Adaptability and integration influence on adaptive capacity of small- and medium-scale construction organisations

Abstract

Purpose

Adaptability and integration (ADI) are the core ingredients for environmentally sustainable construction (ESC), which preserves the ecology from unsupported human activities. However, the approach is lagging in developing countries, which has led to studying the influence of ADI on the adaptive capacity of small- and medium-scale construction organisations.

Design/methodology/approach

The research employed a quantitative methodology, collecting 400 responses as a sample size. A construct of 14 influential factors concerning ADI within the Ghanaian small and medium-scale construction industry was developed. The data obtained from participants underwent analysis using SPSS version 26. The validity of the study’s findings was assessed by applying structural equation modelling (SEM) within the AMOS software.

Findings

It was evident that innovation advancement and ongoing training and evaluations significantly influence ADI for adaptive capacity. Moreover, the system internally and vulnerability (SIV) and perceived need for implementation (PNI) sub-scales were the main latent components for best construction practices.

Practical implications

Ghana’s small- and medium-scale construction organisations have yet to fully recognise the importance of ADI in enhancing adaptive capacity for the best ESC. However, the results indicated that ADI constructs will significantly influence implementation outcomes to ensure ESC.

Originality/value

The originality of this research also resides in identifying how ADI affect small- and medium-scale construction organisation’s ability to ensure ecologically sustainable building practices.

Citation

Pim-Wusu, M., Aigbavboa, C., Adu Gyamfi, T. and Didibhuku Thwala, W. (2024), "Adaptability and integration influence on adaptive capacity of small- and medium-scale construction organisations", Frontiers in Engineering and Built Environment, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/FEBE-01-2024-0003

1. Introduction

Adaptability and integration regarding environmentally sustainable construction (ESC) refers to the capability of the construction industry to have an adaptive capacity to change conditions whilst executing expertise within firms (Waidyasekara and Senaratne, 2020). The explanation includes companies enabling smooth integration with innovative approaches and allowing the organisation’s assessment and management tools to be integrated directly into current systems by creating a cohesive and efficient workflow (Otter, 2019). Gunatilake and Perera (2018) subscribe that the sustainable construction method tailors the integration of innovative processes to fit the specific needs to ensure that the services complement the team in implementing sustainable development goals (SDGs) with the commitment to providing detailed documentation and support to assist the organisation in implementing ESC with ease and confidence. Adaptability and integration (ADI) in the construction industry for ESC practices are indispensable; however, their influence is a crucial problem (Ranadewa and Siriwardena, 2021). He et al. (2021) explain ADI by matching them to the social inclusiveness system that allows construction professionals to share common knowledge; more so, their influence is lacking. Construction companies, particularly small- and medium-sized ones in developing nations, lack the flexibility and integration necessary to pursue the functional criteria that are necessary to ensure the successful accomplishment of the ESC agenda, as Agyekum et al. (2022) demonstrate. The Ghanaian construction sector was the focus of research on adaptation capacity by Pim-Wusu et al. (2023), however, the study did not account for the influence of ADI.

To develop effective guidelines for a paradigm shift that will foster effective adaptation to ESC, particularly the small- and medium-scale construction in Ghana, the study country, Pim-Wusu et al. (2022a, b and c) contend that a study on the influence of ADI should be started.

Debrah et al. (2023) researched developing countries’ perspectives on the drivers for green city development, focussing on Ghanaians; the study identified 19 drivers without ADI, which seems to be a big gap. Agyekum et al. (2022) examined the key drivers for green building projects in Ghana and how best sustainable construction could be achieved and attract finances; eight key important drivers were mentioned. However, ADI never emerged as factors. In light of this, the study aimed to ascertain how Ghana’s small- and medium-scale construction industry can adjust regarding ADI. The analysis offers a succinct indicator variable for the integration and adaptation constructs taken from the literature.

2. Literature review

2.1 Theory that underpinned adaptability and integration

Rogers developed the diffusion of innovation theory in 1962 as a step-by-step approach in which innovation is implemented within certain mediums over some time amongst the people involved in society. The spread of messages and awareness of new ideas, such as ESC adaptation, results in communication diffusion. Communication is a process by which information is exchanged between participants within the construction industry to circulate through mutual understanding (Rogers, 2003).

