Safety Culture in the Spanish Nuclear Power Plants through the Prism of High Reliability Organization, Resilience and Conflicting Objectives Theories
Round 1
Reviewer 1 Report
Review of applisci-1026438
Safety Culture in the Spanish Nuclear Power Plants through the Prism of High-Reliability Organization, Resilience, and Conflicting Objectives Theories
This paper investigates safety culture in nuclear power plants through the use of surveys distributed to a very large sample in nuclear power plants in Spain.
I have very mixed thoughts and feelings about this paper.
Introduction:
On one hand, I really liked the review of the theoretical perspectives of safety and safety culture presented in the introduction (learned some things and I actually know this literature relatively well) as well as the large sample surveyed. Conversely, I was extremely disappointed in the following:
- there was no synthesis of the different theoretical perspectives nor was there any position taken regarding what attributes of each perspective the researchers/authors were going to use.
- The researchers say, “taking as analysis framework the traits of the three theoretical approaches on the safety of high-risk organization…” but they do not really put together any kind of a framework and more importantly,
- they do not connect their methods or measures with their goals and the theoretical perspectives—at all.
- At the end of the introduction, I felt like I was left hanging… how is what you all did going to make a difference, why do we need to know this, how do the measures related to HRO, resilience, conflicting objective perspectives, or safety culture? I guess generally what’s the overall point of this study.
- and then to add insult to injury, after this beautiful summary of these important theoretical perspectives, the researchers give a definition of safety culture from an online pamphlet?! There are probably hundreds of referred references that have definitions of safety culture. Better to have one of those.
Methods (formatting note, measurably more readable for text to be left-justified in tables):
As I mentioned before, it is excellent that you had such a large sample size and that you identified types of facilities and types of workers.
- needed to describe why you identified types of facilities and workers. Why is this level of information important to know? I can guess but you need to say it. Why is the total N different for Location and organization (also for Contract)?
- with regard to your measures - no idea what these have to do with the theories you reviewed or the supposed framework you’re investigating
- I do not understand at all what is going on with the BARS. What is a “graded behavioral scenario” and what is a “scale point?” The table in the results show “high” “med” “low” but then it is described that workers consider that the organization highly prioritizes safety in favor of procedures through the attention to safety measures. How can you get that from a High, medium, or low response?
Results
I LOVE that the effect size is provided but am confused at how inconsistently the significance is provided. Maybe this is a different way of giving results but it is not even consistent in the sections of the result section. However, you say you did ANOVAs so typically this is presented as F(dfb, dfw) = Fscore, p = pvalue, eta squared = effect size. Or at least report the p’s.
- there is no analysis with regard to if any of the variables are different from each other in the first section
- no error bars in the bar graph
- no labels for red, yellow, green in BARS figures
- VERY large N differences for BARS and scales. Why is this? On page 7, you mention “representative sample,” is this the reason for the big difference? How was the sample chosen? Why not the entire sample from the facility?
- No in-text reference for Figure 2
Discussion
- Because you had not set up anything with regard to how your measures were going to tell us anything new, the whole thing after the end of the introduction just seemed like a mess to me honestly.
I know this is harsh and I am not trying to be condescending but it was frustrating because I was disappointed. There is clearly a very very good paper in here somewhere. You just have to explicitly articulate the constructs from the theories in which you’re interested (maybe how they’re related and how that hasn’t been explored before); explain what we need to know about those constructs (are you moving theory forward; are you making nuclear power safer; etc), tell us how these measures have been effectively used to assess those constructs in the past (or maybe that they haven’t been but should have been or could have been); tell us what to expect if you’re right; show us what you found; reflect back on what you thought. What I’m offering in this paragraph is just guidance. You do not need to address or respond to each point. I’m hoping though it might better elucidate some of the things that could help you find the good paper that is hiding in the current one somewhere.
Author Response
Please see attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper presents a very interesting research topic, that is, to link safety culture, high reliability organization, resilience engineering, and conflicting objectives. But there are three key parts missing:
- there is no background information about why we need to conduct this research. For example, is it because the accident rates of Spanish nuclear power plants have increased?
- about the research methodology, a chart/figure is recommended to show the relationship between the four theories proposed in the beginning and the survey scales (i.e. Safety, Risk Perception, Safety Conscious Work Environment, and Organizational Resilience) and BARS scales. The reviewer would also like to see all the survey questions as an appendix in the end.
- Did the authors use factor analysis? Factor analysis is important for safety culture research.
Author Response
Please see attachment
Author Response File: Author Response.pdf
