Background

In order to optimize cost-efficiency when collecting posture data in field studies, accurate cost data are needed. A few studies have assessed costs of different measurement methods, but they did not address the uncertainty of the cost estimates. Information on the uncertainty of cost estimates is key input when deciding which method to use.

Aim

This study aimed atassessingthe uncertainty in estimates of costs for collecting posture data by inclinometry, observations and self-report.

Method

The study wasbased on data collected at a Swedish paper mill (Heiden et al. 2017)¹. Using a model developed by Trask et al. (2014)², costs were calculatedfor measuring trunk and upperarm postures of twenty-eight workers during three full shifts using inclinometers, observations from recorded videos, and workers’ self-reports from a questionnaire. For each measurement method, the uncertainty of the actually observed total cost was assessed by determining the range of costs between an assumed best case (lowest cost) and worst case (highest cost) using scenario analysis.

Results

Observation was the most expensive method (€41499) and also showed a large uncertainty in the cost estimate (€19089–€87154). Self-reports had the lowest cost (€9156) with the smallest uncertainty (€3941–€27473). The overall cost for inclinometry was €16851 with best and worst cases €8567 and €60313, respectively. The actual costs of inclinometry and self-reports in the conducted study were reasonably close to the best case.

Discussion

In their study of flight baggage handling, Trask et al. (2014) concluded that inclinometry was more expensive than observation when measuring trunk and arm postures. In the present study, we found observation to be the more expensive. Given the uncertainty in cost estimates, both results are plausible. Notably, in Trask et al. (2014), video recordings were restricted to 4 hours, whereas in the present material, video recordings covered full 7 -to 12-hour shifts. The time allocated to collecting video material for observations will significantly affect the cost comparison of methods, and this may serve as an example that further investigations of separate cost components for each method are warranted.

Conclusions

Based on a meticulous cost assessment, we found that cost estimates for observation of working postures were the most uncertain, followed by inclinometry and self-reports. A better understanding of data collection costs and their uncertainty, and thus of how to properly identifyan optimal measurement method, requires a deeper analysis of the cost model, and of the contributions of separate cost components to the overall cost

¹ https://doi.org/10.1093/annweh/wxw0262

² https://doi.org/10.5271/sjweh.3416