One of the big discussions in impact evaluation at the moment is the value that RCTs (randomized controlled trials) bring by providing an unbiassed estimate of the mean net effect. Which brings to mind that classic quote by John Tukey:
Far better an approximate answer to the right question, which is often vague, than an exact answer to the wrong question, which can always be made precise.
- The future of data analysis. Annals of Mathematical Statistics 33 (1), (1962), page 13.
This week’s Friday Funny ponders the relative value of precise measures and estimates:
A museum visitor was admiring a Tyrannosaurus fossil, and asked a nearby museum
employee how old it was. “That skeleton’s sixty-five million and three years, two months and eighteen days old,” the employee replied. “How can you know it that well?” she asked. “Well, when I started working here, I asked a scientist the exact same question, and he said it was sixty-five million years old – and that was three years, two months and eighteen days ago.”