Thursday 26 March 2015

#Germanwings Pilot selection

In the absence of hard data on Germanwings’ selection and monitoring procedures, at least at the moment, (they now say they use psychological assessments) I have looked at large data sets on US fighter pilots. If one compares intelligence measures with personality inventories, then intelligence has the lead:

The predictive validity of cognitive ability and personality traits was examined in large samples of U.S. Air Force pilot trainees. Criterion data were collected between 1995 and 2008 from 4 training bases across 3 training tracks. Analyses also examined consistency in pilot aptitude and training outcomes. Results were consistent with previous research indicating cognitive ability is the best predictor of pilot training performance. There were few differences across training tracks, bases, and years, and none was large. Overall, results illustrated the consistency of the quality of pilot trainees as assessed by cognitive ability and personality trait measures, and the consistency of these measures in predicting training performance over time. This consistency results in a more stable training system, enabling greater efficiency and effectiveness.

 

 

image

https://drive.google.com/file/d/0B3c4TxciNeJZQnVZWnpiU2RNeVU/view?usp=sharing

It is probably best to look at primary training for the most representative results (after that one is progressively dealing with very good pilots) but the personality measures contribute less strongly, only in terms of Openness to Experience and Conscientiousness. If you accept the fully corrected scores in the last column of each training level then intelligence is the most substantial factor. Remember that the applicants are already highly selected for intelligence, so these correlations are surprisingly large.

Of course, being a fighter pilot is more demanding than commercial flying, but the tasks are broadly the same.

23 comments:

  1. We'd expect IQ to correlate with training performance. I'd be interested to see career performance.

    Openness to experience as measured in the Big Five is confounded with IQ. Another reason why people should switch to the HEXACO.

    ReplyDelete
  2. Hi James,

    We're accustomed to seeing IQ surpass personality measures, but this is in no small part due to our difficulty in accurately measuring personality. If we leveled the playing field, as it were, by demanding that the study authors simply asked people to self-rate their IQ, these IQ x performance correlations would be cut by a factor of two or three. Also, as JayMan points out, Openness is confounded with IQ (although HEXACO Openness is still confounded with IQ; the HEXACO is better than the Big Five, but not perfect). That O correlated negatively with pilot performance despite all of these roadblocks preventing such a correlation from appearing is to me very interesting - the true negative correlation between O and flying ability is likely to be far, far larger than found here.

    ReplyDelete
    Replies
    1. Like your suggestion that, for real comparability personality should be measured against self-estimated IQ. Of course, I would really prefer that personality were rated on observer assessments.One of the best recoveries from a 4 engine flame out because of volcanic dust was done by a pilot famed for his methodical, unimaginative approach. http://en.wikipedia.org/wiki/British_Airways_Flight_9

      Delete
  3. Was General Patton sane? MacArthur? Why did the previously brave James VII and II break down and flee from William of Orange?

    The problem is perpetual.

    ReplyDelete
  4. Flying airplane needs a lot of knowlage about physics and problems solving skills, which is what IQ predict for.

    IC

    ReplyDelete
  5. warning - the following is in poor taste, yet true.

    we should measure bladder size - if the pilot had a larger bladder the co-pilot could not have taken over the plane (b/c the pilot would not have left the cockpit).

    perhaps pilots can be made to wear texas catheters during flights. female pilots would of course, need foley catheters.

    sorry for the statements above - delete if need be - i just felt it's not getting mentioned.

    ReplyDelete
    Replies
    1. I already tweeted some days ago that a portable toilet would have got round the problem. Since urination is more frequent than defecation, a simple pack of absorbent would do the trick. By the way, natural functions are not in poor taste, they are.......natural.

      Delete
  6. My wife points out that a photo in the paper suggests that the chap had dyed his hair. Unsound, therefore.

    ReplyDelete
    Replies
    1. Hard to judge. If it was a recent change, say in the days before his final flight, I would agree that it could be diagnostic.

      Delete
  7. @James - Probably, the cause of this pilot's behaviour is more likely to be associated with SSRI-type 'antidepressant' drug use than with 'depression' or deficient selection - SSRIs certainly, but rarely, can cause the kind of state which leads to violent suicide of a strange and uncharacteristic type - but depression, not really.

    http://www.davidhealy.org/wp-content/uploads/2012/05/2000-Healy-Healthy-Volunteer-Suicide1.pdf

    ReplyDelete
    Replies
    1. I knew a case, made a suicide attempt, survived. Hard to associate it directly with medication. Have not read what medication pilot was on, and for how long. Cannot understand why he was totally silent going down. Would have expected some tearfulness, or some comments of some sort.

