PARCC Comparison for MIDDLE SCHOOLS in MoCo and Upper Northwest

[Anonymous]

Anonymous wrote:

Anonymous wrote:Compute individual performance metrics for each subgroup. E.g., what percentage of white students scored a 4 or 5? What percentage of hispanic students scored a 4 or 5?

Next, get composition data for each school. E.g., at Pyle 76% of test takers were white, 1.8% were black, etc.

Compute an aggregate school measure where the performance data is is the % scoring a 4 or 5 in that subgroup at Deal and the weight is the % of test takers in that subgroup at Pyle. Then add these products for each subgroup together. Do this for each school.

Wouldn't it be easier to compare the scores of each subgroup?

No, because you can't roll that up into a summary measure to compare the schools as a whole, which I think was OP's goal.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:OP, your analysis is indeed shaky. The fact hat you made the large leap by assuming mixed race means black/white and black/white kids at Deal fall into bad crowds (black) at Deal kind of threw it for me. You then go on to defend your not using other DC schools because guy said there were a handful of schools that aren't horrible. That's not the same as saying Brent is as good as Janney etc. You didn't even reference Ross (probably the best school in DC) and you didn't reference Shepherd (and adjust for demographics as you did with other school). My guess is because it's too black for you.

OP here.

You caught me. I'm a not-so-closeted racist. I've been trying to advance my white supremacy agenda by omitting certain schools from my volunteer analysis. I would have succeeded had your powers of discernment not caught me. Bravo to you. The community is much better for your presence and owes you a debt of gratitude. (Just to test your skill, I've hidden a message within this paragraph. See if you can find it.)

With the exception of Hearst, which was an omission, I chose the schools closest to the supposed best MoCo schools. (And for the MoCo schools, I chose the ones with data closest to DC.) My goal was to help inform someone considering moving to one side of the border or the other. There is no point in including Churchilll feeders since the peso looking to move into WCHS feeder zone is unlikely to be considering also staying inside the DC border.

I would compare Shephard to the Silver Spring schools, not the western MoCo schools since, again, they are geographically proximate.

You are welcome to perform analysis on any schools you choose. We will wait with bated breath.

As for mixed-race, I simply compared them. Lacking any data or credible knowledge about the potentially different composition of mixed-race families, this or omission are the only reasonable choices.

Problem is you didn't "simply compare" mixed race. You drew quite a huge inference.

I quote, "Finally, an odd note: multiple race (two or more) advanced math students perform better at the MoCo schools than at the DC schools. This may confirm the fears of parents of mixed-race children that their children fall into the “wrong” peer group when surrounded by a more diverse student body. It is certainly something worth monitoring going-forward."

Deal is more diverse (more black) and did worse in this category so you're assuming the mixed race kids at Deal hang around the wrong peer group (ie, all black kids) and therefore do poorly on tests. There is no way to get away from that statement.

You do realize that most AA are of mixed race!

This tired line again. Being part something isn't the same as being part of something (please re-read that a few times before replying). You and I both know that the term "mixed race" is generally used in this country to describe people with parents who are two different races. You are confusing that group with AA who have some (often not obvious) non-African ancestry.

[Anonymous]

Anonymous wrote:The Census should start counting Obtuse Americans.

So many of them, yet so systematically overlooked.

I don't get it.

[Anonymous]

Anonymous wrote:

Anonymous wrote:The Census should start counting Obtuse Americans.

So many of them, yet so systematically overlooked.

I don't get it.

Counted

[Anonymous]

Bumping this because it was linked in another thread and the analysis is excellent. I never saw it the first time and it's worth reading (analysis is on first page with some qualifiers afterward). Thank you OP!

[Anonymous]

OP, there is a huge huge assumption underlying your post -- that demographic categories are monolithic. You are assuming that SES and cultural capital holds constant across the demographic groupings in the data. It's an interesting way to crunch the data but it's not standard for the ed. field.

