Golf scoring as a measurement problem

Posted on by

Following up on our link to an article about educational measurement, Eric Loken pointed me to this:

On the Criteria Corporation blog we [Loken] just posted a look at golf tournament scores. If you take the four rounds as if they were four repeats of the same test, or four parallel items on a test, the usual psychometric analyses would yield a terrible reliability coefficient. The problem of course is restriction of range of true scores among the world’s best golfers. We figured since the US Open (this weekend) is sometimes called the Ultimate Test we’d offer a little psychometric analysis of golf.

Despite having published an article on golf, I know almost nothing about the sport–I’ve never actually played macro-golf–so I’ll link to Eric’s note without comment.

6 thoughts on “Golf scoring as a measurement problem”

  1. If the analysis is correct, it casts Tiger Woods in a new light. His latent ability dominance must be extraordinary if he dominates when ability is measured poorly.

  2. Just to mention that the post actually demonstrates that golf is not just luck. If it were, roughly half the correlations would be negative, rather than uniformly positve, if small. That said, way out in the tail of talent, random individual variation is considerably larger than underlying skill variation,yielding small (positive) day-to-day correlations.

  3. Jonathan,

    But nobody things golf is just luck, right??? That's what I call a null hypothesis that isn't worth even thinking about!

  4. I hope I made it clear that I don't think golf is just luck! The point is that if a typical golf tournament were viewed as a series of tests, it would have horrendous measurement properties.

    The fault isn't with the game, it's with the near perfect competition. Because the players are so tightly clumped at the top of the ability distribution, measurement error seems to dominate from week to week.

    And yes, you could calculate the probability of a golfer from the professional ranks winning 30% of his tournaments against 150 opponents. And it would very likely confirm that Tiger is out of this world.

  5. Your comment about measurement error separating winners and losers in the far right tail reminds me of my days as a cross-country runner. My classmates would ask how I did, and I'd say 15th, or what-have-you. The next question was invariably "Out of how many?" Thus the lay statistician understands the important of he underlying distribution as well, at least as far as it is accounted for by sample size.

Comments are closed.