3.1. Drivers of sustainable practices

Sustainability practice within small enterprises in the garment and textile industry in Ghana is not just a matter of aspiration but also founded on the ability of enterprises to move across systemic constraints. The motivators of sustainable practices in this case happen to be a reaction to a group of interrelated obstacles that are either institutional, material or financial. The invention of such innovations as the use of vegetal dyes or the use of textile waste is typically created in times of shortage [17]. Alternatives, including the community-based interventions, such as cooperatives and resource-sharing programs, were also regarded as a possible way out of the current constraints in the economy and the infrastructure [13–18]. Furthermore, the information about cultural issues, historical approaches, and informal learning strategies tends to play a central role in sustaining the environmentally responsible local and contextually embedded behaviours [16]. Thus, the facilitators of the sustainable practices are not unilateral antecedents but are relationally constructed through the dynamic-boundary negotiation of the constraints that foreshadow hybrid models of sustainability based on the grassroots innovation and socio-economic resiliency.

3.2. Challenges to green sustainability

The concept of sustainability being implemented within the garment and textile sector, which, in this particular case, involves small-sized manufacturing units in developing economies, faces a mountainous load at the systemic level. These include bad regulatory governance, bad coordination of the institutions, bad waste management and lack of proper sustainability policy on the industry [19,20]. The practice with small-scale producers in Ghana is usually informal and does not feature in national agendas, resulting in peripheral green practices. Local artisans’ networks, local cooperatives and NGOs are developing locally owned recycling systems with new forms of circular production [21]. This has consequences of proving that despite the imposition of constraints by system-level issues, it can also encourage local sustainability changes. They describe such issues that are shaping the design and development of sustainability drivers according to [15], contingent upon the collaboration and exchange of information. We have hypothesised, based on this discussion, that;

1. H1: Sustainability challenges have a significant relationship with drivers of green sustainable practices.

3.3. Limited green sustainability materials access

The sustainable materials are significant in terms of green practices undertaken by the garment and textile industry, as they are conditional on the use of more environmentally friendly materials like organic cotton, biodegradable dyes, etc. In less developed economies, these materials cannot be easily obtained due to the absence of agricultural investment, poor infrastructure to process them, and fragmented markets [22,23]. This makes them employ artificial or foreign raw materials, thereby lowering the sustainability. Similar to big data, which requires advanced structures, small amounts of green materials require combined interventions to be achievable. A study by [24] discovered that responsible sourcing and eco-labels are better due to the existence of green materials. The access is not only low, but it also weakens the technical feasibility, and it also diminishes the impact of other sustainability drivers like consumer demand. The conflict of material constraints and economics, technology, and policy rates affects the value chain structure and scalability of the sustainable practices [25]. Therefore, the following hypothesis is used in the study;

1. H2: Limited access to green sustainability materials has a significant relationship with drivers of sustainable practices.

3.4. Economic constraints

Economic constraints play a central but often ambivalent role in shaping sustainable practices within developing countries’ garment and textile industries. Low profit margins, limited access to capital, and high relative costs of sustainable inputs and technologies act as key barriers to the adoption of green manufacturing. Economic constraints serve as an important factor in achieving sustainability in the small-scale garment and textile sector in Ghana, since it implies that the industry is constrained by financial concerns that influence the choice of materials, labour, technology, and cost. The initial costs, which may be prohibitive when it comes to small enterprises, may include the implementation of sustainable practices through the purchasing of eco-materials, cleaner technologies, or certifications [26,27]. Similar to the case of big data analytics, the need to become sustainable should be complemented by the development of particular financial policies and strategies [28]. Otherwise, sustainability initiatives will be slowed down without such support. However, the recent studies suggest that those economic limitations have already led to some adaptive measures, including innovation hubs that are donor-based and collaborative funding projects that enable the possibility of sustainable experimentation despite the financial limitations of the environment [13]. In addition, economic constraints can either compound or lessen the performance of other sustainability concerns, and this proves the plausibility of financial competence in the sustainability continuum. Since it was in line with the SDT, in this study, the hypothesis was:

1. H3: Economic constraints have a significant relationship with the drivers of sustainable practices.

Based on the reviewed literature, the study therefore proposed the following conceptual framework, as seen in Fig 1 below.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Proposed conceptual framework.