However, Dearing (2008) came up with the theory of dissemination and implementation model to fuse and systematically facilitate evidence-based intervention to utilise efforts to spread and integrate methodologies out from the diffusion of innovation theory, which relates more to ADI. In ESC, the dissemination and implementation model can be adapted to replicate technology transfer from developed countries' construction organisations to developing countries. Dearing (2008) emphasises that the adaptation process usually takes place in communication, broadcasting the message to the broader members of the construction industry without a direct reaction from them. The explanation above correlates with the small- and medium-scale construction organisation’s adaptive capacity. Otter (2019) agrees that ADI are “needs” for those in the built environment, and they must feel confident when presenting evidence. Small- and medium-scale construction firms must understand what their peers are learning about ADI. Agyekum et al. (2022) suggest that there must be a sense of continuity in belonging to the construction industry when members change. The dissemination and implementation model emphasises that the small- and medium-scale construction organisations tie together the built environment adopters. Any credible professionals and stakeholders from the built environment sector must be part of the innovation process through ADI. Pim-Wusu et al. (2023) suggest that those with such credibility must contribute their bit through ADI. Dearing and Cox (2018) opine the likely success in ADI would depend on the construction sector’s perspective of change agents. If the agent of change correctly identifies the built environment leaders serving as sources of example, then the adaptability is likely to materialise. Dearing and Cox (2018) conducted empirical studies and observed that new ideas and practices relating to ESC usually spread freely through interpersonal contacts. The arrangement primarily consists of mutual communication between members of the construction industry. The new practices’ essential influences include social contact with small- and medium-scale construction organisations, social interaction with industry members and interpersonal communication between professionals in the built environment (Dearing, 2008). Arthur-Aidoo et al. (2016) use “network analysis” to describe social interactions between small- and medium-scale construction organisations, built environment professionals and all stakeholders involved in discoveries. The expression is about identifying individuals in small- and medium-scale construction firms during an adaptive capacity process. From these observations, opinion leaders can initiate the new idea and function as role models and supporters of the approach (Raouf and Al-Ghamdi, 2020).

The dissemination and implementation model underscores how construction organisations could facilitate evidence-based intervention to utilise efforts to spread and integrate methodologies once sufficient analysis justifies the knowledge acquired in one sector. The theory underpins the influence of ADI on adaptive capacity for construction organisations to deliver ESC practices.

2.2 Adaptability and integration factors influencing adaptive capacity for ESC

According to Sultan and Alaghbari (2021), Debrah et al. (2023), Chen et al. (2024), Van Ellen et al. (2021), Manewa et al. (2016) and Yu et al. (2018), for ESC to take place ethically, developing nations must address issues of ADI to ESC in a specific manner, regardless of the social structures. Lean construction is suggested as the future path for ESC by Ebekozien et al. (2023), Atombo et al. (2015), Pim-Wusu et al. (2023) and Stålberg and Fundin (2018) as a means of fostering ADI principles to support sustainable construction communication capabilities. Besides lean construction design and materials handling, high resource standards and innovation advancement construction methods, sustainable construction project delivery needs a critical role in fostering ESC development (McDermot et al., 2020; Hamida et al., 2023; Otter, 2019; Chen et al., 2024; Manewa et al., 2016). Atombo et al. (2015), Debrah et al. (2021), Pim-Wusu et al. (2023) and Fiyinfoluwa et al. (2022) found that the inner factors/morals of clients and developers can actively modify pledges to ADI to drive the social structures of the construction industry towards sustainability. For small- and medium-sized construction enterprises to effectively engage with one another and embrace green practices, Waidyasekara and Senaratne (2020), Pim-Wusu et al. (2022b), Otter (2019) and Hamida et al. (2023) identified ADI of the industry’s active reworking commitment. According to the study’s findings, “taking action” to enhance environmental conditions is required by standards and regulations that support ADI. Nonetheless, emerging nations must allow ESC ADI to happen independently based on organisational goodwill (Pim-Wusu et al., 2022a; Agyekum et al., 2022; Raouf and Al-Ghamdi, 2020; Chen et al., 2024). According to McDermot et al. (2020), Pim-Wusu et al. (2022c), Otter (2019), Hamida et al. (2023) and Yu et al. (2018), ecologically sustainable buildings in poor nations may require ADI to support literacy level and understanding along a route towards sustainable development (SD) and ESC. In this study’s context, and based on the above definitions, ADI are operationalised as “the key elements/conditions/situations that need to be created to make it possible for developing countries to adopt ESC successfully”. It is equally important for countries to seek ESC ADI, which would steer the head of the concept and give the necessary guidance for involvement in policymaking that hinders possible ADI.

According to Cummings et al. (2015), He et al. (2021), Pim-Wusu et al. (2023) and Rashidian et al. (2023), small- and medium-scale construction firms must have extensive views and enthusiasm to execute projects by minimising pollution, i.e. preventing noise and dust by fitting silencers and dust collectors to machinery and equipment. Cummings et al. (2015), Raouf and Al-Ghamdi (2020), Hamida et al. (2023) and Fiyinfoluwa et al. (2022) illustrate that small- and medium-sized construction companies need effective interaction to provide innovative solutions that work in the community. Sultan and Alaghbari (2021), Dearing and Cox (2018), Otter (2019), Debrah et al. (2023) and Askar et al. (2021) argue that to build sustainable cities and communities, businesses should increase their involvement in policymaking of reasonably priced natural ecological cycle materials and combine organisational goodwill with longer-lasting construction materials. Wang et al. (2013), Tang et al. (2020), Dalirazar and Sabzi (2020) and Askar et al. (2021) submit that a lack of ADI mechanisms and knowledge discrimination could hinder clients’ understanding of SD and be a barrier to reasonably practicing ESC. Moreover, Van Ellen et al. (2021), Taherkhani (2023), Gunatilake and Perera (2018) and Rashidian et al. (2023) postulate a need for ADI of organisations and ignore the class of individual richness of human resource planning for construction projects, which is also a clear message.

