Original Research

The roles of social support and sense of community in promoting psychological health at work

William K. Senayah, Florence Ruth A. Ntreh, Lebogang M. Ramalepe, Lawrence L. Mamabolo

Received: 19 Jan. 2026; Accepted: 16 July 2026; Published: 01 Oct. 2026

Copyright: © 2026. The Author(s). Licensee: AOSIS.
This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).

Abstract

Orientation: The garment industry remains labour-intensive and exposes workers to psychological risks that are often overlooked in social compliance systems.

Research purpose: The study examined how support from colleagues, supervisory support and sense of community shape psychological distress among garment workers in Ghana.

Motivation for the study: Prior research has largely focused on wages, physical safety and compliance, while the psychological health of workers in the garment industry has remained underexplored.

Research approach/design and method: A quantitative cross-sectional survey was conducted on 416 garment workers in Ghana’s Free Zone hub. Data were analysed using partial least squares structural equation modelling.

Main findings: Support from colleagues and supervisors strengthened a sense of community at work. Supervisory support and sense of community significantly reduced psychological distress, while the direct effect of colleague support on distress was not significant. Sense of community mediated the relationship between social support and psychological distress.

Practical/managerial implications: Factory managers should train supervisors in supportive communication and create workplace systems that promote belonging, peer interaction and community-building.

Contribution/value-add: Grounded in positive psychology, the study extends psychosocial health literature by showing that workplace relational resources, especially supervisory support and sense of community, are critical for reducing distress among garment workers.

Keywords: social support; psychological health; sense of community; psychological distress; positive psychology.

Introduction

The apparel industry is one of the most labour-intensive sectors globally, and its supply chains continue to raise concerns about decent work, worker safety and psychological well-being (Bae & Kuruvilla, 2025; Ribeiro, 2024; Wickramasingha & Smith, 2025). In Ghana, garment production with free-zone and export-oriented firms is important for employment and industrial development. Yet, repetitive tasks, production pressure, job insecurity and hierarchical supervision often characterise the work. These conditions make psychological health a relevant labour and management concern rather than only a general welfare issue. Although social compliance audits have become common in garment supply chains, they tend to focus more on visible, measurable labour standards such as wages, working hours and physical safety (Jones, 2024; Marslev & Whitfield, 2025; Seidu et al., 2024). Less attention is given to less visible psychological risks, including anxiety, emotional strain, psychological distress and workers’ sense of belonging (Engin & Parlak, 2024; Uddin et al., 2025). This gap is particularly important in Ghana, where broader evidence suggests increasing concern about mental health and workplace well-being (Awortwe et al., 2024; Elliason, 2022; Owusu et al., 2024; Rizzieri & Hannawa, 2024).

The study, therefore, examines whether support from colleagues and supervisors, together with workers’ sense of community, can serve as protective psychological resources for garment workers. Anchored on positive psychology theory, the study argues that supportive relationships and community bonds can build positive workplace resources that reduce psychological distress. Specifically, the study addresses three questions: (1) How does social support from colleagues (SSC) and supervisors influence psychological health among Ghanaian garment workers? (2) How does the sense of community influence workers’ psychological health? and (3) How does the sense of community mediate the relationship between social compliance in the emerging-economy garment-making supply chain?

The study also supports the broader goals of sustainable developmental goal (SDG) 3 on good health and well-being, SDG 8 on decent work and the African Union Agenda 2063 aspiration for shared prosperity and well-being. However, these frameworks are used as contextual justification, while the study’s central focus remains the Ghanaian garment workplace.

Literature review

Theoretical framework

Positive psychology emerged as a distinct field in the late 1990s, led by Martin Seligman and colleagues, as a response to psychology’s traditional focus on pathology and dysfunction (Seligman & Csikszentmihalyi, 2000). Historically, psychological research prioritised diagnosing and treating mental illness, often overlooking the conditions and resources that enable individuals and communities to thrive (Eds. Lopez & Snyder, 2009). Positive psychology seeks to address this gap by emphasising the scientific study of positive experiences, individual strengths and the environmental factors that contribute to human flourishing (Fredrickson, 2004; Seligman, 2011).

The core tenets of positive psychology include the exploration of positive emotions (e.g. hope, gratitude, joy), positive individual traits (e.g. resilience, optimism) and positive institutions (e.g. supportive workplaces and communities) that foster well-being and optimal functioning (Fredrickson, 2004; Peterson & Seligman, 2004). Positive psychology emphasises promoting strengths and enhancing well-being as buffers against psychological distress while fostering sustainable well-being even amidst adversity (Gable & Haidt, 2005). The framework was developed to expand the scope of psychology beyond deficit models by systematically investigating what makes life meaningful and fulfilling (Seligman & Csikszentmihalyi, 2000).

