So interesting. I think the 0-25 group is more nuanced. If the cut off for reporting is 11 students, in the smallest grade schools, you're potentially taking about a weighting towards schools that are majority a single race. So that begs questions like, who is in small, single race schools? How would you expect them to perform? One very big difference is that you're then likely talking about an immigrant population that potentially has more language focused needs to succeed vs. historically underprivileged community which maybe needs more resources to succeed. I don't think it's necessarily just sample size we're observing here, and there is danger to these communities if we don't stop and think about why, and how they might learn best. Anyway, I think it's important because we notice differences in performance. If these differences are brushed off, we might never ask the questions to help us understand why and what can be done about it.
I have been advocating for K-8 schools for years. Stable school communities are beneficial for kids, parents, and teachers and adding three extra years of community building is invaluable. Speaking from experience as a mother of six children who have attended either a k-5, k-8, and/or 6-8 I can attest to the total upheaval that transition is during a really emotionally intense time in a kids development. Im sure we can all agree, middle school is the armpit of education. For some reason someone thought it was a good idea to take a bunch or kids all going through the same hormonal upheaval and put them in the same building. And many of these kids are coming from smaller schools with teachers who have been with them for years, to giant middle schools where they feel nameless, faceless, overwhelmed, and isolated.
Yes, since all the relationships go in the same direction, you have clearly demonstrated that there is no relationship. And since they all go in the same direction one could get the impression that maybe larger schools are better.
It seems clear that there is no evidence that small schools are good for student learning. When does a weak correlation become no correlation even if all of them are in a positive direction.
Well, there are statistical ways to calculate whether a relationship exists or is just noise. I avoid using many statistical terms in my posts because (a) I'm writing for a lay audience and don't want to put them off by jabbering about p-values and null hypotheses, and (b) I'm aware of my own limits as a statistician and don't want to start making mistakes in a bid to sound smarter than I am. What works in my favor here is that if someone wants to say there is a relationship between two variables (School size and student growth in this case), they have to prove it. The default assumption is that there is no relationship and that's all I was trying to show here.
So interesting. I think the 0-25 group is more nuanced. If the cut off for reporting is 11 students, in the smallest grade schools, you're potentially taking about a weighting towards schools that are majority a single race. So that begs questions like, who is in small, single race schools? How would you expect them to perform? One very big difference is that you're then likely talking about an immigrant population that potentially has more language focused needs to succeed vs. historically underprivileged community which maybe needs more resources to succeed. I don't think it's necessarily just sample size we're observing here, and there is danger to these communities if we don't stop and think about why, and how they might learn best. Anyway, I think it's important because we notice differences in performance. If these differences are brushed off, we might never ask the questions to help us understand why and what can be done about it.
I have been advocating for K-8 schools for years. Stable school communities are beneficial for kids, parents, and teachers and adding three extra years of community building is invaluable. Speaking from experience as a mother of six children who have attended either a k-5, k-8, and/or 6-8 I can attest to the total upheaval that transition is during a really emotionally intense time in a kids development. Im sure we can all agree, middle school is the armpit of education. For some reason someone thought it was a good idea to take a bunch or kids all going through the same hormonal upheaval and put them in the same building. And many of these kids are coming from smaller schools with teachers who have been with them for years, to giant middle schools where they feel nameless, faceless, overwhelmed, and isolated.
Yes, since all the relationships go in the same direction, you have clearly demonstrated that there is no relationship. And since they all go in the same direction one could get the impression that maybe larger schools are better.
It seems clear that there is no evidence that small schools are good for student learning. When does a weak correlation become no correlation even if all of them are in a positive direction.
Well, there are statistical ways to calculate whether a relationship exists or is just noise. I avoid using many statistical terms in my posts because (a) I'm writing for a lay audience and don't want to put them off by jabbering about p-values and null hypotheses, and (b) I'm aware of my own limits as a statistician and don't want to start making mistakes in a bid to sound smarter than I am. What works in my favor here is that if someone wants to say there is a relationship between two variables (School size and student growth in this case), they have to prove it. The default assumption is that there is no relationship and that's all I was trying to show here.