In an investigation of the nature of conventional construction methods, Fonseca et al. (2020), Atombo et al. (2015), Van Ellen et al. (2021) and Othman and Abdelwahab (2018) looked at several strategies that might promote a pragmatic approach to ADI into the practice of sustainability in construction. When they combined the systems of industrialised nations, they discovered that the integrated concurrent with constant training and evaluations might enhance the performance of building projects and make it easier to include sustainable building practices. In this case, research outlined the objectives, exercises and sustainability factors for each stage of the project cycle, from need assessment to upkeep and operation.

Stålberg and Fundin (2018), Ranadewa et al. (2021), Agyekum et al. (2022) and Rashidian et al. (2023) have observed that disparities in wealth distribution do not tally with ADI; therefore, small- and medium-scale construction organisations must be encouraged to have organisational goodwill to achieve sustainable practices in the industry. Developed countries are in the lead; developing countries, especially Ghana, need ADI to influence the adaptive capacity to drive small- and medium-scale construction firms for ESC. The literature has captured several ADI factors that could influence the adaptive capacity for ESC as all-inclusive variables; however, Table 1 below shows the constructs generated from the literature to test small- and medium-scale construction firms’ suitability for ESC.

3. Methodology

The research approach - The study employed a quantitative methodology, utilising a literature review and a field questionnaire survey conducted through Google Forms. A literature review was conducted to determine the variables within the body of available literature. Quantitative research evaluates quantities or variables in a study to produce generally relevant responses (Leedy and Ormrod, 2015; Garg, 2019). The objective was to find and confirm linkages to contribute to the existing body of knowledge.

The study population - Neuman (2014) and Cooper and Schindler (2013) define the study population as every member of the group who captures the researcher’s attention. The target demography for the field survey was participants affiliated with professional associations representing the construction sector in each of Ghana’s 16 regions. The databases of the relevant institutions were searched for these individuals. Public sector organisations, consultants and contractors employed the specialists in question. Along with being in good standing within the civil division, there were 1,000 members of the Ghana Institution of Engineering (GhIE). The division’s participants included the 1,800 Ghana Institution of Surveyors (GhIS) members in good standing employed with consultancies, contractors and government agencies. As an illustration, the Ghana Institute of Architects (GIA) has members who work in the building business as consultants, contractors or public sector companies in good standing. About 2,000 members are included on the GIA register. The targeted population consisted of 4,800 members of the GhIE, GhIS and GIA, as they possess significant knowledge in the field and are the main decision-makers regarding the nation’s SDG targets.

Sampling procedure and sample size – Malhotra (2021) defined the sample size as the number of participants in a sample after accounting for statistical power needs and data collection costs. According to Pim-Wusu et al. (2023), two statistical features in advanced investigations are impacted by sample size: the study’s control over validating the major constructs and the accuracy of estimations. According to Arthur-Aidoo et al. (2016), the demographic characteristics, sample alignment with the study, and the necessary level of accuracy all influence the optimum sample size for a given research project. It is crucial to remember that the sample size was not the same as the total population of the chosen professions. The intended data analysis should assess the characteristics of the population, the appropriate sample size and the level of precision necessary to achieve research objectives, as Neuman (2014) suggested. The sample size for the study was calculated using the following calculations:

n=X2NP(1−P)e2(N−1)+X2P(1−P),

Where:

• n = sample size;

• X² = chi-square value with 1 degree of freedom (equal to 3.841);

• N = population size;

• P = population proportion (assumed to be 0.5) and

• e² = margin of error (0.05).

With 4,800 people, X² = 3.841, p = 0.5 and e² = 0.05, if all this is substituted in the formula n = 355.7. However, because structural equation modelling (SEM) was used, which requires a large sample size, the study rounds up the sample number to 500 to gather large data, boosting the results. According to a study by Ebekozien et al. (2023) on the training requirements of professionals working in the built environment, larger sample numbers increase the accuracy of population estimates (SEM). According to Pim-Wusu et al. (2023), a sample size of 300 participants or more is considered normal for SEM analysis; 500 participants or more yield perfect findings, and 1,000 participants or more yield great results. For this study, 400 out of the 500 questionnaires administered were retrieved, accounting for 80% of the sample size. According to Neuman (2014), the planned data analysis should identify the ideal sample size, the degree of accuracy required to meet the study goals and the characteristics of the population. Dependability of the research: experts in the building sector tested the data collection instruments. The questionnaires for the study were piloted so that any necessary revisions could be made. A pilot survey can help researchers better understand or refine research ideas, select the most effective approach and estimate the time and resources required to conduct a more comprehensive study, according to Ashley (2020) and Adu Gyamfi et al. (2022). A Cronbach alpha test was used to determine the reliability of the study variable to ensure that the data are reliable and suitable for measuring the research goal. Per Norušis (2011), the Cronbach alpha cut-off value should be 7.0 or greater to guarantee the variables’ reliability. The research yielded a Cronbach alpha coefficient of 0.864.