Positive psychology offers a robust lens for examining workplace well-being among apparel workers in Ghana by focusing on psychosocial resources that enhance resilience and well-being in challenging work environments. Specifically, this framework supports the exploration of social support and sense of community as protective factors that can mitigate psychological distress in precarious employment settings (Fredrickson, 2004). By prioritising enabling factors, the framework enables the study to explore how positive experiences and workplace resources can enhance well-being and adaptive functioning despite structural challenges.

Positive psychology has been applied across African contexts to explore workplace well-being and resilience. For instance, Rothmann (2013) used positive psychology to examine work engagement among South African employees, demonstrating that supportive work environments promote well-being and reduce burnout. Amponsah-Tawiah and Mensah (2016) applied positive psychology principles to occupational health in Ghana’s mining sector, linking supportive workplace practices to enhanced employee well-being. Cloete and Jeggels (2014) highlighted the role of supportive supervision in reducing distress among nurses in South Africa, aligning with positive psychology’s emphasis on positive institutions.

Recent African scholarship has further illustrated the utility of positive psychology in workplace and community settings. Konlan et al. (2022) found that organisational support in Kenya reduced burnout among health workers, while Agarwal et al. (2020) demonstrated that workplaces with a higher sense of community fostered informal support networks, which are critical for managing stress. Wissing et al. (2020) explored psychosocial well-being in South Africa, identifying how communal values and social connections foster resilience and life satisfaction. Mert et al. (2022) applied positive psychology frameworks to examine resilience among Turkish Fortune 500 companies, emphasising the importance of social support systems in buffering workplace stressors. Similarly, Theron et al. (2022) highlighted resilience processes among South African youth and communities, emphasising the role of social connectedness and support.

Further, Appiah et al. (2021) demonstrated that community support reduced psychological distress in Ghana, while Asiamah et al. (2023) found that social support and community ties reduced distress among informal workers. Elliason (2022) highlighted how social support in Ghanaian workplaces enhances well-being and mitigates stress, while Bonsu et al. (2025) found that supportive organisational cultures reduced psychological distress among Kenyan workers. A scoping review by Agyapong et al. (2023) observed that positive workplace practices reduced burnout among teachers, while Oti-Boadi et al. (2024) linked social support to psychological well-being among Nigerian university staff.

Applying positive psychology to this study allows for an examination of how social support and sense of community within workplaces act as psychological resources that foster well-being among apparel workers, even in environments marked by long hours, financial stress and job insecurity (Fredrickson, 2004). This framework also aids in interpreting findings related to psychological distress, framing distress not solely as an outcome but as a condition that can be mitigated through interventions and workplace practices that promote positive experiences (Fredrickson, 2004). For instance, the broaden-and-build theory within positive psychology posits that positive emotions expand individuals’ thought-action repertoires, enabling the building of enduring social and psychological resources that buffer against workplace adversity (Fredrickson, 2004).

This theoretical framework is relevant to the study, as it aligns with global and local goals of promoting decent work and well-being, and emphasises practical approaches to enhancing positive resources in workplaces to reduce psychological distress and improve well-being among apparel workers in Ghana. By applying positive psychology, the study contributes to evidence-based recommendations for creating supportive, connected work environments that enable workers to thrive amid challenging conditions.

Empirical review

The apparel industry in Ghana, like many emerging economies, is a labour-intensive sector fraught with human rights challenges, including precarious working conditions, excessive workloads and systemic power imbalances (Siaw et al., 2025). Despite the adoption of social compliance audits aligned with international frameworks like the United Nations Guiding Principles on Business and Human Rights (UNDPs, 2011), these mechanisms often fail to address the psychological health of workers, a critical yet underexplored dimension of well-being (Ma, 2023). This review argues that social support from colleagues and supervisors, coupled with a strong sense of community, is an indispensable protective factor for fostering psychological resilience and well-being among Ghanaian apparel workers. Grounded in positive psychology, these factors mitigate the adverse effects of high-pressure work environments and offer a pathway to sustainable, ethical labour practices. The industry’s persistent neglect of psychosocial health, driven by superficial compliance measures, underscores the urgent need for empirical research and practical interventions to prioritise workers’ mental well-being.

