Safety compliance and construction waste reduction in Ghana with implications for sub-Saharan Africa: the role of deep and surface practices

Purpose Despite growing recognition that worker behaviour drives construction waste generation, no study has empirically tested whether disaggregated safety compliance dimensions directly predict waste reduction outcomes, particularly in Sub-Saharan Africa, with evidence from the Ghanaian construction environment. This study aims to examine the impact of deep safety compliance (DSC) and surface safety compliance (SSC) on construction waste reduction (CWR) in Ghana's building sector, addressing an empirically untested relationship within Ghana and a similar Sub-Saharan Africa (SSA) context as a broader scope. Design/methodology/approach A positivist, quantitative research design was adopted. Data were collected from 101 valid responses from Ghanaian construction professionals through a structured questionnaire using a five-point Likert scale. All three constructs, namely DSC, SSC, and CWR, were measured through respondents' self-reported perceptions rather than objective site-level waste audit data. Partial Least Squares Structural Equation Modelling (PLS-SEM) via SmartPLS 4 was employed to test two hypotheses and evaluate explanatory and predictive model performance. Findings Both DSC (ß = 0.648, p < 0.001) and SSC (ß = 0.186, p = 0.012) positively and significantly influenced CWR. DSC demonstrated substantially stronger effects, with the model explaining 59% of CWR variance. DSC presents a theoretically plausible pathway through intrinsic commitment, proactive hazard anticipation, and adaptive decision-making, while SSC contributes through procedural discipline that limits accident-induced material damage. Originality/value This study provides the first empirical evidence linking disaggregated safety compliance dimensions to construction waste reduction in SSA. It extends High Reliability Organisation and Socio-Technical Systems theories into the waste management domain, offering a dual-benefit framework for resource-constrained construction environments in Ghana and similar SSA contexts. The findings are based on perceptual data from Ghanaian construction professionals and offer context-specific empirical evidence with broader implications for similar developing-country settings.