Survey administration – The study’s literature provided a theoretical underpinning that influenced the creation of survey questions. Figure 1 illustrates the study methodology. The questionnaires for the study were administered using Google Forms because many respondents were required to answer questionnaires online. The study obtained a higher return rate from Google Form administration because the researchers contacted each respondent on the phone and made regular follow-ups.

Data analysis technique - The responses of 400 participants were examined using SPSS version 26. The software supported several analytical methods, including tools like mean, standard deviation (SD), frequencies and percentages for descriptive analysis. Information processing was further aided by the exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) approaches used for extensive data assessment in the study. It is possible to evaluate the reliability of the data and the information by employing SEM.

4. Analysis and results

In this part of the analysis, the respondents’ backgrounds were based on their positions, years of experience, highest educational qualification and company nature. The respondents’ backgrounds were analysed using frequencies and percentage distribution.

From Table 2, regarding the position held by respondents, 67 respondents, representing 16.75%, were company directors, whereas 65 respondents, representing 16.25%, were project managers. Also, 64 respondents, representing 16.00%, were site managers. Moreover, 64 respondents, representing 16.00%, were field engineers, whereas 49 respondents, representing 12.30%, were consultants. Again, 32 respondents, representing 8.00%, were consultant supervisors, 28 respondents, representing 7.0%, were quantity surveyors and 24 respondents, representing 6.0%, were architects.

Regarding years of experience, 98 respondents, representing 24.5%, had worked for 6–10 years, whereas 97 respondents, representing 24.3%, had worked for 11–15 years. Also, 94 respondents, representing 23.5%, had worked for 16–20 years, and 69 respondents, representing 17.3%, had worked for 20 years and above. Moreover, 42 respondents, representing 10.5%, had worked for 1–5 years.

With regards to the highest educational qualification, the majority of the respondents, as represented by 201 (50.3%), had master’s degrees, 144 (36.0%) had bachelor’s degrees, 43 (10.8%) of the respondents had a doctoral degree and 12 (3.0%) of the respondents held higher national diploma (HND).

With regards to the nature of the company, 103 (25.8%) were consultancy forms, 102 (25.5%) were D2K2 contractors, 95 (23.8%) were public sector firms, 64 (16.0%) were D3K3 contractors and 36 (9.0%) were D4K4 contractors.

4.1 The indicator variables for adaptability and integration construct

This section of the analysis seeks to examine the ADI factors. Items under this section were analysed using mean and SD. Table 3 presents the findings;

The result shows that the respondents recorded a mean score of 3.85 (SD = 1.02) for the extent to which innovation advancement influences adaptive capacity development. Again, constant training and evaluations recorded a mean extent of 3.78 (SD = 1.03). Moreover, effective interaction recorded a mean level of extent of 3.76 (SD = 0.96). A mean extent level of 3.75 (SD = 1.02) was recorded for the literacy level and understanding. Active reworking commitment recorded a mean level of extent of 3.72 (SD = 1.02). Again, the communication capabilities recorded a mean level of extent of 3.68 (SD = 0.99). Additionally, views and enthusiasm recorded a mean level of extent of 3.67 (SD = 0.99). In addition, a mean extent level of 3.63 (SD = 1.09) was recorded for knowledge discrimination. Involvement in policymaking as a factor recorded a mean level of extent of 3.64 (SD = 1.00).

Again, organisational goodwill recorded a mean level of extent of 3.57 (SD = 1.02). Again, that class of individual richness recorded a mean level of extent of 3.53 (SD = 1.02). The social structures recorded a mean level of extent of 3.48 (SD = 0.94). Inner factors/morals recorded a mean level of extent of 3.44 (SD = 0.97). Furthermore, Disparities in wealth distribution recorded a mean level of extent of 3.42 (SD = 1.04).

4.2 Exploratory factor analysis: dimensionality of adaptability and integration (ADI) construct

ADI construct was assessed by grouping the 14 variables into identifiable scale components using EFA (Pallant, 2013). The definition of the extraction and rotation approach was maximum likelihood with Varimax rotation, which assesses the construct. Bartlett’s sphericity and Kaiser–Meyer–Olkin (KMO) tests were used to confirm the sample’s adequacy and applicability. Rehbinder (2011), Agumba (2013) and Farrington (2009) all point out that the KMO and Bartlett’s test of sphericity are used to establish sample adequacy. Concerning the KMO cut-off value of 0.70 and Bartlett’s test of sphericity of p < 0.05, as suggested by Hair et al. (2022), the study obtained a KMO figure of 0.864 and Bartlett’s test of sphericity of p < 0.000, which make the data to be analysed using EFA. Once more, an investigation was done to determine how many components are included in the ADI constructs. Table 4 demonstrates how the ADI characteristics have been loaded into Factors 1 and 2. The Eigenvalues of Factors 1 and 2 were 2.638 and 1.254, respectively. Still, 44.795% of ADI variations may be explained by the two components. As a result, all 14 components (ADI 1, ADI 2, ADI 3, […], ADI 14) were verified for loading and measurement.