Social support from colleagues: A cornerstone of psychological resilience

Social support, defined as the emotional, instrumental and informational resources provided by workplace relationships (Cohen & Wills, 1985), is a well-established buffer against psychological distress. In the Ghanaian apparel industry, workers face repetitive tasks, stringent production quotas and financial insecurity, all of which elevate risks of anxiety, burnout and emotional exhaustion (Seidu et al., 2024). Social support from colleagues offers a critical counterbalance, fostering emotional resilience and job satisfaction. Studies in African contexts demonstrate this effect, as Konlan et al. (2022) found that peer support among Ghanaian health workers reduced burnout, while Obeng et al. (2025) showed that informal peer networks in Ghanaian factories enabled workers to share coping strategies, significantly lowering stress levels. These findings highlight that collegial support creates a sense of solidarity, enabling workers to navigate the psychological toll of demanding work environments.

Critics may argue that SSC is limited in its impact, particularly in hierarchical or competitive workplaces where interpersonal conflicts can arise (Viswesvaran et al., 1999). However, this overlooks Ghana’s cultural context, where communal values foster natural alliances among workers. Obeng et al. (2025) argue that even in high-pressure settings, peer support fosters micro-environments of trust, reducing feelings of isolation and enhancing emotional well-being. Moreover, positive psychology’s broaden-and-build theory suggests that positive interactions among colleagues, such as sharing encouragement or practical assistance, expand workers’ emotional resources, enabling them to better cope with stressors (Fredrickson, 2004). By prioritising collegial support, apparel firms can cultivate a workplace culture that mitigates psychological distress and aligns with global decent-work goals.

Social support from supervisors: A catalyst for worker well-being

Supervisory support, encompassing guidance, recognition and emotional encouragement from managers, is equally critical for psychological health in high-risk industries like apparel manufacturing. Supervisors act as gatekeepers of workplace resources, shaping workers’ perceptions of fairness and support (Darling & Cunningham, 2022). In Ghana, where power imbalances between workers and management are pronounced (Seidu et al., 2024), supportive supervision can mitigate feelings of marginalisation and enhance job satisfaction. For instance, Cloete and Jeggels (2014) found that supportive supervision reduced distress among South African nurses, a finding echoed in Ghana’s mining sector by Amponsah-Tawiah and Mensah (2016). These studies underscore that supervisory support is a pivotal resource for fostering psychological resilience.

Opponents might contend that supervisory support is inconsistent in fast-paced apparel factories, where managers prioritise production over employee well-being (Yucel et al., 2025). While production pressures are real, this critique ignores the potential for supervisors to integrate supportive practices within existing workflows. Gudeta et al. (2025) argue that brief, intentional supervisory interventions, such as regular check-ins or recognition programmes, can significantly enhance workers’ psychological health without disrupting productivity. Positive psychology emphasises that such interventions, by fostering positive emotions, build enduring psychological resources (Fredrickson, 2001). Thus, supervisory support is not only feasible, but also essential for creating a psychologically healthy workplace.

Sense of community: Fostering belonging and collective resilience

A sense of community, characterised by feelings of belonging, mutual commitment and emotional connection (McMillan & Chavis, 1986), is a powerful determinant of psychological health, particularly in Ghana’s collectivist cultural context. In the apparel industry, where workers often face isolation and job insecurity, a strong sense of community fosters trust, inclusion and collective coping mechanisms. Evidence from African workplaces supports this. Agyei et al. (2025) found that communal values in Ghanaian workplaces reduced psychological distress, while Maria Najjuka et al. (2020) demonstrated similar effects in Uganda. Seidu et al. (2024) specifically highlight that in Ghanaian textile factories, a sense of community – cultivated through shared goals and informal support networks – correlates with lower anxiety and higher life satisfaction among workers. These findings affirm that community is a vital resource for psychological resilience.

Sceptics may argue that fostering a sense of community is challenging in apparel factories due to high turnover, transient workforces and competitive pressures (Ashill & Sinha, 2004). However, this perspective underestimates the adaptability of communal practices in African contexts. Gyan and Kwakye (2025) show that informal communal activities, such as group lunches or mutual aid systems, naturally emerge among Ghanaian workers, strengthening emotional bonds and reducing stress even in transient settings. Positive psychology further supports this, suggesting that communal environments amplify positive emotions, enabling workers to build resilience against adversity (Fredrickson, 2004). By intentionally fostering community, apparel firms can create workplaces that prioritise psychological well-being

The synergistic interaction of social support and sense of community

The interplay between social support and sense of community is a critical yet underexplored mechanism for enhancing psychological health. Positive psychology posits that supportive relationships thrive within communal environments, while a strong sense of community amplifies the availability and impact of social support (Fredrickson, 2001). Musanje et al. (2024) found that social support embedded in strong community contexts significantly reduced psychological distress among informal workers. Similarly, communal environments enhance the efficacy of both collegial and supervisory support, as workers perceive support as more authentic within a cohesive community (Appiah et al., 2021; Darling & Cunningham, 2022). This synergy creates a reinforcing cycle in which the community strengthens support networks, which in turn deepen communal bonds, fostering resilience.

