Answers are grounded in this paper and cite their sources.
Evaluating knowledge retention over time is critical for assessing the long-term effectiveness of any training program. This is particularly relevant in the construction industry, where workers must not only acquire but also retain and consistently apply safety training knowledge to remain safe while at work. Despite its importance, there is limited guidance on how knowledge retention should be evaluated in construction, and most previous studies have relied primarily on quantitative assessments. This paper demonstrates the application of a concurrent triangulation mixed-methods approach for evaluating knowledge retention through a field-based case study involving five delivery methods: pre-recorded video, lecture, interactive lecture, flipped lecture, and interactive lecture with hands-on activities. Of approximately 600 workers who participated in the initial training, 33 completed the one-month follow-up assessment and interview, providing the data for this illustrative case study. Results indicated significant one-month improvements in hazard recognition for the pre-recorded video and interactive lecture with hands-on activity groups, and significant improvements in high-energy hazard recognition for the pre-recorded video, lecture, flipped lecture, and interactive lecture with hands-on activity groups. Although the magnitude of general hazard recognition change differed across delivery methods overall, adjusted pairwise comparisons did not identify statistically significant differences between specific training groups. Together, the findings illustrate how integrating quantitative and qualitative evidence produces richer interpretation of retention rather than either approach alone, while providing practical considerations for researchers designing future retention studies.
No takes yet. Share an insight, caveat, or question.
Raeisinafchi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: