"background": "The adoption of advanced manufacturing systems in developing economies is critical for industrial growth, yet robust methodological frameworks for analysing adoption rates are lacking. Existing studies often fail to account for the hierarchical structure of plant-level data, where operational units are nested within firms, leading to potentially biased inferences. ", "purpose and objectives": "This case study presents a methodological evaluation of applying multilevel regression modelling to measure the adoption rates of manufacturing systems. Its objective is to demonstrate the model's utility in capturing variance at both the plant and firm levels, providing a more accurate analytical tool for engineering management research in an African context. ", "methodology": "A longitudinal dataset from a sample of manufacturing plants was analysed using a two-level hierarchical linear model. The core statistical model is specified as y{ij = \0j + \1x1ij + eij, where \0j = \00 + \01z1j + u0j. Here, i denotes plants and j firms. Robust standard errors were calculated to ensure inference reliability. ", "findings": "The analysis reveals that firm-level resources, such as access to technical training, explain approximately 40% of the firm variance in adoption rates. A significant positive relationship was found at the plant level between workforce digital literacy and the likelihood of system adoption (p < 0. 01, 95% CI 0. 15, 0. 42). ", "conclusion": "Multilevel regression is a statistically sound and practically valuable methodology for analysing the nested data structures inherent in industrial adoption studies. It successfully disentangles the influence of factors operating at different organisational tiers. ", "recommendations": "Researchers and policymakers should employ multilevel modelling for similar industrial analyses to avoid ecological fallacies and aggregation bias. Future data collection initiatives should be designed to capture variables at both the plant and parent-firm levels. ", "key words": "Hierarchical linear
Uwimana et al. (Sat,) studied this question.