https://doi.org/10.1371/journal.pone.0338270.g001

4.1. Data collection and sampling

Quantitative approach was deployed in the study. The use of the quantitative approach was to minimise the researchers’ bias through standardised tools like surveys and statistical analysis, ensuring consistent, replicable results across large samples. The sample was drawn from a list of registered Garment and Textiles SMEs in the manufacturing and production sectors from the National Board for Small-Scale Industries (NBSSI) regional office, Western region. SMEs in the manufacturing and production sectors within five districts were used for the data collection, specifically Essikado-Ketan, Takoradi, Efia, Kwesimitsim and Sekondi. The selected districts are notable for their business operations within the region. Respondents were drawn using simple random sampling from the accessible population, collected via online and offline questionnaires, to ensure representation across the targeted districts [29–32]. Both online and offline data collection approaches were deployed. A cross-sectional research design was used in the study since the data were collected once for its analysis. To be more specific, 398 structured questionnaires were administered through both online and offline means. Out of this, 350 responses were received, with a non-sampling error of 35 questionnaires. This affirmed that the 35 non-sampling error questionnaires indicated incomplete questionnaires among other factors, which makes it invaluable for it to be used for the final data processing and analysis. Given this, a total of 315 valid responses were used in the data processing and analysis. A five-point Likert scale, particularly (1 = strongly disagree, 5 = strongly agree) was adopted. The structured questionnaire was answered by owner-managers of the selected SMEs in the manufacturing and production sector. Participants used in the study’s data collection were assured of confidentiality and also promised that the data gathered was for academic purposes. Given this, participants’ informed consent was sought. A total of 45 respondents were used in the pilot study process to ensure corrections and modifications of the study’s research questions. Four months were used in gathering the needed data from January to April 2025. In addition to that, the participants were assured of anonymity, which served to motivate them to provide sincere feedback and to protect their identity. The study applied the partial least square structural equation modelling (PLS-SEM) 4.0 version, which is suitable when complex predictive equations are used [33]. This allowed latent variable tests and path connections, bearing in mind the non-normality of organisational research data. Table 1 below depicts the details of the respondents’ profiles.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Respondents’ Profile.

https://doi.org/10.1371/journal.pone.0338270.t001

4.2. Measurement and instrument development

All the items or scales used in the questionnaire were adapted and modified from existing studies, such as Challenges to Sustainability [34–36], Limited Access to Materials [37,38], Economic constraints [39,40], and Drivers of Sustainable Practices [41,42]. The questionnaire design followed best practices to minimise common method bias in measurement error, as indicated by [43]. This was achieved through clear wording, logical question ordering, and optimised item sequencing to ensure high data quality and respondent understanding. To further mitigate potential bias, questions measuring independent and dependent variables were distributed across different survey sections. This procedural remedy, supported by prior research, reduces artifactual covariance [44]. The questionnaire underwent pilot testing and expert review to evaluate its reliability and validity. It was subsequently revised based on the feedback received. The reliability of the construct was measured using Cronbach alpha for internal consistency.

4.3. Common method bias

To address potential common method bias (CMB), the researchers implemented both procedural and statistical controls. Following [45], the study used clear construct items, assured participant confidentiality and anonymity on the cover page, and informed respondents of their voluntary participation and right to withdraw at any time. Additionally, we conducted multicollinearity analysis using variance inflation factor (VIF) values to assess common method variance (CMV). No VIF values exceeded the critical threshold of 10, and all constructs surpassed the recommended minimum of 0.5, as per [46,47]. These findings confirm that multicollinearity and CMV do not pose significant concerns. Table 2 reports the minimum CMB observed.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Construct, Indicator, Loading, VIF, CR, AVE, and CA.

https://doi.org/10.1371/journal.pone.0338270.t002

5.1. Assessment of the measurement model

Partial Least Square Structural Equation Modelling (PLS-SEM) was used to examine relationships among latent constructs. The measurement model was evaluated in several steps, including factor loadings, convergent validity, and discriminant validity. First, indicator factor loadings were inspected to ensure they met the minimum threshold of 0.5, retaining only those that improved overall model fit [48,49]. Next, internal consistency was assessed using Cronbach alpha and composite reliability (CR), while convergent validity was confirmed by average variance extracted (AVE), following [48]. All constructs exceeded the acceptable internal consistency threshold of 0.7 and demonstrated strong convergent validity. Table 2 summarises key measurement items, including factor loadings, Cronbach alpha, CR, AVE, and VIF values for all constructs.

Discriminant validity was established using the Fornell-Larcker criterion, which requires the square root of each construct’s AVE to exceed its correlations with other constructs. Results showed all constructs had AVE values above the recommended 0.5 threshold, with inter-construct correlations lower than their respective AVE square roots. This confirms the constructs.

5.2. Assessment of the structural model and hypothesis testing

The third step in the structural model analysis involved hypothesis testing and R² calculation, following [50]. Before this, collinearity among predictor variables was assessed using variance inflation factor (VIF) values, which ranged from 1.3 to 1.7, within acceptable limits, confirming effective model performance.