      Delete
  8. There is apparently evidence in his apartment of depression medications: http://www.straitstimes.com/news/world/europe/story/germanwings-flight-4u9525-crash-medicine-psychological-illness-found-co-pilo

    Germany has rules about privacy. As the evidence reported in the press strongly suggests he took steps to hide his condition from his employer, what can be done to prevent such terrible acts in the future?

    ReplyDelete
    Replies
    1. first, we need proper data on his medical history. If we can never get that then it is harder to judge, but airlines could make full voluntary disclosure a condition of employment, and monitor pilots even more carefully. There is a real problem with finding reliable detection techniques, which might do more harm than good if they put pilots under extra pressure.

      Delete
  9. "being a fighter pilot is more demanding than commercial flying": I suspect that, rather, it's demanding in different ways. The airline pilot should presumably be calm, stolid, cautious; a man with patience and stamina for the long term. That sounds unlike fighter pilots. Anyway, the questions about this nutter have got little to do with his competence, have they? Or maybe time will reveal more and more twists to the tale.

    ReplyDelete
    Replies
    1. The world, so far, has had good luck in picking nuclear missile submarine captains. Apparently the Navy looks for calmness.

      Delete
    2. The analogy may be rather obvious, but flying a fighter is probably more like shooting with a shotgun, flying an airliner like shooting with a rifle. When I once tried clay pigeon shooting, the instructor told me that my experience with rifles would put me at a disadvantage.

      Delete
    3. Dear Steve, A psychiatrist friend talked, informally and off the record, to a Polaris captain about whether he would "press the button" and he replied laconically "It would take more than an order to make me do it".

      Delete
    4. Fighters not only have to be able to fly, but also be able to kill. Some Falklands pilots found that, to their surprise, in real combat that the second was difficult. Fighter pilots know they have to get in first. Airline pilots know that most of the time nothing happens, but they must be ready to react quickly and appropriately if it does.

      Delete
  10. During the 2004 election, I did a lot of research into the military officer qualification tests taken in the 1960s by George W. Bush and John F. Kerry. I wrote:

    How did Bush do? In estimating his IQ, we can probably throw out his high score (the 95th percentile on Officer Quality) and his low score (25th percentile on Pilot Aptitude) because those tests don`t measure IQ very directly. Instead, we should concentrate on his Verbal Aptitude (85th percentile), Quantitative (65th), and Navigator Aptitude (50th). In fact, those three are fairly similar in subject matter to the three parts of the Naval OQT that Kerry took: Verbal Analogies, Arithmetic Reasoning, and Mechanical Comprehension, respectively.

    The Officer Quality score was derived by combining Bush`s score on the 60 item Quantitative Aptitude subtest, the 60 item Verbal Aptitude subtest, with the 100 item Officer Biographical Inventory. The latter was a personality test that asked about "past experiences, preferences, and certain personality characteristics related to measures of officer effectiveness." It inquired into enthusiasm for sports and hunting, and was only vaguely correlated with IQ.

    (A retired Air Force test psychologist told me that this section was later dropped because women did very poorly on it, and urban and suburban youths didn`t do as well as country boys. "It was politically incorrect, but"—he recalled wistfully—"It was a predictor of success as an officer.")

    http://www.vdare.com/articles/this-just-in-kerrys-iq-likely-lower-than-bushs

    The retired military psychometrician told me that now banned Officer Biographical Inventory asked a lot of questions about how much you engaged in blood sports and competitive sports, especially contact sports. So, it was a personality test, but one focused on your commitment to winning and willingness to kill.

    ReplyDelete
    Replies
    1. I remember reading your Bush/Kerry comparison at the time. When I lectured to some physicists about IQ I got them to estimate Bush's IQ (they pitched it very low, about 65th percentile) and I was able to give them your estimate.
      On the biographical inventory, it is a great pity that good predictors are dropped for bad reasons.

      Delete
  11. In general, being an airline pilot is not a terribly well-paid job these days, at least in the U.S. It does have a lot of applicants relative to its pay, though, because it's an impressive job.

    ReplyDelete
  12. I was under the impression that fighter pilots often became airline pilots after retirement from the military.

    How does personality correspond with morality? Is it a sign of mental illness to plot and plan to kill hundreds of people? At what point is it acceptable to say evil, rather than ill? I know enough about history to know that at most points in time, it would be dangerous to rely upon the kindness of strangers. We are not naturally benevolent, in my opinion.

    The more intelligent someone is, the better equipped he is to evade procedures.

    ReplyDelete
    Replies
    1. Fighter pilots used to go into commercial airlines, and it was because of their dominance that co-pilots had to be trained in crew resource management. I was reading about the key crash today: copilot too abashed to mention they were running out of fuel. Chicago airport, I think.

      Delete