Critics might argue that this synergistic effect is less pronounced in industries with fragmented workforces, where transient employment disrupts community cohesion (Winson & Leach, 2002). However, positive psychology’s broaden-and-build theory counters this by suggesting that even short-term positive interactions can build lasting psychological resources (Fredrickson, 2001). Bellingan et al. (2020) found that brief communal interventions, such as team-building workshops, significantly improved workers’ perceptions of support and reduced stress, even in high-turnover settings. This evidence underscores that the interaction between social support and sense of community is a powerful mechanism for enhancing psychological health, regardless of workforce dynamics.

Despite strong evidence that social support and a sense of community protect mental health, Ghana’s apparel industry continues to overlook these psychosocial resources in favour of narrowly focused compliance audits that prioritise physical conditions and economic outputs (Welmon, 1998). This neglect is concerning, given the high rates of anxiety and depression among the general population in Ghana (Amu et al., 2021). Studies show that integrating psychosocial support – such as supervisor training, community-building and peer mentoring – can reduce stress, improve morale and enhance productivity (Cloete & Jeggels, 2014; Darling & Cunningham, 2022). Furthermore, embedding such practices within ethical business models improves worker retention and brand reputation, offering long-term strategic benefits (Węziak-Białowolska et al., 2019).

Research design

Study design and settings

This study was anchored in a quantitative cross-sectional survey, which influenced the sampling, data collection and interpretation. Using a cross-sectional survey, the study collected numerical data to develop a framework and interpret the study variables. This study used a quantitative cross-sectional survey because the objective was to measure the association among social support, sense of community and psychological distress among a relatively large group of workers at a single point in time. The design was appropriate for testing the proposed mediation model using partial least squares structural equation modelling (PL-SEM). However, because the data were collected at one point in time, the findings do not necessarily establish temporal or causal effects; therefore, the relationships are interpreted as statistical associations. The study was conducted among free-zone companies that are the mainstay of Ghana’s apparel exports. These companies, designated as Tier 1 garment firms by the Ministry of Trade, Agribusiness and Industry, employ thousands of workers, mostly female, to produce clothes such as uniforms, women’s wear, pet products and accessories. Together, these companies make up the largest apparel exporters from West Africa, with export destinations in the United States, Italy, Germany and the United Kingdom.

Sample and sampling technique

The study relied on attendance records to make sample size and sampling decisions. Given that the population was considered finite, Krejcie and Morgan’s (1970) sample size determination table was used to determine the study sample size. Thus, considering that attendance records showed that, on average, all the Tier-1 companies had approximately 9000 workers, Krejcie and Morgan’s (1970) sample size determination recommends a sample of 384. Thus, to account for potential errors, incomplete responses and other unforeseen circumstances, 450 questionnaires were distributed, yielding a response rate of 92.44%. To assess non-response bias, the characteristics of early and late respondents should be compared using available demographic information such as sex, job category, age and shift status. Missing data were screened before analysis. Cases with substantial missing responses on a scale were excluded from the relevant analysis, while the extent of missingness was reported in the result tables. This explains why the analysis of psychological distress used 374 valid cases after 42 incomplete responses to the psychological distress items were excluded. The simple random sampling technique was applied to select respondents. The sampling frame was developed from attendance records of eligible workers in the Tier-1 garment firms. Each eligible worker was assigned a unique identification number, and respondents were selected at random until the target sample was reached. Where selected workers declined or were unavailable, replacements were made using the next randomly selected number from the sampling list. These respondents were contacted via WhatsApp to obtain their consent to participate in the study after being briefed on the study’s aims and objectives. WhatsApp was used because workers were distributed across production units and shifts, and the platform provided an accessible channel for sharing the Google Forms questionnaire after consent was obtained. To reduce selection bias, invitations were sent only to workers identified in the sampling frame, reminders were sent uniformly, and participation remained voluntary. The authors acknowledge that workers without reliable smartphones or internet access may have been less likely to participate, and they note this limitation.

Sample

The sample comprised 416 respondents (208 [50.0%] males and 203 [48.8%] females; five [1.2%] did not disclose their sex) across the production line of a garment manufacturing company at an industrial hub in Accra, Ghana. Most of the respondents were machinists (78.8%), with the remaining being supervisors, production managers, quality controllers and accountants. Also, most did not run shifts (67.8%), while 30.0% did; 9 (2.2%) respondents did not respond to their shift status. The respondents ranged in age from 18 to 49 years, with an average age of 24.69 (standard deviation [s.d.] = 5.12) years, and an average of 3.27 (s.d. = 2.28) years of working in garment manufacturing.

Measures

The three main variables were measured in the study: social support (from colleagues and supervisors), sense of community at work (SCW) and psychological distress. The measures for social support and sense of social community at work were adapted from the second version of the Copenhagen Psychosocial Questionnaire (COPSOQ II: [Pejtersen et al., 2010]), while psychological distress was measured with an adapted version of the Kessler Psychological Distress Scale (Kessler 10: [Kessler et al., 2003]). The COPSOQ II items were adapted for the Ghanaian apparel workplace without altering the underlying construct. For sense of community, items were rephrased from questions to statements to match the response format and improve clarity. For the Kessler-10 scale, one item was removed because it loaded negatively against the item structure. To minimise validity threats from the adaptations, the revised measures were reassessed using Cronbach’s alpha, composite reliability (CR), average variance extracted (AVE) and Heterotrait-Monotrait (HTMT) ratios. The results showed acceptable CR, convergent validity and discriminant validity, supporting the use of adapted scales in the context.

Table 1 presents the reliability and validity evidence for the scales used in this study. We present evidence of construct reliability using Cronbach’s alpha and CR, convergent validity with the AVE, and discriminant validity using the HTMT ratio of the latent variables. Heterotrait-Monotrait ratios between any pair of latent variables should be less than 0.90 to establish sufficient discriminant validity (Henseler et al., 2015). The results showed that all the HTMT ratios were below the 0.90 criterion. The evidence indicated that the measures demonstrated adequate construct reliability and validity.

Social support from colleagues

Social support from colleagues was measured with the three-item scale from the COPSOQ II. The items require respondents to indicate how often they experience the indicated situations at work, and responses are rated on a 5-point Likert-type scale ranging from 1 (not at all) to 5 (all the time). A sample item is, how often do you get help and support from your colleagues at work? Total scores on the scale are obtained by summing the ratings across the three items. The scores ranged from 3 to 15, with lower scores denoting receiving less support from work colleagues. Pejtersen et al. (2010) reported a reliability coefficient of 0.70, and we obtained Cronbach’s alpha of 0.70 and a CR coefficient of 0.83. We also obtained convergent validity evidence of 0.62 using the AVE.

Social support from supervisors

We measured social support from supervisors (SSS) with the three-item scale from the COPSOQ II, which required respondents to indicate the frequency with which they receive support from their immediate supervisors at work. The responses are rated on a five-point Likert-type scale, ranging from 1 (not at all) to 5 (all the time). The scale composite scores range between 3 and 15, with higher scores indicating greater support from supervisors. A sample item is: ‘How often do you get help and support from your nearest supervisors?’ Pejtersen et al. (2010) report a reliability coefficient of 0.79. We obtained a Cronbach’s alpha of 0.69, a CR coefficient of 0.82 and an AVE of 0.61.

Sense of community at work

We measured the perceived SCW with a three-item scale from the COPSOQ II. We slightly modified the original sentence structure from question-type items to statements for respondents to indicate the frequency with which they perceive such an atmosphere at their workplace. For instance, an original item, ‘Do you feel part of a community at your place of work?’ was rephrased to read, ‘I feel part of a community at my place of work’. Responses are rated on a five-point Likert-type scale, from 1 (not at all) to 5 (all the time), with total scores ranging from 3 to 15. Higher scores indicate a stronger perception of social community at the workplace. Pejtersen et al. (2010) reported a reliability coefficient of 0.85, and we obtained Cronbach’s alpha of 0.64, a CR of 0.81 and an AVE of 0.58.

Psychological distress

We adapted the ten-item Kessler Psychological Distress Scale in this study. One item of the original scale (item 8) loaded negatively on the item total and was omitted, leaving nine items for the current study. The items require respondents to indicate how often they felt the situations or conditions presented in the statements in the past 4 weeks. The responses are rated on a five-point Likert-type scale, from 1 (none of the time) to 5 (all of the time). Total scale scores ranged between 9 and 45. Low scores indicate low levels of psychological distress.

We modified the Victorian Department of Health (2011) cut-off scores used as a guide for screening for psychological distress, given that we deleted one item that did not load adequately. For the original ten-item scale, Kessler-10 scores of 10–19 indicate that the individual is likely to be well (not distressed), 20–24 (likely to have mild distress), 25–29 (likely to have moderate distress) and 30–50 (likely to have severe distress). For our nine-item scale, scores of 9–17 indicate that the individual is likely to be well (not distressed), 18–22 (likely to have a mild distress), 23–26 (likely to have a moderate distress) and 27–45 (likely to have a severe distress). We obtained a Cronbach’s alpha coefficient of 0.88, CR of 0.90 and an AVE of 0.51.

