The Hawthorne effect is the idea that people perform better when they’re being studied. The name comes from studies conducted at Western Electric’s Hawthorne Works facility. Increased lighting improved productivity in the plant. Later, lowering the lighting also increased productivity. The Hawthorne effect says that the productivity increase wasn’t due to changes in lighting per se but either the variety of changing something about the plant or the attention that workers got by being measured, a sort of placebo effect.
The Alternative Blog has a post this morning entitled Hawthorne effect debunked. The original Hawthorne effect was apparently due to a flaw in the study design; correcting for that flaw eliminates the effect.
The term “debunked” in the post title may imply too much. The effect in the original studies may have been debunked, but that does not necessarily mean there is no Hawthorne effect. Perhaps there are good examples of the Hawthorne effect elsewhere. On the other hand, I expect closer examination of the data could debunk other reported instances of the Hawthorne effect as well.
The Hawthorne effect makes sense. It has been ingrained in pop culture. I heard a reference to it on a podcast just this morning before reading the blog post mentioned above. Everyone knows it’s true. And maybe it is. But at a minimum, there is at least one example suggesting the effect is not as wide-spread as previously thought.
It would be interesting to track the popularity of the Hawthorne effect in scholarly literature and in pop culture. If the effect becomes less credible in scholarly circles, will it also become less credible in pop culture? And if so, how quickly will pop culture respond?


{ 4 comments… read them below or add one }
Jason Dyer 06.16.09 at 12:56
This educational study saw the effect:
http://numberwarrior.wordpress.com/2008/02/13/curious-results-in-india-merit-pay-study/
John S. 06.16.09 at 15:44
I always considered the key feature of the Hawthorn effect to be the totally counterintuitive (and now shown to be erroneous) conclusion that productivity increased when the lighting levels were lowered. That’s what Levitt and List seem to have debunked. In the Indian schools study mentioned in the previous comment, I was expecting to read that the measures of success improved after teachers’ salaries were lowered, similar to the way those measures improved in other schools after teachers’ salaries were increased. But that’s not what was shown, so I don’t think the study validates the Hawthorn effect.
Gene Harris 06.16.09 at 19:20
I wonder if the Hawthorn effect kicks in if your company has announced layoffs or a reduction in force?
Jason Dyer 06.17.09 at 17:10
Hmm, I guess I was interpreting the definition differently then.
In the case of the educational study the observational effect had arguably a greater affect than the money; it’s just the money didn’t have a negative impact.