[Anonymous]

Anonymous wrote:OP, there is a huge huge assumption underlying your post -- that demographic categories are monolithic. You are assuming that SES and cultural capital holds constant across the demographic groupings in the data. It's an interesting way to crunch the data but it's not standard for the ed. field.

I don't understand this criticism. OP is only assuming subgroups are similar across the schools. That is, white students at Deal are similar to white students at Pyle, Westland and Hardy, or that Asian students are comparable across these four schools. This is reasonable given the schools being compared are all within three miles or so of the DC-MD border. They're all within a four mile radius from Westmoreland Circle.

[Anonymous]

Anonymous wrote:OP, there is a huge huge assumption underlying your post -- that demographic categories are monolithic. You are assuming that SES and cultural capital holds constant across the demographic groupings in the data. It's an interesting way to crunch the data but it's not standard for the ed. field.

What more do you expect the OP to do with the data available? All analyses like these have major assumptions.

[Anonymous]

Thank you! We currently rent in AU Park (ie we can easily move) and I have often wondered if we were short changing our kids by going to deal. Now I can rest easy and not consider another move. I really appreciate your analysis.

[Anonymous]

Anonymous wrote:

Anonymous wrote:OP, there is a huge huge assumption underlying your post -- that demographic categories are monolithic. You are assuming that SES and cultural capital holds constant across the demographic groupings in the data. It's an interesting way to crunch the data but it's not standard for the ed. field.

I don't understand this criticism. OP is only assuming subgroups are similar across the schools. That is, white students at Deal are similar to white students at Pyle, Westland and Hardy, or that Asian students are comparable across these four schools. This is reasonable given the schools being compared are all within three miles or so of the DC-MD border. They're all within a four mile radius from Westmoreland Circle.

+1

It is not reasonable to assume nationwide that all white kids are similar. White kids in appalachia versus white kids at prep school on the upper east side, etc.

But when comparing the DC area schools that OP compared, it's fair enough to treat the racial demographic as a broad proxy. It's infortunate that this is appropriate, says a lot about racial inequality in DC. But it's appropriate.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:OP, there is a huge huge assumption underlying your post -- that demographic categories are monolithic. You are assuming that SES and cultural capital holds constant across the demographic groupings in the data. It's an interesting way to crunch the data but it's not standard for the ed. field.

I don't understand this criticism. OP is only assuming subgroups are similar across the schools. That is, white students at Deal are similar to white students at Pyle, Westland and Hardy, or that Asian students are comparable across these four schools. This is reasonable given the schools being compared are all within three miles or so of the DC-MD border. They're all within a four mile radius from Westmoreland Circle.

+1

It is not reasonable to assume nationwide that all white kids are similar. White kids in appalachia versus white kids at prep school on the upper east side, etc.

But when comparing the DC area schools that OP compared, it's fair enough to treat the racial demographic as a broad proxy. It's infortunate that this is appropriate, says a lot about racial inequality in DC. But it's appropriate.

I get what the first poster is saying. OP could have avoided any confusion by stating outright this analysis was only useful when comparing scores of wealthy white children, not all children.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:OP, there is a huge huge assumption underlying your post -- that demographic categories are monolithic. You are assuming that SES and cultural capital holds constant across the demographic groupings in the data. It's an interesting way to crunch the data but it's not standard for the ed. field.

I don't understand this criticism. OP is only assuming subgroups are similar across the schools. That is, white students at Deal are similar to white students at Pyle, Westland and Hardy, or that Asian students are comparable across these four schools. This is reasonable given the schools being compared are all within three miles or so of the DC-MD border. They're all within a four mile radius from Westmoreland Circle.

+1

It is not reasonable to assume nationwide that all white kids are similar. White kids in appalachia versus white kids at prep school on the upper east side, etc.

But when comparing the DC area schools that OP compared, it's fair enough to treat the racial demographic as a broad proxy. It's infortunate that this is appropriate, says a lot about racial inequality in DC. But it's appropriate.

I get what the first poster is saying. OP could have avoided any confusion by stating outright this analysis was only useful when comparing scores of wealthy white children, not all children.

