Phil La Duke's Blog

Fresh perspectives on safety and Performance Improvement

What Can The Hawthorne Electric Studies Teach Us About Worker Safety

By Phil La Duke

Last week I spoke at the National Safety Council where the behemoths of safety gather. I saw some really cool new products—work pants with need pads sewn into them, the latest in ceramic cutting technology, and even an amazing device that prevents industrial vehicle and pedestrian accidents—but I also heard a lot of the same old drivel some repackaged but mostly unadulterated snake oil sold in largely the same package, and a fair amount of the same old hackneyed arguments.  There isn’t much new under the sun, at least not in worker safety.

One of the supposed “tried and true” safety tactics is conducting behavior observations.  Behavior observations lie at the heart of many safety management systems. Exactly how these practices are performed can vary widely from organization to organization and the efficacy of these practices similarly varies from location to location.  I’ve remained largely silent (well as silent as I can ever be) on the practice, because I know many smart safety professionals for whom I have the utmost respect who value behavior observations as important components of their overall safety tactics and strategy.

The thinking that drives behavior observations is that a supervisor (or in some cases, another worker or a safety professional) watches a worker do his or her job after which the observer offers tips on how to do the work more safely. I’m over simplifying, but not a lot.  Proponents of the behavior observation believe that the combination of intervention in cases of unsafe behavior and positive feedback for safe behavior reduce injuries. Those that support Behavior Based Safety proudly point to their love of scientific study and organizational psychology but in doing so they ignore one of the most important studies of the workplace in history, those at the Hawthorne Western Electric Company (and other research conducted by conducted by Fredrick Taylor and others.) For five years, researchers studied the effects of physical, social, psychological, and environmental factors would influence the productivity[1] of workers.  The most famous finding was the dubbed the Hawthorne effect — which referred to an increase in worker productivity produced by the psychological stimulus of being made the focus of the study. One could easily extrapolate that workers too can be temporarily manipulated into working safely, but the results are neither lasting nor indicative of a lasting behavior change. If the Hawthorne Effect is true of safety, than it doesn’t matter whether the feedback is positive or negative, skilled or unskilled, well articulated or grunted out by a bonobo, the behavior of the worker will temporarily improve.

But there were other findings as well, researchers found lesser known phenomena, like the fact that research in and of itself alter the behaviors that are studied.  This phenomenon has been replicated many times in many other studies.

But perhaps more germane to safety, researchers concluded:

  • Productivity is a group activity. The Hawthorn researchers found that the relationship between the supervisors and the workers played an important role in workplace productivity. It should surprise no one that workers with good relationships with their supervisors will tend to report hazards more frequently.  If the worker believes his or her supervisors care about their safety they are far more inclined to bring safety concerns to the attention of leadership.
  • Team Norms Directly Influence Worker Productivity. Researchers have known for a hundred years that workers set the expectations for fair day’s work; but Hawthorn researchers were the first to demonstrate and describe this phenomenon. Similarly, work groups set the expectations of safety and safe work practices. This is the essential core of a corporate culture—that the work group set the rules, even those associated  with worker safety.
  • Worker skill is a poor predictors of his or her job performance. The Hawthorne study found that while worker capabilities provide some clues as to the future performance of a worker (in terms  of the physical and mental potential of the worker), exactly how well the worker will perform (again this applies to safety) the real performance is strongly influenced by social factors; it’s less about whether or not the worker is observed, and more about how the worker interacts with his or her peers.
  • The workplace is a social system. Fredrick Taylor and his colleagues viewed, the work place as a social construct; a system composed of many unpredictable and interdependent elements.

Beyond the Hawthorne Effect, proponents of behavior observations also ignore several key truths:

  • Workers behave differently when they are being studied. From the Hawthorne Electric studies in workplace productivity to studies with animals, researchers now know that the behavior of a research subject is significantly changed simply by the act of observing them.  The tale, it would seem, is tainted in the telling.
  • Observations are essentially shoddy training needs assessments.  In those cases where the worker is acting unsafely (or more likely less safely) the result is either that the worker is doing so because he or she needs to be trained in the correct procedures, or the worker has made an error.  Since we know that human error cannot be prevented through behavior observations, the act of observing workers is akin to doing a slip shod training needs assessment.

 Observations are expensive, pointless, and provide little information that could not be gathered more effectively through another method.  It is an overblown, quasi-scientific reaction to a problem that can be readily addressed through an easier and cheaper approach.

[1] It’s important to note, that while the Hawthorne researchers were studying productivity, it’s not that far from safety.  Safety is the product of a robust and efficient process; that is to say, a process cannot be considered efficient or productive if it produces poor quality or injures workers.  For the purposes of this post, productivity = safety.

Filed under: Safety, , , ,

7 Responses

  1. A great post, I read about The “ Hawthorne effect” in 1991, I went through different stages of what I understood about its findings, that workers knew they were being studied;so acted and performed differently. I came to many conclusions because it served my purposes and beliefs. I like you see these safety behaviourist’s at meetings, talking of the “second mind” when I wonder some of them have one.
    But I do consider that the workers performed both in productivity and safety performance to a higher degree because they were involved,considered and informed on issues they otherwise would not have been. Many talk the talk, whilst they preach and dictate one thing, they do absolutely everything that contradicts the same thing they profess in another area, hypocrites . Thanks for your alter ego style that bucks the sycophantic remarks and teachings of others.


    • Phil La Duke says:

      Thanks Raymond, both for reading and your comments. I agree that involved workers are safer workers. In fact, I think worker engagement is the key to safety.
      When it comes to safety, knowledge is power, good training and a good relationship with both management and co-workers are also key.


  2. Jack Benton says:

    Reblogged this on EHS Safety News America and commented:
    Great post and excellent read by Phil La Duke!


  3. Jim Tassell says:


    I’m struck by the similarity between observation, in the way you describe it here, and the process of accident investigation. In the latter we try to look below the superficial at underlying issues, admittedly sometimes more successfully than others. We try very hard to avoid the superficial “silly so and so, he/she should have known better” sort of “operator error” response. But yet in observation that is exactly the level at which one is working. Or have I missed something?


    • Phil La Duke says:

      I think that when done properly, a holistic observation of workplace hazards exactly parallels a good accident investigation. There are two basic differences: context and timing. We should be observing the process and identifying the system and behavioral issues BEFORE there is an injury (while obviously conducting incident investigation in response to an incident.) In the second difference, context, we should view observations of the workplace in a much broader context the we would in an incident investigation.

      Furthermore, I want to be clear, I am not advocating Behavioral Observations, which I believe are largely a waste of time and resources. Rather an overall, walking FMEA approach that identifies, contains, and corrects the multiple causes of hazards (and thus injuries). One should never land on “operator error” as a root cause, rather one should consider, isolate, and manage all the performance shaping elements of a work environment that increases the risk of injuries. In other words, we can’t stop until we identify all the factors that made human error more probable or that increase the most likely severity of any injuries that should occur.

      I hope that clears it up a bit. Thanks for reading and thanks for your comments.



  4. wish more had entered into this debate, very few in positions to make real changes, actually do so, they follow trends of compiling data/figures and stats , yet really do nothing proactive to make REAL change


    • Phil La Duke says:

      It’s a shame really, because there’s great information that if interpreted incorrectly can tell a us a great deal about our processes. Unfortunately, many of the people who gather data do so inappropriately or just for its own sake. We don’t need data for data’s sake


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s



Guest blogs

La Duke in the News


Press Release

Professional Organizations


Safety Professional's Resource Room

Social Networking


Web Resource


Get every new post delivered to your Inbox.

Join 1,110 other followers

%d bloggers like this: