Skip to content

Ellen Langer: expert on, and victim of, the illusion of control

It all started when Lee Sechrest pointed me to this post by James Coyne. Sechrest wrote:

I know you have enough to do, and if you do not get to this…well, no problems. It is a blog by Jim Coyne taking apart a “classic” study in social psychology, originally published in the early ’70s.

Implausible effects on mortality of giving plants to old people to take care of. The study became the foundation for a whole movement of giving “control” to nursing home patients and other such. The study has been cited nearly 500 times. Except that a later erratum revealed a statistical error that rendered the results not significant. The erratum has only been cited 6 times. The study was never withdrawn.

The second author of the study went on to become the first tenured female professor at Harvard and she earned many (justified) awards, fellowships, and so on for her work on mindfulness.

The first author, not so often mentioned in connection with the plant study, got on the fast track in administration and became Provost at Yale, President at U Pennsylvania, and is now President of the Rockefeller Foundation.

Oak trees from little acorns grow.

This stunned me, not in a good way. I associate Ellen Langer with her wonderful article, The Illusion of Control, in the classic Kahneman, Slovic, and Tversky book. So it’s a bit disappointing to see this.

A few days later Sechrest followed up with this:

Sadly, the story gets worse. Sort of hard to believe, but Coyne is quite careful. Lots of people hate him, but they don’t do so based on his mistakes. The Coan he refers to in his blog is one of our students and a good friend. When I read Coyne’s reference to his work, my thought was that he can’t be right. Then I read the whole blog and some articles: he’s right.

Here’s what Coyne wrote:

When an Ivy League psychologist promotes quackery, is it time to stop accepting her ‘scientist’ branding?

Following up on my previous dissection of her nursing home study, I [Coyne] take a skeptical look at what passes for science in the intense promotion of Ellen Langer’s planned study attempting to shrink tumors of metastatic breast cancer patients at a Mexican resort.

Ellen Langer’s identification as an eminent, well-published Harvard psychologist is an important part of her branding and the promotion of herself and her products. The promotion is infused with references to her 40 years of research. Yet, she assumes none of the responsibility that goes with being a scientist. She does not consistently submit her work to peer review. She makes references to unpublished studies, even those that have remained so for many years. Some of her studies are described only in an ostensibly peer reviewed journal, Perspectives on Psychological Science, but with insufficient details to allow any independent evaluation of her claims. There is evidence of deliberate, selective publication in her direct quotes in the New York Times article. She talks about ongoing studies in ways that suggest biases being introduced by her monitoring incoming data. She is flippant in presenting her theoretical model and the sources of her hypotheses. There are discrepancies between claims that she makes to the media and what is available in published accounts of her research. And finally, she is dismissive of the basic responsibilities of a scientist conducting biomedical research to justify the work with reference to plausible mechanism and to provide patients with an accurate sense of the evidence base supporting or not supporting treatments.

Langer has published in scientific journals, but she is not otherwise acting like a scientist. One of her studies involved telling hotel chambermaids that their work involved lots of exercise, with the result that the women lost weight without increasing their actual exercise. These implausible results are reported in the peer-reviewed journal Psychological Science, with Langer explaining results with the same sort of mental frame explanation. If I am skeptical, do I owe it to her to carefully examine this article before dismissing it? I don’t think so. Her other activities establish sufficient prior probabilities it will not be worth the effort.

There is a way in which, once a researcher is eminent enough, he or she becomes a “brand.” Is this good or bad, I don’t know. Linus Pauling, Roger Penrose, Jared Diamond, that Bell Curve guy.

To get back to Langer, though: this is just sad. She was famous (to me) for her 1975 article on the illusion of control, “defined as an expectancy of a personal success probability inappropriately higher than the objective probability would warrant.” And then, in her own research, she herself becomes subject to the inclusion of control.

