Australian Lefty on Politics, Governance, Science and Info Management

Ignorance is power

Posted by Dave Bath on 2009-06-10

Error bars are never valued by those who would benefit from them the most.

A while back I (2008-12-21) mentioned an article from Harvard Business School Working Knowledge that showed people looked more favorably on artful dodgers who avoided questions, than a stumbling truth teller.  Depressing.

There’s another paper to file under "we’re all doomed": "Humans prefer cockiness to expertise" (New Scientist 2009-06-10) – even if the cocksure no-nothing has a proven bad record.

So, the best chance a political party has of winning a seat is to get the most arrogant dissembler they can find, preferably one ignorant of their own ignorance.  Religious zealots will beat subtle theologians for market share.

And so much for the persuasiveness of nuanced opinions and list of caveats.  No-one wants to listen.

Geoscientists, carefully laying out their thoughts on climate change, haven’t a hope in hell of getting past the electorate and the politicians.

Metaphysical arguments and the admission of doubt will find it hard to compete against the simplistic unreasons given by ardent theists – and the more idiotic the theist, the more unaware of their own ignorance, the more convincing they will be!

It’s a bit like my rule about people grading themselves between 1 and 5 out of 5 in various domains.  Ninety-nine percent of people rate themselves as "3", because those who should be a "1" don’t know the extent of their ignorance, while those who should be a “5” have an excellent grasp of the gaps in their knowledge.  Ask me what I don’t know about unix (I used to teach it) and I’ll take a good hour to list things.  Ask ken (met him once at an AUUG conference) what he doesn’t know about unix and he’d probably take a week to give you just the bullet points.

From the New Scientist article:

The research, by Don Moore of Carnegie Mellon University in Pittsburgh, Pennsylvania, shows that we prefer advice from a confident source, even to the point that we are willing to forgive a poor track record.  Moore argues that in competitive situations, this can drive those offering advice to increasingly exaggerate how sure they are.  And it spells bad news for scientists who try to be honest about gaps in their knowledge.

The study used a game, players guessing the weight of people in photographs, and able to "buy" advice (a bit like being offered lifelines in quiz shows).

In the later rounds, guessers tended to avoid advisers who had been wrong previously, but this effect was more than outweighed by the bias towards confidence.

This should have been no surprise.  The baseless "It’ll take two weeks and cost 50K" is preferred by managers rather than "It’ll probably take a couple of weeks, cost between 40K and 60K, but if x, y, or z pops up, that could stretch time-to-delivery to about a month, while if q,r and s don’t surface, it might only take a week".

So I can now state categorically that all J’s on the Myers-Briggs should be taken out and shot, so we P’s have a chance of running things properly.

OK, so maybe we don’t kill those with borderline J-ness, perhaps only those in the top 3 or 4 deciles of J-ness, …. and maybe not if there is a hope of upping their P-ness in the next 5 through 20 years… but maybe we keep just a few J’s alive, even though I can’t think right now why we’d need them, there might be some value in having a few breeding specimens…

Curses!  Foiled again!  Hoist with my own P-type!

See Also

  • Yep. I’m an off-the-scale NTP, with a little i at the front, which is why I prefer personality typing tools (I once wrote psychometric software used by clinicians, so I’m an informed user of such tools) that allow graduated rather than yes/no responses. Grab a copy (Microsoft EXE file included, but unix source available) of the DDLI which is pretty darn good – and even has an "If I’m wrong, there is a chance you are an XXXX because…" section, or at least try this – less subtle than the DDLI, but very much better than most Yes/No MBTI tools.

13 Responses to “Ignorance is power”

  1. […] Ignorance is Power. […]

  2. Jacques Chester said

    You should know better than to rely on Meyers-Briggs, Dave. It’s about as scientific as leeches and mercury injections.

  3. Isn’t your 1-5 rule the same as the Dunning-Kruger effect?

    Given the number of cognitive biases in play (see it is hard to imagine expertise having a major impact on people’s ability to influence. Except, of course, expertise in influencing people. Dale Carnegie said something along these lines last century.

  4. memeweaver said

    Agreed on the 1 to 5 scale of rating one’s ignorance/knowledge. Knowing the gaps is knowing much more.

    Not entirely related, but on your comment regarding employing the most arrogant dissembler in order to have the most influence, I heard on the radio a couple of days ago that the Bush administration apparently employed the magician David Blaine to do interior design in the Pentagon. Apparently, no one is quite sure what exactly his role involved, but that it probably had something to do with special effects.

    I am an INTJ that has already bred. So, look out P-heads. We’re going to judge you into oblivion, if you’re not nice to us.

  5. Dave Bath said

    Peter Williams@3
    Yes it is probably a manifestation of the Dunning-Kruger effect. Didn’t know the name. And thanks for the link to that cognitive biases page.

    Myers-Briggs (and similar things) will be useless if the test is useless, either because the Ps can’t by nature respond accurately to a YES/NO question (hence my liking for the DDLI), there is no subtlety to the questions so you can answer to get the result you want, or there is no quantitative reporting so you cannot take into account borderliners.

    However, it’s not a bad classification scheme. With a good test, taking into account the different types a person might be (where different combinations of type components can cause similar responses), it does allow useful guidance for individuals (if you are a 99% “P” don’t accept an offer to swap your tech lead job for a project manager position – any increase in pay will be spent on antihypertensives) and for putting together teams that are likely to be successful.