According to Table 5, all of the items had factor loadings for their components that were greater than 0.5 when factor loading was set at a threshold of 0.5, which is higher than the ideal value of 0.40, as stated by Field (2013) and Hair et al. (2022). This implies that the elements were correctly stacked on top of the determined build factors. It’s noted that, from the sub-scale in Table 5, the first component consisting of seven items recorded a threshold exceeding 0.5. These items, namely “organisational goodwill”, “inner factors/morals”, “views and enthusiasm”, “social structures”, “active reworking commitment”, “literacy level and understanding” and “class of individual richness”, measure the construct of the system internally and vulnerability (SIV). Seven items recorded a threshold, exceeding 0.5 for the second component from the sub-scale. These items, namely “effective interaction,” “constant training and evaluations,” “communication capabilities”, “involvement in policymaking”, “innovation advancement”, “knowledge discrimination” and “disparities in wealth distribution”, measure the construct of perceived need for implementation (PNI).

After extracting the two components using the EFA, the adjusted item-total correlation for each of the two components’ items was extracted using the recommended 0.30 cut-off value. The analysis determined that the items accurately measured the components because the first component, SIV, had a Cronbach’s alpha of 0.801, and the second component had an alpha of 0.807 PNI, as shown in Table 5. These numbers demonstrate an adequate level of internal reliability, as stated by Nunnally and Bernstein in 1994.

4.3 Structural equation modelling (SEM) for adaptability and integration construct

A CFA was administered after the constructs’ satisfactory performance in demonstrating unidimensionality and reliability using EFA. According to Hu and Bentler’s (1999) advice, the factors under consideration are satisfaction with goodness-of-fit analysis based on a three-statistics approach to fit indices. Using sample data, the ADI model produced an S-Bχ2 of 4.577 with 53 degrees of freedom (df) and a probability of p = 0.0000. A strong match was suggested by the chi-square value, which showed a considerable divergence of the sample data from the proposed model. It is important to remember that the chi-square test is subject to sample size variations and functions better as an index of descriptive fit than a rigorous statistical test (Kline, 2023).

Based on its Comparative Fit Index (CFI) score of 0.914, which is greater than the 0.90 cut-off requirement, the ADI factor model is accepted. Comparably, Table 6 shows that the Normed Fit Index (NFI) value of 0.973 is greater than the NFI ≥0.90 cut-off value. That NFI has done well is notable, then. With a satisfied cut-off value of less than 0.80, the Parsimonious Normalised Fit Index (PNFI) has a good-fit score of 0.578. Furthermore, the Root Mean Square Residual (RMSR) of 0.039 is less than the acceptable value of 0.05, and the Goodness-of-Fit Index (GFI) value of 0.903 is greater than the cut-off level of 0.90. To verify the model’s regularity, other fit indices, including the Parsimonious Comparative Fit Index (PCFI), Root Mean Square Error of Approximation (RMSEA) and Incremental Fit Index (IFI), were used, as Table 6 shows.

The correlation values, standard errors and test statistics of the final 12-indicator model for the ADI components are displayed in Table 7. Statistical significance is indicated by all correlation values being less than 1.00 and p-values being less than the significant value of 0.05. It is decided that the estimations are statistically significant and reasonable. With a parameter coefficient of 0.710, the indicator variable with the variable effective interaction (PNI1) has the greatest standardised coefficient. Most parameter estimates exhibit strong correlations with values near 1.00, indicating a high degree of linear relationship between the unobserved and indicator variables (SIV and PNI). The R square values also show that the factors account for a more significant proportion of the variance in the indicator variables, with values nearing the target value of 1.00. In light of the ADI factors, the results imply that the indicator variables significantly predict the unobserved components (SIV and PNI).

Figure 2 indicates CFA model for ADI construct, which has a two-factor model and thus SIV and PNI factors. However, the correlation values, standard errors and test statistics for the final 12-indicator model of the ADI factors are shown in Table 8, indicating statistical significance because every correlation value and every p-value is below the significant value of 0.05 and 1.00, respectively. Based on statistical significance, the estimations are deemed plausible. Equipped with the indicator variable PNI1 (effective interaction), the parameter with the greatest standardised coefficient, 0.702, is linked. Accordingly, the PNI component of the ADI components reveals that effective interaction is a significant aspect. According to the findings, there is a PNI inside the concept, largely due to various aspects, one of which is effective interaction.