I’d like to believe that, if someone were to just talk with Langer and explain what’s going on, she’d say, something like: “Wow—you’re right! I totally fell into that trap, the illusion of control. Damn!” and it wouldn’t be too late for her to rectify things. Realistically, though, nahhhh, I can’t imagine it will happen. Presumably she’s had lots of chances already to see what was going on, and if she hasn’t taken the opportunity so far, she’s probably not going to do it now. I can only assume that she “knows” that her scientific hypotheses are true, so the strength of the evidence would seem to her to be nothing but a minor technicality.

P.S. More here.


  1. Corey says:

    Now you see that evil will always triumph, because good is dumb.

  2. Steen says:

    Match-up for the next seminar speaker bracket: Gauss (1) vs. Charles A. Murray. Gauss in a walk, later eliminated by Linus (1; Torvalds or Pauling) in the final four.

    “She says she also vaguely recalls switching from the negative to the positive after someone suggested that in psychology people studied their own worries. ”I didn’t want to be thought of as mindless,” she jokes, ”so I changed the name of what I was studying from mindlessness to mindfulness.”

  3. Keith O'Rourke says:

    Maybe a quick visible success for this new incumbent at the Rockefeller Foundation?

  4. Steve Sailer says:

    I blogged about Ellen Langer here:

    “Here’s an NYT Magazine article from last month on a Harvard psychology professor named Ellen Langer who is somewhere on the genius-crank continuum.

  5. zbicyclist says:

    For counterpoint, we have this research, from the esteemed Huffington Post:

    Ironically, the author of this study purporting to show that pessimists live longer is Lang.

    (I lost track of how many people sent me this article, often with the notation “You’ll live forever!”)

  6. mschulte says:

    Some early attempts to replicate the ‘Langer effect’ in risky choice.

    We talked about replicability a while ago :)

  7. Rahul says:

    Is this an example of tenure allowing you to do whatever wacky stuff you want to without risk of censure?

    What safeguards does the University system have against crackpot ideas being disseminated under it’s name?

  8. This reminds me of Amy Cuddy’s Ted talk (slogan: “Fake it till you make it”). There is a lovely graphic starting around 11:59 with “high power” people and “low power” people’s testosterone levels (pg/ml; is that a trillionth of a gram per ml?) shown in a barplot (nope, no estimates of uncertainty, that would only distract us).

    If I recall correctly, you have to wave your hands around wildly above your head to increase your testosterone levels, so just before an interview, all you have to do is go to the bathroom and do this thing she says, and you’re all set: low power gears up to high power.

    This person is also at Harvard.

    So what’s with Harvard? First that IQ guy, then these people? It’s almost like a parody. I’m sure there are really great people there, but how come nobody is telling them to cut it out?

  9. sbk says:

    When I was in graduate school at Harvard, 3 of my fellow students were working on Langer’s age reversal study among elderly (i.e., provide an environment containing artifacts from @ 40 years past and those older folks within will show physical signs of age reversal).

    The students involved were busy buying such things as old magazines from local bookstores, etc. with which to populate the location the older folks would temporarily be housed in.

    According to the students, they were asked to measure dozens of age-relevant outcomes (e.g., rate of finger nail growth) and among the many assessed, found evidence of a small subset consistent with the age-reversal hypothesis. The positive outcomes were reported and walla — the hypothesis was “confirmed”. (Disclaimer — this information was related to me by the graduate students. I personally did not witness any of it (save for buying old magazines) first hand. And I did not read the article).

    What is upsetting to me is not just the experimental approach reported by the students. It is, to me, that this research was published by a psychological journal. What exactly is the theory being tested? That mindset can reverse physical aging? How does this occur? Etc.? But, why go on? I see no coherent theory being tested.

    This, again as I see it, is the sad state of contemporary psychology and its infatuation with the rewards that accrue from media accolades. The search for understanding would seem to take a second (or third, or fourth) seat. In this light, replication failure issues are far from our biggest problem facing he discipline.

Leave a Reply