Analysis

Before administering the questionnaire, which had been converted into a pilot study, 50 interns in a factory setting were assessed for item clarity, questionnaire flow, completion time and ease of responding via Google Forms. Feedback from the pilot indicated that the substantive items were understandable; therefore, no construction items were substantially changed. Minor revisions were made to the instruction sections to improve clarity. Pilot participants were not included in the final sample. Following the successful pilot, the questionnaire was uploaded to respondents’ WhatsApp pages for them to respond. The workers were given 2 weeks to respond and complete their responses. The questionnaire included consent to participate in the study and to publish the findings (in writing). Following successful data collection, responses were securely stored on personal computers, protected by passwords that ensured anonymity and confidentiality.

The main analytical procedure was PLS-SEM using SmartPLS (Jordan & Troth, 2020). The SPSS software was used to run descriptive statistics (means, s.d., skewness and kurtosis), tests of normality and tests of multicollinearity assumptions. The PLS-SEM was suitable because the model involved latent variables, mediation analysis and prediction-oriented relationships. The sample size of 416 exceeded the minimum requirement commonly applied for PLS-SEM and was adequate for the number of constructs and paths in the model. Before structural model testing, the measurement model was assessed through Cronbach’s alpha, CR, AVE and HTMT ratios. Multicollinearity and common method bias were also assessed using VIF values, all of which were below the 3.30 threshold. Bootstrapping was used to estimate the significance of direct and indirect effects.

Ethical considerations

Ethical clearance to conduct this study was obtained from the Accra Technical University and Directorate of Research, Innovation, Publication and Technology Transfer (Ref. No. RE 111 – 2025-DRIPTT). Hence, all procedures performed in studies involving human participants were in accordance with the ethical standards of DRIPTT, the national research committee and the 1964 Helsinki Declaration and its later amendments or a comparable ethical standard. Written informed consent was obtained from all individual participants involved in the study. Participation was voluntary; informed consent was obtained before questionnaire completion, and participants were assured of anonymity and confidentiality. No identification information was reported in the manuscript. The Google Forms data were downloaded and stored on a password-protected computer accessible only to the research team, and the results were reported in aggregate form.

Results

The study sought to ascertain if social community at work would significantly mediate the relationship between social support from colleagues and supervisors and psychological distress of workers at garment production companies at the central industrial hub of Ghana, Accra.

Assessment of the possible presence of common method bias

We were mindful of the potential effects of common method bias or common method variance in cross-sectional survey research. We collected the study data at a single point in time for all measures. Some researchers suggest that this may indicate the presence of common method bias (Jordan & Troth, 2020), which could affect the coefficients in an analysis and, hence, the findings. We therefore took steps at the pre-data and data collection stages to avoid the potential for common method bias.

Firstly, we used a temporal separation strategy, dividing the questionnaire into three sections and providing concise, clear instructions for respondents. We also included two items intended to minimise response set bias. Furthermore, the items for the construct were few, making it convenient and less time-consuming for respondents. Responses from respondents who did not respond appropriately to those items were excluded from the analysis.

Secondly, notwithstanding these pre-data-collection measures, we also conducted a post-data-collection assessment (Memon et al., 2023) to assess the possible presence of common method bias. Methodologists such as Hair et al. (2019), Kock (2015) and Kock and Lynn (2012) recommended the use of collinearity diagnostics, whereby a variance inflation factor (VIF) value less than 3.3 indicates the absence of common method bias. All the VIFs were below the 3.30 criterion, indicating no possible presence of common method bias. We also assessed the potential for common method bias using Harman’s single-factor test. We performed an unrotated principal component factor analysis containing all 18 measurement items. The results presented in Table 2 indicated that the first factor accounted for only 30.98% of the total variance, which is below the recommended 50% threshold. Therefore, common method variance was not a concern in our dataset.

Preliminary analysis

We evaluated the distribution of the Psychological Distress (dependent) variable using visual inspection of a histogram (Figure 1) and a Normal Q-Q plot (Figure 2). The histogram showed a positively skewed distribution with a prominent peak at the lower end of the scale, suggesting an expected floor effect, with most participants reporting lower levels of distress. While the visual distribution deviated from a perfectly symmetric bell curve, the Normal Q-Q plot showed that the mid-range data points still fit the diagonal reference line reasonably well. Given the relatively large sample size (N = 374), the Central Limit Theorem applies, rendering the data sufficiently robust for parametric statistical analysis and meaningful results. Also, the skewness and kurtosis statistics in Table 3 were below 1.0, suggesting an acceptable distribution of the scores.

FIGURE 1: Histogram with normality plot of psychological distress (dependent variable).