The finding implies an intense linear relationship between the indicator and unobserved variables, such that most parameter estimations had high correlation values closer to 1.00 (SIV and PNI). A strong correlation exists between the underlying constructs and the measured indicators, as seen by the high degree of linear association. Furthermore, the R square values’ proximity to the ideal 1.00 value indicates that the factors (SIV and PNI) account for a sizable percentage of the variation in the indicator variables. This suggests that the components of the ADI construct have been found to have a substantial impact on the variation shown in the indicators that have been measured. The findings, taken together, support the validity and reliability of the ADI construct in capturing the intended dimensions of ADI within the study’s context by indicating that the indicator variables within the construct are strong predictors of the unobserved components (SIV and PNI).

5. Discussion of findings

The 12 factors were categorised into two sub-scales, namely SIV and PNI after CFA analysis.

5.1 Sub-scale 1: system internally and vulnerability (SIV)

On the sub-scale “system internally and vulnerability (SIV)”, six factors were found from the CFA and grouped into a single construct. “Organisational goodwill”, “views and enthusiasm”, “inner factors/morals”, “social structures”, “active reworking commitment” and “literacy level and understanding” are amongst them. These components assess the SIV. This confirms the findings of Pim-Wusu et al. (2022a), Agyekum et al. (2022) and Raouf and Al-Ghamdi (2020) that ADI would be enhanced by the SIV. The results are consistent with previous research, highlighting the multifaceted characteristics of the factors affecting ADI in the adaptive capacity (Fonseca et al., 2020; Cummings et al., 2015; Raouf and Al-Ghamdi, 2020). With a 0.64 factor loading, organisational goodwill was rated as the highest variable under the SIV sub-scale. This finding is consistent with the findings of Chen et al. (2024) and Raouf and Al-Ghamdi (2020), who found that organisational goodwill guarantees that emerging nations permit ESC ADI to occur independently. Furthermore, Rashidian et al. (2023) have noted that to achieve sustainable practices in the industry, small- and medium-sized construction organisations need to be encouraged to have organisational goodwill. The SIV sub-scale rated views and enthusiasm as the second highest variable, with a factor loading of 0.61. This finding is consistent with He et al.’s (2021) assertion that small- and medium-sized construction companies need to have broad views and enthusiasm to carry out projects by minimising pollution, i.e. preventing noise and dust by equipping machinery and equipment with silencers and dust collectors. The results align with the findings of Pim-Wusu et al. (2023) and Fiyinfoluwa et al. (2022) that inner factors/morals of clients and developers can actively modify pledges to ADI to drive the construction industry towards sustainability. Inner factors/morals were the third-rated variable under the SIV component, with a factor loading of 0.60. The social structures variable came in at number four under the SIV component. This finding is consistent with the findings of Debrah et al. (2023), Sultan and Alaghbari (2021) and other scholars who have noted that to ensure ESC, developing countries' social structure provision must address issues of ADI.

5.2 Sub-scale 2: perceived need for implementation (PNI)

EFA was also used to identify and combine six components into a single construct on the subscale “perceived need for implementation (PNI)”. Amongst them are “knowledge discrimination”, “participation in policymaking”, “effective interaction”, “ongoing training and assessments” and “communication skills”. Effective interaction was the first PNI variable identified by the CFA model. It had a factor loading of 0.70. This result is consistent with the findings of Hamida et al. (2023) and Fiyinfoluwa et al. (2022), which show that effective interaction is necessary for small- and medium-sized construction companies to provide creative solutions that benefit their communities. The variable for constant training and assessments was ranked second with a factor loading of 0.67; this finding resonates with the findings of Othman and Abdelwahab (2018), who suggested that constant training and evaluations could improve building project performance and facilitate the inclusion of sustainable building techniques. The CFA model identified the third PVI variable, participation in policymaking, with a factor loading of 0.67. This finding is consistent with the findings of Debrah et al. (2023) and Askar et al. (2021), which suggest that to construct sustainable cities and communities, businesses should become more involved in policymaking regarding reasonably priced natural ecological cycle materials and combine organisational goodwill with longer-lasting construction materials. Additionally, the CFA model identified communication skills as the fourth PNI variable, with a factor loading of 0.65. This finding aligns with the findings of Stålberg and Fundin (2018), who suggest that developing flexibility and integration principles through communication capacities supports sustainable building.

6. Conclusion

Ghana’s small- and medium-scale construction ability to adapt to ESC was the focus of the study, which attempted to ascertain the influence of integration and adaptation. A twelve-influential factor construct offers a thorough summary of how respondents view the significance of each aspect. The study used a quantitative methodology. The findings help understand the complex facets of integration and adaptation in the unique setting of Ghana’s building sector easier. Furthermore, supporting the validity and dependability of the findings’ relevance in the context of adaptive capacity in the construction sector is the statistical significance of the links amongst the dimensions, which correlates with the central idea of the literature. Thus, innovation advancement and ongoing training and evaluations significantly impact ADI for adaptive capacity, ensuring ESC in Ghana’s small- and medium-scale firms. These findings from the EFA and the SEM results support this assertion. Ghana uses ecologically friendly building techniques that contribute to advancing the global agenda for SDGs, which makes this study novel.

