3 ms·
Every single student in that lottery had to be entered by a parent concerned about their education. That already presents a selection bias that will deliver be
by dublinben 11y ago
Every single student in that lottery had to be entered by a parent concerned about their education. That already presents a selection bias that will deliver better testing results than the entire student population.
- ThomPete 11y agoAnd every single student who want to go to the public school in Tribecca have to have an address there to be zoned for it. If you want to talk about selection bias thats your case not a lottery. Plenty of kids in Harlem goes to Success Academy. Most of the kids in Williamsburg comes from fairly low-income families. That was the very premise of the school and why it's popular. Watch the documentary The Lottery on Netflix it's quite illuminating.
- drumdance 11y agoMy recollection from Freakonomics is they found that the children who lose the lottery still do as well as those who get in. The differentiating factor seems to be parents who care enough to enter the lottery in the first place. Their research was based on magnet schools in Chicago in the nineties. I don't know how well that model correlates with Success Academy.
- ThomPete 11y agoIt's I believe the Hawtorne Effect I believe https://en.wikipedia.org/wiki/Hawthorne_effect https://en.wikipedia.org/wiki/Hawthorne_effect
- drumdance 11y agoI don't believe that applies in the Chicago study because they just looked at data retrospectively. There were no observers when the events were taking place.