    The problems with Myers-Briggs usage (especially of poor tests) is all the J types who ignore the subtleties and treat it like a “silver bullet”, often for inappropriate situations. (I have a low opinion of the utility of silver bullets. They only work on werewolves and werewolves don’t exist.)

    Any employ/no-employ decision based on MBTI and similar, without understanding the context of the person and their roles, is negligent. Can you imagine an Australian boutique software testing company being smart enough to have 75% of their testers on the autism spectrum?

    For the record, leeches are very useful in modern medicine, reducing the oedema after nasty surgery/transplants when the veins/lymphatics aren’t functioning, risking both pressure necrosis and poor oxygen supply to the affected part. Hg, on the other hand, was a treatment justified by homeopathic philosophies and thus with no redeeming features. However, mercurial diuretics had some value before more modern drugs if the need to get rid of fluid was greater than the risk of toxicity.

  6. Jeez, Dave, I’m an INTJ and you want to shoot me? And you’re an INTP? But we have so much in common – I’m just more sure of it than you. :) Join me, Dave, hold reason as your only absolute, pick up that line of logic and run with it as far as it will go. It’s quite a ride. Much better than being uncertain and undecided. I think you’d like it.

    Seriously, I don’t find that study all that surprising. In the absence of being able to perform your own knowledgeable analysis, and when being put on the spot for a relatively quick answer, it’s pretty logical to go with confidence. More than likely, in the history of humanity, more good decisions have been made this way than bad ones.

    The problem is people who are very good at generating an appearance of confidence to push their own agenda, who use this particular human attribute to mislead others, usually for their own benefit. Just look at the AGW moguls like Gore.

  7. Regarding the Myer-Briggs Type Indicator, I agree it is a very useful tool. I have no training in psychology, and all my management training is quite low level i.e. primarily military (and I’m an engineer and INTJ!) so I use it as a rule-of-thumb for selecting personalities for specific functions when putting a team together. I’ve found it very (perhaps extremely) useful and successful.

    Incidentally, when deciding on what functions I need in the team I use the military Common Staff System as a rule-of-thumb. It works well for constructing organisations of all types (not just military).

    Both of these systems are well worth committing to memory IMHO. When you just need a little bit of direction to get things moving, or you’ve drawn a mental blank, you can always turn to them in an instant, and this gets enough momentum up for your own rational thinking to take over.

  8. Dave Bath said

    Memeweaver and Mick:
    Your “NT”-ness stops you from being put up against the wall! (And I hope you’ll note that the post started with a straight face and gradually changed, so that by the end I was giving myself buccal stretchmarks).

    Mick: as an Engie, I’m sure you like error bars, confidence levels, yada yada.

    Memeweaver: The “yes no yes no but” comment on another post indicates that you aren’t at least a TOTAL J.

    And I wouldn’t worry too much. Even if you are tied up in front of the wall, the P posse will probably be undecided about the color of the blindfold, the brand of cigarette, the optimal size of the squad, the best weapon, the most humane place to put the bull’s-eye that you’ll die of old age first.

    So, maybe we /can/ join forces and start wiping out another personality type: the PHB. (See this, this or this if you are unfamiliar with Dilbertian Psychosocial Theory). I’m getting out my protractor, ruler, and whatever-you-measure-albedo-with as we speak.

  9. I’m an ENFP – but on the borderline of J. I need deadlines otherwise I never hand things in – but I do tend to meet the deadline once I have it. The difficulty I face in life is knowing when to stop researching. As far as I’m concerned, there is always more knowledge out there.

    I have the opposite of cockiness – imposter syndrome (as detailed in a post here. The more I know, the more I doubt my own knowledge. I need a cocky person to present my research for me.

  10. Just read this article about the role of cognitive bias in decision making on the NPR website. It’s from the Nudge folks who are now embedded in the US government.

    It looks like a rational response to cognitive bias might be as central to the Obama administration as symbolism was to the Carter administration. It will interesting to see if a Duane Delacourt figure emerges.

  11. Jacques Chester said

    The problem is that MBTI isn’t based on anything empirical; it’s based on Jungian archetypes. It categorises people, but so do horoscopes, with about the same degree of efficacy.

    MBTI is the QWERTY keyboard of personality tests. Not very good, but well-known and popular. For reasons I am not able to understand, it is wildly popular amongst two very different audiences: smart nerds and HR morons. I presume the former because it provides a neat and tidy framework, and the latter because it provides them some shortcuts to avoid any semblance of human thought.

    The premier personality theory at the moment is the Five Factor Model. It’s imperfect, debatable and possibly incomplete. But critically it is also the only empirical categorisation model so far developed. Every other test is based in a hand-waving theory, most of them Freudian in origin. The 5FM at least derives from two independent, empirical sources — dictionary studies and factor analysis of correlations between other personality tests.

    Of course, worse than the MBTI is the “Four Colours” model I had foisted on me at one workplace. The HR manager couldn’t for the life of her work out why I didn’t fit into any of the colours. I decided not to tell her that a two-dimensional model of personality is even less descriptive than astrology.

  12. Dave Bath said

    Jacques: gotta preferred url to 5FM (esp any src) or do i just hit wikipedia? Sounds interesting, even to my Platonist parts.

  13. Jacques Chester said

    Dave — Wikipedia has an OK intro. Otherwise check out a personality textbook from your local library.

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

%d bloggers like this: