I appreciate you taking the time to write this James. It was refreshing to read.
However, I am not sure that impact is often measurable. What evidence do you have for that intuition? Can you provide any examples to illuminate this point?
My prior is that measurement in the social sciences is really, really hard, and people often update too much on social science statistics. And that the social world has a lot of uncertainty. It seems that people update too much on them because the social scientists who end up promoting findings to the public aren’t that forthcoming about their limitations and the ongoing disagreements in the field. One examples is social media’s affect of minors’ mental health, and the ongoing debate between Jon Haidt and Candice Odgers.
James’ post did provide an example: the Maternal Health Initiative shut down due to their pilot results not yielding the 10% shift in contraceptive uptake their CEA informed them to aim for. Maybe this wasn’t what you were looking for?
That’s an example of a stat illustrating impact, but I’m looking for examples that support the statement it is often the case that if you have impacts there are stats to illustrate that.
I appreciate you taking the time to write this James. It was refreshing to read.
However, I am not sure that impact is often measurable. What evidence do you have for that intuition? Can you provide any examples to illuminate this point?
My prior is that measurement in the social sciences is really, really hard, and people often update too much on social science statistics. And that the social world has a lot of uncertainty. It seems that people update too much on them because the social scientists who end up promoting findings to the public aren’t that forthcoming about their limitations and the ongoing disagreements in the field. One examples is social media’s affect of minors’ mental health, and the ongoing debate between Jon Haidt and Candice Odgers.
James’ post did provide an example: the Maternal Health Initiative shut down due to their pilot results not yielding the 10% shift in contraceptive uptake their CEA informed them to aim for. Maybe this wasn’t what you were looking for?
That’s an example of a stat illustrating impact, but I’m looking for examples that support the statement it is often the case that if you have impacts there are stats to illustrate that.