Sorry… we were wrong! By Nick McRoberts

by Didier Marlier on Wednesday November 6th, 2013

One of the most difficult things that we all face at some point is having to admit that we were wrong. I don’t mean minor mistakes like predicting the weekend weather incorrectly that we can shrug off by mumbling “errare humanum est”. I’m referring to the difficult, embarrassing trial of laying aside something we believed in, supported and maybe even preached, when the evidence finally convinces us beyond a doubt that we were wrong.

Often our first reaction is to glance around to see if someone has noticed. Did someone see me run that red light? Has anybody noticed the printer that’s “not working” is unplugged? And if we “got away with it”, do we really have to admit it?

What if it turns out that our mistake was visible and we are just plain “wrong” – no justifying it after the fact, no putting a spin on the interpretation, no blaming someone else? What do we do then? Say nothing and hope it will go away? Silently change our stance?

The capacity to recognise and act on the knowledge that we are wrong, and that we have misled people that trusted us, is critical in science, in business, in relationships and even in art.

But it’s a lot rarer than it should be. And while we are very good as a species at noticing other people’s mistakes, we don’t seem to be as comfortable with our own, or putting in place the corrective measures that are necessary.

It’s the other half of that well known quote in Latin that should give us pause:

“Errare humanum est, perseverare diabolicum“. To err is human, but to persevere is diabolical.

It seems that we often prefer to cling to a sinking hypothesis rather than call for help or try to swim for shore. “The captain must go down with his ship!” This kind of stubbornness is endemic in academia and business, in religion and music.

This week, the difficult decision of what to do in that kind of situation has come to our door at Enablers.

Many of you will have heard us speak with passion of the work of Losada, Heaphy and Fredrickson who made great waves in the field of psychology beginning in 2004. They proposed that there needed to be a specific and quantifiable ratio of positive to negative behaviours in order for a team to flourish. They performed a number of experiments to test their hypotheses and their work was published in the American Behavioural Scientist in 2004 (1) and the American Psychologist in 2005 (2) .

This week, Liz a psychologist colleague of ours, drew our attention to a strongly worded critique published in July 2013 in the American Psychologist (3) . Nick Brown, a graduate student at the time, had read the papers and felt that something was “off” in the statistics and mathematics. He shared his concerns with Alan Sokal, a professor of physics at New York University. Sokal made a name for himself as a critic of poor academic standards in the 1996 “Sokal affair” in which he sent a scientific article “liberally salted with nonsense” to a leading journal to see if they would publish it. (They did). Together with Harris Friedman, a professor of psychology at Florida University, Brown and Sokal went through the papers in detail, concluding:
“We examine critically the claims made by Fredrickson and Losada (2005) concerning the construct known as the “positivity ratio.” We find no theoretical or empirical justification for the use of differential equations drawn from fluid dynamics, a subfield of physics, to describe changes in human emotions over time; furthermore, we demonstrate that the purported application of these equations contains numerous fundamental conceptual and mathematical errors. The lack of relevance of these equations and their incorrect application lead us to conclude that Fredrickson and Losada’s claim to have demonstrated the existence of a critical minimum positivity ratio of 2.9013 is entirely unfounded. More generally, we urge future researchers to exercise caution in the use of advanced mathematical tools, such as nonlinear dynamics, and in particular to verify that the elementary conditions for their valid application have been met”.

Their examination has led to a published partial retraction of the paper with Fredrickson distancing herself from Losada’s work (he was largely responsible for the mathematics).

Didier contacted Losada who stands by his work but does not want to enter the “academic debate”.

This leaves us, the Enablers community, with a decision to make: how to deal with the fact that the academic basis for an important part of our practice has been publicly called into question?

After consulting all our partners, the consensus is that, stripped of the mathematics, the key concepts put forward by Losada, Heaphy and Fedrickson have been incredibly useful to us in coaching teams and leaders. We remain convinced that adopting the “right” behaviours in the workplace is of crucial importance.

So how do we avoid “throwing out the baby with the bath water” while remaining honest and transparent?

First the transparency: For those of us inclined to use impressive sounding phrases like “conclusively proved” or “scientifically validated”, it is a welcome reminder that dogmatic positions should not exist in science and must be avoided, perhaps especially in the “soft” sciences.

The same issues exist however in physics, mathematics, medicine and all the other “hard” sciences. Newton’s mechanics (4) are very useful in predicting how everyday objects will behave, but give the wrong answers if they are very big, very small or moving very fast. Not all of Einstein’s ideas (5) turned out to be correct. Bohr’s model (6) of the atom is a useful teaching aid despite being totally “wrong” from a quantum point of view.

And even in the arts: to take just one example, Beethoven wrote a few really bad pieces of music (7).