To address potential common method bias (CMB) inherent in cross-sectional survey data, preventive measures were implemented per Scott and Bruce (1994). Harman’s one-factor test revealed three eigenvalues exceeding 1, accounting for 70% of total variance, with no single factor dominating. Recognising Harman’s test limitations, we additionally applied the Partial General Factor approach within the PLS model, as recommended by [51]. Once measurement model reliability and validity were confirmed, explanatory power was evaluated via R² values (Table 3, Fig 2). Hypothesis testing results, as seen in Table 4, indicated that two of the hypotheses were insignificant, while one was not.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Discriminant Validity.

https://doi.org/10.1371/journal.pone.0338270.t003

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Hypothesis Testing.

https://doi.org/10.1371/journal.pone.0338270.t004

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Estimated model.

https://doi.org/10.1371/journal.pone.0338270.g002

7.1. Theoretical implications

This study advances understanding of sustainable practices in Ghana’s garment and textile industry by highlighting how contextual challenges influence adoption. It bridges gaps in sustainability literature specific to developing economies. First, the study reinforces and extends theories and concepts of sustainability, particularly research on the manufacturing industry. The findings demonstrate that sustainability functions not merely as a voluntary strategic tool but as a necessary response to institutional, environmental, and operational threats. This supports the core tenet of Sustainable Development (SDT) theory that organisational behaviour is shaped by external pressures such as regulations and norms [58,59]. Theoretically, this implies that for firms in emerging markets, pursuing sustainability is a legitimacy-seeking mechanism aimed at aligning operations with international standards, thereby extending the theory’s applicability to contexts where global supply chain pressures intersect with local institutional voids.

Second, the study advances resource-based by introducing a scarcity-driven innovation mechanism. While organisational resources traditionally emphasise how unique and valuable resources shape competitive advantage [60], this study shows that resource scarcity itself can act as a catalyst for sustainable innovation. The findings imply that limited material access does not simply constrain firms; it can force creative problem-solving and eco-innovation [61]. The theoretical implication is significant for developing economies where scarcity becomes an endogenous driver of sustainability rather than merely an exogenous barrier. This reframes resource-based assumptions to account for how resource constraints, when institutional pressures are present, generate novel sustainable practices.

Third, the study challenges and refines conventional economic determinism. Contrary to the prevailing assumption that financial costs are the primary barrier to sustainability adoption [62], this study finds that firms may pursue sustainable initiatives despite economic disadvantages. This theoretical implication is twofold: (a) non-financial drivers such as supply chain dynamics, regulatory pressure, and reputational concerns can override economic constraints [63]; and (b) existing sustainability frameworks, largely developed in high-income contexts, overemphasise financial logic and underemphasise socio-institutional pressures.

7.2. Practical implications

The conclusions drawn from this study have implications for the various actors in the garment and textile industry. Industry actors can learn from the study that sustainability programs can be easily put in place even in situations of resource scarcity. Companies are advised to analyse the constraints that face them as opportunities to come up with innovations that can help sustainably address the issue. Scarcity of resources is not just viewed as a constraint but can also lead companies to come up with innovative ways of doing things. This is because the use of recycling and the reuse of products has been associated with sustainable production programs [64]. In addition, companies will increase their competitiveness in the international market due to increased sustainability compliance requirements in such markets.

From the policy standpoint, the research indicates that any effort aimed at promoting sustainability needs to transcend beyond the provision of financial instruments. Although finances are essential, more attention should be devoted to increasing access to sustainable material resources and providing the required infrastructures for ensuring sustainable production. The key role here can be played by policymakers who need to facilitate efficient logistics channels and promote the production of environmentally friendly materials domestically while minimising obstacles related to importing sustainable materials. This corresponds to international policy advice, which highlights the role of systemic approaches to developing sustainable logistics and locally sustainable industries [65,66]. Additionally, innovation-related policies can substantially boost the process of transformation towards sustainability.

For development agencies and other assisting entities, the results of the analysis suggest the necessity to implement capacity-building programs that would emphasise knowledge building and resource access issues. Collaboration efforts between domestic companies and foreign organisations can help reduce the current gaps and increase the use of sustainable practices within the industry, which is discussed in sustainable development models [67]. These collaborations can also help to gain new skills and technologies essential for achieving sustainability. Finally, the results obtained suggest that the shift in perception towards sustainability is crucial. Instead of treating it as an additional cost to sustain, it should be treated as a source of competitive advantage and a means to achieve better results in the future. This point of view is also supported by previous evidence indicating that sustainability can generate shared value for businesses and improve performance outcomes [68]. Hence, managers should start considering sustainability as an integral part of their business activities.