FIGURE 2: Normal Q-Q plot of psychological distress (dependent variable).

Table 3 presents the means, SD, skewness and kurtosis statistics of the constructs. Given that the scales for social support from colleagues and supervisors, as well as the sense of community, each have 3 items and are scored on a 5-point scale, the scores ranged from 3 to 15. A cut-off score of 9.0 (midpoint between 13 and 15) was used to determine high and low levels on these variables. A score above 9.0, therefore, indicated a high level on the measure. The results indicated that mean scores on all three variables were above 9.0, indicating generally high levels of social support from colleagues and supervisors and a perceived SCW among the respondents.

We further analysed the frequencies and percentages of the respondents reporting high and low levels on the constructs. The analysis of levels (Table 4) showed that six respondents did not respond to some items across the three constructs. Approximately 51% and 65% of the respondents reported high levels of support from colleagues and supervisors, respectively, with 76% reporting a high level of perceived SCW.

For the psychological distress measure, we used a nine-item scale, and the scores were categorised into levels of distress. As indicated earlier, mean scores between 9–17 indicate that the individual is likely to be well (not distressed), 18–22 (likely to have mild distress), 23–26 (likely to have moderate distress) and 27–45 (likely to have severe distress). This is consistent with the Victorian Department of Health’s (2011) guide. The overall mean of 19.72 suggested that workers were generally mildly distressed.

The results in Table 5 indicated that 42 respondents did not respond to some items on the psychological distress scale, leaving 374 for analysis. This number is still large enough for the analysis, with any adverse effect on the results and generalisability. The results showed that approximately 41% of the respondents were likely to be well, while 21% were likely to be severely distressed. Thus, about 59% of the respondents were likely to be mildly, moderately or severely distressed.

Structural path model evaluation

The structural model of PLS-SEM helps evaluate and test the relationships among variables. Figure 3 presents the results of the direct relationships and their corresponding significance values in parentheses.

FIGURE 3: Structural analytical model showing direct relationships with their corresponding level of significance.

The results presented in Table 6 showed that support from supervisors and colleagues jointly accounted for 36.3% of the variance in SCW. In contrast, support from supervisors and colleagues, together with SCW, explained 14.7% of the variance in psychological distress. The results indicated that SSC (β = 0.259, t = 5.423, p < 0.001, 95% confidence interval (CI) [0.170, 0.355]) and from supervisors (β = 0.452, t = 9.846, p < 0.001, 95% CI [0.360, 0.542]) significantly and positively predicted SCW, with support from supervisors being a stronger predictor with a moderate effect size.

Also, while SSS had a significant negative relationship with psychological distress (β = −0.224, t = 3.577, p < 0.001, 95% CI [−0.349, −0.105]), support from colleagues did not significantly predict psychological distress (β = 0.115, t = 1.862, p = 0.063, −0.004, 0.239). Meanwhile, SCW had a significant negative relationship with psychological distress (β = −0.256, t = 4.059, p < 0.001, 95% CI [−0.382, −0.135]).

The results presented in Table 7 further showed that SCW partially mediated the relationship between both SSC and psychological distress (β = −0.066, t = 3.238, p = 0.001, 95% CI [−0.113, −0.032]) and between support from supervisors and psychological distress (β = −00.116, t = 3.719, p < 0.001, 95% CI [−0.183, −0.060]). The mediation role in the relationship between colleagues’ support and psychological distress was competitive, rendering the total effect of colleagues’ support on psychological distress non-significant. However, support from supervisors showed a complementary partial mediation, making the total negative effect of support from supervisors on psychological distress significant and stronger. The mediation role is said to be competitive when the direct and indirect relationships are both significant, but in opposite directions (e.g. while the direct effect is positive, the indirect effect is negative). On the other hand, when both direct and indirect effects are in the same direction (e.g. both negative or positive), it is said to be complementary.

Discussion

The findings show that both colleague and supervisory support are important workplace resources, but they operate through different pathways. Colleagues’ support for a strengthened worker’s sense of community indicates that horizontal relationships build belonging and shared identity in Ghana’s garment workplace. However, colleague support did not directly reduce psychological distress. One possible explanation is that coworkers can provide empathy, companionship and practical help. Still, they may have limited authority to change structural stressors such as production pressure, workload, pay insecurity or supervisory demands. Thus, peer support may reduce distress indirectly by strengthening the community rather than by directly removing its sources.

By contrast, supervisory support had both a direct negative effect on psychological distress and an indirect effect through sense of community. This suggests that supervisors occupy a more powerful position in the workplace because they influence task allocation, communication, recognition and workers’ perceptions of fairness. In hierarchical work settings such as garment factories, supportive supervision may therefore reduce uncertainty and emotional strain more immediately than peer support.