No, this is the wrong take-away. The presentation is fine. The summary measures presented are a simple and reasonable aggregation of subgroup scores as if each school had the same subgroup distribution. So, if Martians score poorly on these measures (ignoring the reason) and school A has 75% Martians while school B has 10%, then of course school A will have worse overall scores. The analysis accounts for this by calculating what school A's overall scores would have been if they had only 10% Martians like school B.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:OP, there is a huge huge assumption underlying your post -- that demographic categories are monolithic. You are assuming that SES and cultural capital holds constant across the demographic groupings in the data. It's an interesting way to crunch the data but it's not standard for the ed. field.

I don't understand this criticism. OP is only assuming subgroups are similar across the schools. That is, white students at Deal are similar to white students at Pyle, Westland and Hardy, or that Asian students are comparable across these four schools. This is reasonable given the schools being compared are all within three miles or so of the DC-MD border. They're all within a four mile radius from Westmoreland Circle.

+1

It is not reasonable to assume nationwide that all white kids are similar. White kids in appalachia versus white kids at prep school on the upper east side, etc.

But when comparing the DC area schools that OP compared, it's fair enough to treat the racial demographic as a broad proxy. It's infortunate that this is appropriate, says a lot about racial inequality in DC. But it's appropriate.

I get what the first poster is saying. OP could have avoided any confusion by stating outright this analysis was only useful when comparing scores of wealthy white children, not all children.

Ha! Yes, it probably *is* appropriate to assume that the white kids in the analysis are comparable. Is that actually what the analysis is for? That hadn't actually occurred to me, but I can see how there would be many many people on DCUM for whom that would be the primary question in their minds. Ouch.

[Anonymous]

Anonymous wrote:Bumping this because it was linked in another thread and the analysis is excellent. I never saw it the first time and it's worth reading (analysis is on first page with some qualifiers afterward). Thank you OP!

+1000 - thank you OP.

It will be interesting to see the 2016 PARCC results in comparison to your analysis of the 2015 results. Although given the backlash you received, no one would blame you for never posting again any of your future analyses. But I hope you do because I think many of us would welcome your thoughts.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:OP, there is a huge huge assumption underlying your post -- that demographic categories are monolithic. You are assuming that SES and cultural capital holds constant across the demographic groupings in the data. It's an interesting way to crunch the data but it's not standard for the ed. field.

I don't understand this criticism. OP is only assuming subgroups are similar across the schools. That is, white students at Deal are similar to white students at Pyle, Westland and Hardy, or that Asian students are comparable across these four schools. This is reasonable given the schools being compared are all within three miles or so of the DC-MD border. They're all within a four mile radius from Westmoreland Circle.

+1

It is not reasonable to assume nationwide that all white kids are similar. White kids in appalachia versus white kids at prep school on the upper east side, etc.

But when comparing the DC area schools that OP compared, it's fair enough to treat the racial demographic as a broad proxy. It's infortunate that this is appropriate, says a lot about racial inequality in DC. But it's appropriate.

I get what the first poster is saying. OP could have avoided any confusion by stating outright this analysis was only useful when comparing scores of wealthy white children, not all children.

Ha! Yes, it probably *is* appropriate to assume that the white kids in the analysis are comparable. Is that actually what the analysis is for? That hadn't actually occurred to me, but I can see how there would be many many people on DCUM for whom that would be the primary question in their minds. Ouch.

What's your point? It's also safe to assume that the black kids are comparable across the MD-DC border as OP has done. And the asians, and the latinos. The point of OP's analysis is that in the DC region, for whatever reasons, there are racial differences in test scores. Race is a statistically powerful predictor of test score performance in our city. Therefore differences in racial composition across schools leads to misleading comparisons of aggregate test scores between schools. The OP has normalized the data to show us how the schools would perform if they all had the same set of students. And what we see is that Deal looks great on that measure. Which tells us that the instruction etc at Deal is just as good as say Pyle and the difference in test scores can be almost completely explained by differences in demographics.