This is a nice articulation of some important points. In particular, the warning against “measuring only what we can instead of what we should” is a welcome reminder about the importance of clarity about desirable outcomes, not just rigor in measurement.
Alas, the previous parenthetical sentence undermines the effect. “Sometimes we simply can’t wait 20 years to see if, say, an educational program caused a lower assimilation rate.” Well, yes, that’s true. But not just because we have to make policy decisions in the short term, without the luxury of waiting 20 years. Much more importantly, “lower assimilation rate” is neither a useful metric nor even a conceptually coherent one.
Another way of putting the point: Assuming that educational or cultural investments should result in “lower assimilation rate” is the kind of fuzzy thinking that the rest of the article is arguing against.
This is a nice articulation of some important points. In particular, the warning against “measuring only what we can instead of what we should” is a welcome reminder about the importance of clarity about desirable outcomes, not just rigor in measurement.
Alas, the previous parenthetical sentence undermines the effect. “Sometimes we simply can’t wait 20 years to see if, say, an educational program caused a lower assimilation rate.” Well, yes, that’s true. But not just because we have to make policy decisions in the short term, without the luxury of waiting 20 years. Much more importantly, “lower assimilation rate” is neither a useful metric nor even a conceptually coherent one.
Another way of putting the point: Assuming that educational or cultural investments should result in “lower assimilation rate” is the kind of fuzzy thinking that the rest of the article is arguing against.
Jon A. Levisohn
Brandeis University