Carvalho F, Melo R B
Work. 2015;51(3):591-600. doi: 10.3233/WOR-141878.
In many enterprises the semi-quantitative approach turns out to be the available and most suitable technique to perform a risk assessment. Despite its advantages, we cannot disregard the existing gap in terms of validation of this type of applications.
This paper reports a study about risk assessments' reliability, namely both inter-coder (reproducibility) and intra-coder (stability) reliability of the semi-quantitative approach.
This study comprised 4 fundamental stages. Data collection relied on free and systematized observations and made use of video recording, documental research, analysis grids and questionnaires specifically developed for this purpose. A set of different analysts were asked to use four semi-quantitative risk assessment methods (in two different moments) to estimate and assess six risks identified in two tasks accomplished to produce Airbags. The Krippendorff's Alpha Coefficient (α K) was the agreement measure selected to evaluate both inter-coder and intra-coder consensus.
The preliminary results revealed a general low concordance (α K < 0.6) for both reliability evaluations. Apparently there are no relevant differences between the risk assessment results obtained by individuals with different levels of experience or expertise.
This study revealed that the use of the semi-quantitative approach should be done with caution.
在许多企业中，半定量方法是进行风险评估可用且最合适的技术。尽管它有优点，但我们不能忽视这类应用在验证方面存在的差距。
本文报告一项关于风险评估可靠性的研究，即半定量方法的编码员间（再现性）和编码员内（稳定性）可靠性。
本研究包括4个基本阶段。数据收集依靠自由和系统的观察，并利用录像、文献研究、分析网格以及为此专门开发的问卷。一组不同的分析人员被要求使用四种半定量风险评估方法（在两个不同时刻）来估计和评估在生产安全气囊的两项任务中识别出的六种风险。选择克里彭多夫阿尔法系数（αK）作为一致性度量来评估编码员间和编码员内的一致性。
初步结果显示，两种可靠性评估的总体一致性较低（αK < 0.6）。显然，不同经验或专业水平的个人所获得的风险评估结果之间没有显著差异。
本研究表明，应谨慎使用半定量方法。