Finally, it is important to recognise that Brown et al do not “disprove” (or even challenge) the most basic hypothesis underlying Losada, Heaphy and Fredrickson’s work, namely that positive attitudes lead to better outcomes. What they do challenge is the methods used to set a numerical value on the ratio of positive to negative behaviours. Does that sound like a subtle difference? Maybe, but it’s important. Losada may be right about his ratios. However, his mathematics and the papers do not prove it. The papers are flawed, perhaps fatally, but poorly designed research does not mean that the hypotheses are untrue, just that they have not been properly tested.

Our collective position is that we have seen the powerful results that attention to behaviour can have for teams and we will therefore continue to test and employ Losada and Fredrickson’s concepts with our clients. We will however present it as an interesting concept that has strong empirical and anecdotal evidence, rather than as something that has been conclusively proved. Should any new work emerge to challenge that view, we will read it attentively and adjust our views and practice once again.

So what for leaders? I suspect we are all wrong a lot more often than we’d like to admit, and more worrying still, probably more often than we know. What risk do we run if we don’t deal with this? Dozens of business examples spring to mind where the inability to modify or evolve a dogmatic position has led to the demise of a product or company. Bad news doesn’t go away if you ignore it. So how can we create a climate where people feel safe to tell us when they think we’re wrong? How do we stimulate critical and lateral thinking so that multiple perspectives can co-exist or confront each other? How do we prevent the self-centred laziness of “three hundred academic citations can’t all be wrong”, or “we have X% of market share – we can’t be wrong”? Creating conditions for feedback, from your peers, your subordinates and your clients is critical to survival.

We’re all wrong from time to time. The challenge is to recognise it, admit it then do something about it.

Next week we’ll take a look at the misaligned incentives that are the root cause of many of these errors, both in the scientific community and in business.

(Written by N. McRoberts, co-edited by M. Faure, D. Marlier, M. Newman & S. Okunowo)

(1) The Role of Positivity and Connectivity in the Performance of Business Teams: A Nonlinear dynamics Model. Marcial F. Losada and Emily Heaphy. American Behavioural Scientist 2004; 47; 740

(2) Positive Affect and the Complex Dynamics of Human Flourishing. Barbara L. Fredrickson and Marcial F. Losada. American Psychologist, October 2005; Vol 60 No 7

(3) The Complex Dynamics of Wishful Thinking: The Critical Positivity Ratio. Brown, Nicholas J. L.; Sokal, Alan D.; Friedman, Harris L. American Psychologist, Jul 15 , 2013, No Pagination Specified.

(4) C Hellingman (1992). “Newton’s third law revisited”. Phys. Educ. 27 (2): 112–115

(5) Einstein against quantum mechanics: randomness, ignorance and our ignorance about randomness, Juan Pablo Paz, El universo de Einstein: 1905 – annus mirabilis – 2005, Alejandro Gangui (ed.), Editorial EUDEBA, 2007

(6) Helge Kragh (2011). “Conceptual objections to the Bohr atomic theory — do electrons have a “free will” ?”. European Physical Journal H 36 (3): 327

(7) [1] Wellingtons Sieg Op. 91.  Nicholas Cook describes it as, ”[t]here is indeed a rare unanimity of critical judgment [here]: for Maynard Solomon the works [dating from, or shortly before, the Congress of Vienna] are ‘the nadir of Beethoven’s artistic career’, and for William Kinderman the ‘nadir of his artistic achievement’” 19th-Century Music, P3 2003 University of California Press

Share and Enjoy:
  • Print
  • Twitter
  • Facebook
  • StumbleUpon
  • Digg
  • Google Bookmarks
  • LinkedIn

9 Responses to “Sorry… we were wrong! By Nick McRoberts”

    • Obrigado ZaĂŻ…

  1. Great post! congrats to both Nick and Dids on an excellent job.

    One important comment: the Losada line in my view is but one representation of what has been dubbed the “Negativity Bias” – the human brain’s tendency to have negative experiences be more impactful, more salient, and more memorable than positive ones. And for that, there is considerable evidence well beyond Fredrickson and Losada’s work. Here is a link to a paper that provides a thorough review of the phenomenon:

    This bias is explored as well as in Rick Hanson’s latest book “Hardwired for Happiness”. Hanson argues that the brain is like Teflon for the positive and like Velcro for the negative, and suggests a number of practices for counter-acting the tendency.

    Like all the Enablers, I have quoted Losada at length and used his famous 2.9 number in courses and coaching alike. I may drop the name “Losada Line”, but I will definitely continue to highlight the importance of being aware of one’s behaviours and language and the overly powerful impact that negativity can have relative to positivity (on oneself, on others, and on teams), and I think the 3-to-1 ratio is a good starting point for that discussion.

    Once again, well done gentlemen!

