV were significantly and negatively associated with depression and perceived stress among university students. However, because of the potential bidirectional nature of the relationship between citizenship behaviors and mental health, causality could not be inferred. Engagement in citizenship behaviors may contribute to building social networks, acquiring societal knowledge, and fostering a sense of self-worth (Lewicka, 2005), which in turn may mitigate loneliness, depression, and perceived stress. Such behaviors may promote democracy and entrepreneurship (Kennedy, 2006; Kennedy, 2018) and enhance societal welfare (Wojciszke, 2015), all of which may enhance individuals’ well-being (Lewicka, 2005; Wojciszke, 2015). Conversely, individuals experiencing depression and high perceived stress may lack the energy or motivation to engage in citizenship behaviors. This complexity necessitates longitudinal studies to investigate the temporal relationships between citizenship behaviors and mental health.

The present study demonstrated that the associations of the political activity and activity for change factor scores with depression and perceived stress were significantly positive or nonsignificant among university students. Similarly, a study on Polish adolescents reported nonsignificant associations between citizenship behaviors within these two domains and subjective well-being (Zalewska and Zawadzka, 2016). These divergent findings may be attributable to the multidimensional nature of citizenship behaviors, as evidenced by our CFA results. Further, sociopolitical uncertainty in Taiwan may have the influences on the associations of the political activity and activity for change factor scores with depression and perceived stress. Taiwan and China are fraught with tension as Taiwan seeks to maintain its autonomy against China’s growing power (Bukhari et al., 2024). A potential collapse of the status quo between Taiwan and China is developing for several reasons such as the weakening power of the United States in the Asia-Pacific Region, the growing capacity of the People’s Liberation Army as well as on Taiwanese sociopolitical factors (Campagnola et al., 2024). Young adults may who engage in political activity and activity for change may face significant pressures and feelings of powerlessness. Despite these findings, the actual reasons underlying why these two factors exhibit different correlations with depression and perceived stress, compared with other factors in the CBQ-30-GV, warrant further investigation.

This study demonstrated that the CBQ-30-GV had a bifactor structure, comprising seven specific factors and a general factor, and exhibited adequate internal consistency and concurrent validity in a sample of university students. These findings support the CBQ-30-GV as a valid and reliable measure for evaluating citizenship behaviors, both specific and general, in this population. Our analysis revealed variations in factor scores across the seven dimensions of the CBQ-30-GV. The majority of CBQ-30-GV factor scores were negatively associated with depression and perceived stress, whereas the factor scores for political activity and activity for change exhibited either positive or nonsignificant associations with depression and perceived stress. Further research into the association of different citizenship behavior dimensions with depression and perceived stress is essential to understand the psychological well-being of young adults.

The datasets generated and analyzed during this study are available from the corresponding authors upon reasonable request.

AEC-C: conceptualization; CFH: writing the manuscript; KLL, BG and ML: conduct the study; AMZ: validation; BO-M: project administration; CYL: statistical analysis; CFY: data curation and Methodology. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study protocol was approved by the Institutional Review Board of Kaohsiung Medical University Hospital (KMUHIRB-E(I)-20240394). All participants provided informed consent to agree to participate in this study. The study was carried out in accordance with the guidelines of the Declaration of Helsinki.

Not applicable.

This study was supported by a grant from Kaohsiung Municipal Siaogang Hospital (Kmhk-113-09).

The authors declare no conflict of interest. Chung-Ying Lin is serving as one of the Editorial Board members of this journal. We declare that Chung-Ying Lin had no involvement in the peer review of this article and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to Vicente E. Caballo.

Supplementary material associated with this article can be found, in the online version, at https://doi.org/10.31083/BP46509.