6.1 Practical implications and recommendations

This study is an essential tool in planning to fast-track the effective utilisation of resources for construction activities. Proper planning must be undertaken since small- and medium-scale construction is the main stakeholder in an adaptive capacity for ESC. It can also assist small- and medium-scale construction planners in implementing adaptive capacity. Researchers and practitioners can use these details to tailor interventions, policies or training programmes that address the specific factors deemed most critical by industry professionals. The study implies that the environment can be managed prudently by implementing adaptive capacity. In ecologically sustainable buildings, the number of beneficiaries should be taken into account extensively through integration and adaptation. Although the study focussed solely on the effects of AID on the ability of small- and medium-scale construction in Ghana, significant information is still being revealed. However, the study is not without limitations, and it would be ideal, if resources are allowed to carry out a similar study about the other adaptive ability factors covering Ghana’s entire territory.

Figures

Figure 1

Study methodology

Figure 2

CFA model for adaptability and integration

Table 1

Indicators of adaptability and integration construct

Item	Source	Constructs	Independent variable	Dependent variable
1	McDermot et al. (2020), Pim-Wusu et al. (2022c), Otter (2019), Hamida et al. (2023), Yu et al. (2018)	Literacy level and understanding	Adaptability and Integration	Adaptive capacity
2	Sultan and Alaghbari (2021), Debrah et al. (2023), Chen et al. (2024), Van Ellen et al. (2021), Manewa et al. (2016), Yu et al. (2018)	Social structures
3	Atombo et al. (2015), Debrah et al. (2021), Pim-Wusu et al. (2023), Fiyinfoluwa et al. (2022)	Inner factors/morals
4	Cummings et al. (2015), He et al. (2021), Pim-Wusu et al. (2023), Rashidian et al. (2023)	Views and enthusiasm
5	Pim-Wusu et al. (2022a), Agyekum et al. (2022), Raouf and Al-Ghamdi (2020), Chen et al. (2024)	Organisational goodwill
6	Waidyasekara and Senaratne (2020), Pim-Wusu et al. (2022b), Otter (2019), Hamida et al. (2023)	Active reworking commitment
7	Van Ellen et al. (2021), Taherkhani (2023), Gunatilake and Perera (2018), Rashidian et al. (2023)	Class of individual richness
8	Ebekozien et al. (2023), Atombo et al. (2015), Pim-Wusu et al. (2023), Stålberg and Fundin (2018)	Communication capabilities
9	McDermot et al. (2020), Hamida et al. (2023), Otter (2019), Chen et al. (2024), Manewa et al. (2016)	Innovation advancement
10	Wang et al. (2013), Tang et al. (2020), Dalirazar and Sabzi (2020), Askar et al. (2021)	Knowledge discrimination
11	Stålberg and Fundin (2018), Ranadewa et al. (2021), Agyekum et al. (2022), Rashidian et al. (2023)	Disparities in wealth distribution
12	Fonseca et al. (2020), Atombo et al. (2015), Van Ellen et al. (2021), Othman and Abdelwahab (2018)	Constant training and evaluations
13	Sultan and Alaghbari (2021), Dearing and Cox (2018), Otter (2019), Debrah et al. (2023), Askar et al. (2021)	Involvement in policymaking
14	Cummings et al. (2015), Raouf and Al-Ghamdi (2020), Hamida et al. (2023), Fiyinfoluwa et al. (2022)	Effective interaction

Source(s): Authors’ construct

Table 2

Background information of the respondents

		Freq	%
Position held by respondents	Company Director	67	16.75
	Project Manager	65	16.25
	Site Manager	64	16.0
	Consultant supervisor	32	8.0
	Consultant	49	12.3
	Field Engineer	64	16.0
	Quantity surveyor	28	7.0
	Architect	24	6.0
	Planner	7	1.75
Years of experience	1–5	42	10.5
	6–10	98	24.5
	11–15	97	24.3
	16–20	94	23.5
	20 and above	69	17.3
Highest Educational Qualification	HND/Diploma	12	3.0
	Bachelor’s Degree	144	36.0
	Master’s Degree	201	50.3
	Doctoral Degree	43	10.8
Nature of Company	D4K4 Contractor	36	9.0
	D3K3 Contractor	64	16.0
	D2K2 Contractor	102	25.5
	Consultancy firm	103	25.8
	Public sector firm	95	23.8

Source(s): Fieldwork

Table 3

Adaptability and integration (ADI)

Variable	Mean	Standard deviation	Rank
Innovation advancement	3.85	1.02	1
Constant training and evaluations	3.78	1.03	2
Effective interaction	3.76	0.96	3
Literacy level and understanding	3.75	1.02	4
Active reworking commitment	3.72	1.02	5
Communication capabilities	3.68	0.99	6
Views and enthusiasm	3.67	0.99	7
Knowledge discrimination	3.63	1.09	8
Involvement in policymaking	3.64	1	9
Organisational goodwill	3.57	1.02	10
Class of individual richness	3.53	1.02	11
Social structures	3.48	0.94	12
Inner factors/morals	3.44	0.97	13
Disparities in wealth distribution	3.42	1.04	14

