| FTA: "Let’s begin with the Perry Preschool Experiment, which consisted of 123 low income African-American children from Yspilanti, Michigan." and three sentences later... "The differences, even decades after the intervention, were stark: Adults assigned to the preschool program were 20 percent more likely to have graduated from high school and 19 percent less likely to have been arrested more than five times." The above passage contains the only numbers I could find (cited paper's behind a paywall).
Those numbers mean _nothing_. 123 people 20 % more likely to graduate = 67 graduated, 56 did not 67 - 56 The author suggests that 11 people graduated high school because of pre-school. He/she could have said "11 more people graduated high school than did not", but he/she didn't. Tried to obfuscate it. "19 percent less likely to have been arrested more than five times." They obviously tried to find something that reinforced their point instead of looking at reality. They didn't say "19 percent fewer arrests", they said "less likely... more than 5 times". Because the real difference in mean arrests is probably much smaller. Again the study is behind a paywall so I can't read it, but I suspect they're not even wrong. |