The mediating role of sense of community indicates that social support becomes psychologically protective when it is translated into belonging, inclusion and shared workplace identity. This supports positive psychology, particularly the view that positive relationships and supportive institutions build resources that enable workers to cope with adversity (Seligman & Csikszentmihalyi, 2000). It also aligns with the Job Demand-Resource (JD-R) logic, which posits that social resources can buffer psychological strain (Jolly et al., 2021). In collectivist contexts such as Ghana, workplace community may be especially important because workers often interpret support through relational and group-based ties.

The Ghanaian findings are consistent with evidence from other collectivist and African settings, including studies from South Africa, Kenya, Uganda and Nigeria, where belonging, social support and supportive supervision have been linked to reduced distress and improved well-being (Reddy et al., 2022; Sharma & Srivastava, 2022). However, the stronger role of supervisors in this study may reflect the hierarchical organisation of garment factories, where managing behaviour has direct implications for workers’ emotional security. An alternative interpretation should be considered. Because the study is cross-sectional, it is possible that workers with lower distress perceived their supervisors and workplace community more positively. In addition, unmeasured factors such as wages, overtime, job insecurity and family responsibilities may partly explain psychological distress. These limitations should be acknowledged when interpreting the results.

Implications for practice

The practical implications are threefold: firstly, garment firms should train supervisors in supportive communication, fair task allocation and early identification of psychological strain. Secondly, firms should create structured opportunities for peer interaction, mentoring and worker associations that build a stronger sense of community. Thirdly, social compliance systems should be expanded beyond physical safety and labour standards to include psychological risk assessment, referral pathways and workplace mental health support.

Conclusion

Given that social compliance measures currently undertaken in the apparel industry do not fully address the full scope of psychological factors affecting workers, and that research on the psychological status of workers is underexplored, this study sought to examine the psychological health of Ghanaian apparel workers based on support from colleagues and supervisors, and a sense of community. Consequently, five conclusions are drawn from the study, based on the objectives, to achieve the aim of the study. Firstly, SSC positively predicted a SCW. Supervisor support also positively predicted the workers’ sense of community. Supervisors’ support significantly and negatively predicted workers’ psychological distress. Next, a sense of community significantly and negatively predicted workers’ psychological health. Finally, it played a significant mediating role through which social support, especially support from supervisors, influenced workers’ psychological health.

Admittedly, previous studies had examined the nature of work in the apparel industry and determined that workers were exposed to conditions that were susceptible to poor psychological health (Kabir et al., 2019, 2023; Seidu et al., 2024; Węziak-Białowolska et al., 2019) However, the key departure of the current study from those studies is that those studies focused on how disease, disability and most cases ergonomics factors impact psychological health, while the current study connects support from colleagues, supervisors and sense of community to determine not only their influence on psychological health, but also from the perspective of positive psychology to contribute to literature specific to both the apparel industry literature and management of Ghanaian apparel workers. This study thus contributes to the literature by establishing that colleagues, supervisors and the workplace sense of community are essential to workers’ positive psychology, which is critical for addressing stress, anxiety and depression. This study used a cross-sectional design; therefore, causal or temporal conclusions cannot be drawn. The data were self-reported, which may introduce common-method bias, although procedural and statistical checks were used to reduce the risk. The study was also limited to garment workers in Ghana’s garment industry, so generalisations to other sectors or countries should be made cautiously. Finally, other factors, such as wages, overtime, job insecurity and family demands, were not included in the model and may also influence psychological distress.

Acknowledgements

The authors, William K. Senayah; Florence Ruth A. Ntreh; Lebogang M. Ramalepe and Lawrence L. Mamabolo, declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.

William K. Senayah: Conceptualisation, Data curation, Methodology, Investigation, Writing – original draft and Writing – review & editing. Florence R.A. Ntreh: Data curation, Formal analysis, Methodology, Writing – original draft and Writing – review & editing. Lebogang M. Ramalepe: Data curation, Project administration, Writing – original draft and Writing – review & editing. Lawrence K. Mamabolo: Investigation, Project administration, Writing – original draft and Writing – review & editing. All authors reviewed the article, contributed to the discussion of results, approved the final version for submission and publication and take responsibility for the integrity of the findings.

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

The data that support the findings of this study are available on request from the corresponding author, William K. Senayah, upon reasonable request. The data are not publicly available due to the privacy of research participants.

The views and opinions expressed in this article are those of the authors and are the products of professional research. They do not necessarily reflect the official policy of any affiliated institution, funder, agency or that of the publisher. The authors are responsible for the article’s findings and content.

References