Source(s): Authors’ fieldwork

Table 4

Total variance explained by adaptation and integration construct

Component	Initial eigenvalues			Extraction sums of squared loadings			Rotation sums of squared loadings
	Total	% of variance	Cumulative %	Total	% of variance	Cumulative %	Total	% of variance	Cumulative %
1	2.638	23.840	23.840	2.638	23.840	23.840	1.896	13.54	13.54
2	1.254	20.955	44.795	1.254	20.955	44.795	1.436	10.25	23.80
3	0.998	7.138	51.933
4	0.989	6.842	58.775
5	0.978	6.793	65.565
6	0.952	6.513	72.078
7	0.887	5.434	77.512
8	0.862	4.354	81.866
9	0.810	4.297	86.163
10	0.755	4.286	90.449
11	0.705	3.125	93.574
12	0.687	3.104	96.678
13	0.623	2.207	98.885
14	0.583	1.115	100.000

Note(s): Extraction Method: Principal Component Analysis

Source(s): Fieldwork

Table 5

Unidimensionality and reliability of adaptability and integration construct

	SIV	PNI	Corrected item-total correlation	Squared multiple correlation	Cronbach’s alpha
Organisational goodwill	0.704		1.000	0.500
Inner factors/morals	0.674		0.324	0.328
Views and enthusiasm	0.669		0.429	0.357
Social structures	0.663		0.308	0.263	0.801
Active reworking commitment	0.663		0.500	1.000
Literacy level and understanding	0.644		0.332	0.253
Class of individual richness	0.518		0.240	0.288
Effective interaction		0.744	0.602	0.413
Constant training and evaluations		0.728	0.592	0.423
Communication capabilities		0.721	0.580	0.391
Involvement in policymaking		0.708	0.558	0.400	0.807
Innovation advancement		0.706	0.560	0.425
Knowledge discrimination		0.616	0.488	0.308
Disparities in wealth distribution		0.550	0.425	0.290

Source(s): Authors’ fieldwork

Table 6

Robust fit index for adaptability and integration

Fit index	Cut-off value	Estimate	Comment
S – Bχ2		4.577
Df	0≥	53	Acceptable
CFI	0.90≥ acceptable 0.95≥ good fit	0.973	Good fit
PCFI	Less than 0.80	0.601	Good fit
RMSEA	Less than 0.08	0.075	Acceptable
RMSEA 95% CI	0.00–0.08 “good fit.”	0.063–0.087	Acceptable
NFI	Greater than 0.90 “good fit.”	0.945	Good fit
IFI	Greater than 0.90 “good fit.”	0.974	Good fit
PNFI	Less than 0.80	0.578	Good fit
RMSR	Less than 0.05 “good fit.”	0.039	Good fit
GFI	Greater than 0.90 “good fit.”	0.903	Good fit

Source(s): Authors’ compilation of fit index

Table 7

Final conceptual model indicator variables for adaptability and integration

Latent component	Indicator variable	Measurement variable	Label
System Internally and Vulnerability (SIV)	ADI5	Organisational goodwill	SIV1
	ADI3	Inner factors/morals	SIV2
	ADI4	Views and enthusiasm	SIV3
	ADI2	Social structures	SIV4
	ADI6	Active reworking commitment	SIV5
	ADI1	Literacy level and understanding	SIV6
Perceived Need for Implementation (PNI)	ADI14	Effective interaction	PNI1
	ADI12	Constant training and evaluations	PNI2
	ADI8	Communication capabilities	PNI3
	ADI13	Involvement in policymaking	PNI4
	ADI9	Innovation advancement	PNI5
	ADI10	Knowledge discrimination	PNI6

Source(s): Authors’ compilation of indicator variables

Table 8

Factor loading and p-value of adaptability and integration

Hypothesised relationships (path)	Unstandardised coefficient (λ)	Standardised coefficient (λ)	p-value	R-square	Significant at 5% level
SIV1 ← SIV	1.000	0.643	0.00	0.813	Yes
SIV2 ← SIV	0.887	0.599	0.00	0.859	Yes
SIV3 ← SIV	0.914	0.608	0.00	0.970	Yes
SIV4 ← SIV	0.805	0.561	0.00	0.815	Yes
SIV5 ← SIV	0.916	0.591	0.00	0.949	Yes
SIV6 ← SIV	0.874	0.561	0.00	0.915	Yes
PNI1 ← PNI	1.000	0.702	0.00	0.793	Yes
PNI2 ← PNI	1.031	0.673	0.00	0.853	Yes
PNI3 ← PNI	0.957	0.651	0.00	0.823	Yes
PNI4 ← PNI	0.940	0.635	0.00	0.703	Yes
PNI5 ← PNI	1.006	0.667	0.00	0.844	Yes
PNI6 ← PNI	0.811	0.501	0.00	0.851	Yes
SIV ↔ PNI			0.00	0.714	Yes

Source(s): Authors’ fieldwork analysis

We hereby declare that there is no potential conflict of interest and any support from a third party.