What is the meaning of the two graphs? How are they consistent? ETA: the graphs come from here, but that doesn’t help. My guess is that they have different baselines, the first using the previous measurement of 20% and the second using the control group, if it regressed to 30%. But they still don’t look consistent to me.
When I looked at Project Follow-Through, about 10 years ago, when DI was considerably less popular than it is now, the graphs looked very different. No single method beat out all the others, though three or four were clearly better than the others.
What is the meaning of the two graphs? How are they consistent?
ETA: the graphs come from here, but that doesn’t help. My guess is that they have different baselines, the first using the previous measurement of 20% and the second using the control group, if it regressed to 30%. But they still don’t look consistent to me.
When I looked at Project Follow-Through, about 10 years ago, when DI was considerably less popular than it is now, the graphs looked very different. No single method beat out all the others, though three or four were clearly better than the others.