    Cheers, Pierre

    • Thank you Pierre, I’ll read the article. Funnily enough, I personally ever hardly used the mathematics and ratio around the model. But my daily practice, in dealing with potentially high performing teams who are blocked at some point of their development, shows me that the Behaviours => Emotions => Quality of the intellectual debate remains true. Thanls for your comments mate!

  2. Correction: the title of Hanson’s book is “Hardwiring Happiness”, not “Hardwired for Happiness”. My apologies.

  3. I’m the lead author of the article under discussion.

    I want first to applaud you for putting your hand up in this way. I haven’t seen your previous work using the positivity ratio concept, but I know that a lot of training and development organisations have used the concepts extensively – that’s one of the reasons I worked hard to ensure that the errors in the original work could be brought to public attention.

    I also think that your meta-lesson – on the benefits of admitting mistakes – is a powerful one. Workplace culture, whatever the sector, is often dominated by a perceived need to about showing any weakness or fallibility. Over time, this leads to unsupportable stress on individuals and groups, as the cognitive dissonance associated with all the versions of the truth that have to be maintained builds until something gives.

    We didn’t discuss the empirical evidence, but mostly for reasons of space (and to stay focused). In our opinion, the empirical work in this area is not great either. Just to give one example, Fredrickson and Losada’s own studies from the 2005 article show tiny effect sizes, and one if the results is not even statistically significant (although the author claim that it is).

    Of course, nobody’s denying that people who are doing well in life are likely to be those who also experience more putative emotions. But that’s close to bring tautologous. Also, the empirical work to date hasn’t begun to address what might be the direction of causality. Perhaps people who work in highly effective teams become happier.

    Furthermore, let’s not forget that Steve Jobs, arguably the most successful business figure of the last 10 years, was famous for his regular outbursts of negativity; he was a very tough person to work for, but it’s hard to imagine how Apple could have been more successful. And Apple’s main manufacturing supplier, Foxconn, has installed suicide nets in its factories. It’s easy to increase your company’s overall positivity, like any other performance indicator, if you can turn the bad stuff into an externality.

    Again, congratulations for making the non-obvious decision to write this piece.


    PS: You might want to correct the spelling of the surname of Emily Heaphy.

    • Thank you Nick,
      You indeed have presented us with a healthy challenge: the one of walking the talk that we recommend to our clients when it comes to admitting temporary incompetence of having committed a mistake.
      As Nick writes, we stand behind the idea of the model (Behaviours create the emotional space that heavily impacts the quality of the intellectual debate) as this is what we see every day in our practice with senior teams.
      I like the causality challenge and whereas, indeed it looks like the dynamic can go both ways (Behaviours => Emotional space => Intellectual debate’s quality or is it that Results of a intellectual decision => Emotional space => Behaviours?) but what we see is that results based organizations are pretty fragile when problems arise”. Behaviour or values based companies are far stronger and resilient.
      Thank you for reaching out. It is always nice when an author comes and speaks to us directly… Have a good week

  4. I have not seen The Enablers’ team in action with the Losada model so I can only offer an outsider’s perspective on your blog. I appreciate your mea culpa (particularly since it is so rare in our field of consulting). I also think that the blog raises a series of takeaways and a very relevant question.
    The field in which we delve: human behavior, leadership and change management, is not quantifiable. At most we can try to provide relative opaque benchmarks, ratios and correlations. Yet we are constantly in need to assess where the organization is at, where is it heading and what are the key barriers blocking the way. In essence, we are caught in a dilemma of needing good information and having no reliable data at our disposal. On top of this, Daniel Kahneman in his book “Thinking, Fast and Slow” demonstrates our drive to seeing quantifiable patterns in random noise. So we are predisposed to believing in non-existing correlations.
    Given this background, these are my takeaways:
    First, as absolutes do not exist, it is not helpful to give precise figures. This applies to every behavioral model – including Losada’s. So, if we see anything based on precise values, we should not use it at face value.
    Second, just because the model uses an absurdly precise ratio, it does not discount the whole concept. Our decades of experience in trying to generate change in organization tell us that there is a critical tipping point when an organization moves from putting on the brakes to hitting the accelerator. The core question is: when does this happen? Losada’s model points in a certain direction and it might give us some insights and be useful in some circumstances.
    Third, as we navigate in a perpetual, every evolving fog, it is best to use 2 or more models to obtain multiple reference points when trying to ascertain where our clients are at. This is particularly true as each client has its unique culture, set of players and market dynamic. From my own experience, I know how comfortable I get applying a “one model approach” to an assignment. This is particularly true if the model proved useful in a previous project. So, we constantly need to remind ourselves to apply multiple approaches and measurements when supporting a change management process.
    Lastly, the blog raises the important question of how to evaluate behaviors and what makes a behavior relevant or not. This is a conversation worth pursuing.

    • Thank you Doug, fully agree with your points. Thank you, they certainly represent my point of view at this moment…


Leave a Reply