4 ms·

I wonder if the `control group` were students that did not want to be enrolled in the program (introducing significant bias into the results) or students select

by hurin 12y ago

I wonder if the `control group` were students that did not want to be enrolled in the program (introducing significant bias into the results) or students selected randomly to be tracked and not enrolled or given such an opportunity (the ones 60% more likely to commit a violent crime).

m_mueller 12y ago
I wanted to say that as well - however in this particular case I'd say the ends justify the means if they get funding for more tutoring, even if their statistical methods are unsavoury.
- tracker1 12y ago
  Many schools see a ratio of teachers to students around 25:1 (or higher).. they're talking about getting a ratio of 2:1 for even 1/6 of the school day... that's roughly the need for double the number of teachers/tutors over current needs. Of course there are other offsets. I'm not sure I'm completely with the economists on this one, but I can definitely see some of the points. I think doing something like this in the K-8 years would be far more effective than in the H.S. years where kids should start to be more independent. If cities could work with local businesses to allow STEM employees to volunteer (paid by the company) a couple days a month at a local school, in addition to getting parents to volunteer 2-3 days a month, this could get the coverage needed and starting at much younger ages before a lot of the issues start to settle in.
  - m_mueller 12y ago
    > If cities could work with local businesses to allow STEM employees to volunteer (paid by the company) a couple days a month at a local school, in addition to getting parents to volunteer 2-3 days a month, this could get the coverage needed and starting at much younger ages before a lot of the issues start to settle in. I think that's a great idea. Some management types can checkmark 'social responsibility' that someone added to the company's mission statement, and employees get to have some mini sabbatical and feel good about themselves, probably come back with more energy.
rcheu 12y ago
In the paper, they claim that the assignment was properly randomized. It seems like they got authorization to assign tutors to some students, and randomly assigned some percentage of the students tutors.

by hurin 12y ago

m_mueller 12y ago

I wanted to say that as well - however in this particular case I'd say the ends justify the means if they get funding for more tutoring, even if their statistical methods are unsavoury.

tracker1 12y ago
Many schools see a ratio of teachers to students around 25:1 (or higher).. they're talking about getting a ratio of 2:1 for even 1/6 of the school day... that's roughly the need for double the number of teachers/tutors over current needs. Of course there are other offsets. I'm not sure I'm completely with the economists on this one, but I can definitely see some of the points. I think doing something like this in the K-8 years would be far more effective than in the H.S. years where kids should start to be more independent. If cities could work with local businesses to allow STEM employees to volunteer (paid by the company) a couple days a month at a local school, in addition to getting parents to volunteer 2-3 days a month, this could get the coverage needed and starting at much younger ages before a lot of the issues start to settle in.
- m_mueller 12y ago
  > If cities could work with local businesses to allow STEM employees to volunteer (paid by the company) a couple days a month at a local school, in addition to getting parents to volunteer 2-3 days a month, this could get the coverage needed and starting at much younger ages before a lot of the issues start to settle in. I think that's a great idea. Some management types can checkmark 'social responsibility' that someone added to the company's mission statement, and employees get to have some mini sabbatical and feel good about themselves, probably come back with more energy.

rcheu 12y ago

In the paper, they claim that the assignment was properly randomized. It seems like they got authorization to assign tutors to some students, and randomly assigned some percentage of the students tutors.