MAP percentile cutoff for MS magnet lottery?

[Anonymous]

Anonymous wrote:

Anonymous wrote:They could also change the FARMS bands. I think one year they had 5 and then they changed it to 3. There's a lot of ways to manipulate the numbers to get your desired result.

The one strength I need to give them credit for is their amazing ability to manipulate the data to produce their desired results. Heaven forbid they give every student the same test (which they do) and use the data itself to determine who is best prepared for the program. No, they must come up with very complex algorithms to allocate seats away from students with outlier performance and toward students with mediocre scores.

That's because MAP performance is influenced largely by exposure, while they seek expoaure-neutral ability for magnet programs while utilizing MAP for expedience. It is far from the fidelity to capability sought (in many ways), but, when implemented correctly, local norming is a recommended practice to try to account for that disconnect, where teachers in low-FARMS schools are more routinely able to manage cohorts to provide enrichment/exposure to those able than those in high-FARMS schools. This is independent of any more socio-political aim, but it could very well serve that purpoae, as well.

Cue the diingenuous CogAT-is-more-gameable-than-MAP poster...

[Anonymous]

Anonymous wrote:

Anonymous wrote:They could also change the FARMS bands. I think one year they had 5 and then they changed it to 3. There's a lot of ways to manipulate the numbers to get your desired result.

Actually, I think that we all speculated that there were three because we thought it would go title one, focus schools, everyone else. When the freedom of information request finally went in, it turns out there were five and our speculation had been wrong all along.

I was also surprised to see how few schools were in the highest FARMS band. These are not mixed income schools, they are schools were almost every single kid qualifies for free and reduced meals. I think it alleviates some of the complaints about middle class families in high needs schools scooping up all of the spots.

And the greatest local norming differences were seen in the two highest-FARMS groups, with low- and low-moderare-FARMS groups pretty close.

[Anonymous]

Anonymous wrote:

Anonymous wrote:They could also change the FARMS bands. I think one year they had 5 and then they changed it to 3. There's a lot of ways to manipulate the numbers to get your desired result.

Actually, I think that we all speculated that there were three because we thought it would go title one, focus schools, everyone else. When the freedom of information request finally went in, it turns out there were five and our speculation had been wrong all along.

I was also surprised to see how few schools were in the highest FARMS band. These are not mixed income schools, they are schools were almost every single kid qualifies for free and reduced meals. I think it alleviates some of the complaints about middle class families in high needs schools scooping up all of the spots.

That's not exactly true. They revealed in the first year they used these bands but not in a lottery there were a certain number of bands. Then the next year they did a lottery and then they changed the number of bands to make more or less. I'm sure someone ran spreadsheets of the results a number of different ways and they used the one that achieved their desired mix of URMs and ever FARMS students.

[Anonymous]

The problem with using MAP from what we've seen is that the small number of more middle class or wealthier kids at high FARMS schools will be at an advantage over the kids who are actually FARMS so you are missing the actual smart kids with a lot of potential.

I know Cogat is not perfect but it would capture more of those outlier smart kids who with the influence of strong peers and strong teachers could be really successful.

I would bet anything the local norming if you use Cogat would be unnecessary or at least nowhere as intense as it is for MAP. There would be more of an even distribution throughout the county.

[Anonymous]

Anonymous wrote:The problem with using MAP from what we've seen is that the small number of more middle class or wealthier kids at high FARMS schools will be at an advantage over the kids who are actually FARMS so you are missing the actual smart kids with a lot of potential.

I know Cogat is not perfect but it would capture more of those outlier smart kids who with the influence of strong peers and strong teachers could be really successful.

I would bet anything the local norming if you use Cogat would be unnecessary or at least nowhere as intense as it is for MAP. There would be more of an even distribution throughout the county.

This gets at what folks were discussing above. If you actually look at the list of schools and the associated scores, the schools where kids get the biggest boost is a much shorter list and are schools with functionally no middle class children.

So, the kids that get the biggest boost are coming from schools that are profoundly disadvantaged.

The bulk of the schools are in one of the mid levels and the swing between required scores is not actually that large in that middle band.

[Anonymous]

Anonymous wrote:The problem with using MAP from what we've seen is that the small number of more middle class or wealthier kids at high FARMS schools will be at an advantage over the kids who are actually FARMS so you are missing the actual smart kids with a lot of potential.

I know Cogat is not perfect but it would capture more of those outlier smart kids who with the influence of strong peers and strong teachers could be really successful.

I would bet anything the local norming if you use Cogat would be unnecessary or at least nowhere as intense as it is for MAP. There would be more of an even distribution throughout the county.

The idea is that those kids don't get the exposure that their equally-able peers at low-FARMS schools get, so the local norming adjusts for that. Those more directly disadvantaged are boosted by the lower percentile threshold allowed for those receiving services, including individual FARMS designation. A FARMS student at a moderate-FARMS school might get in with a 73rd national percentile score, where their non-FARMS classmate might need to hit the 88th percentile. It ain't perfect, of course.

[Anonymous]

The kids we know who got in to the MS magnets from those high FARMS schools are mostly white and they are from middle class or upper middle class families.

[Anonymous]

Anonymous wrote:The kids we know who got in to the MS magnets from those high FARMS schools are mostly white and they are from middle class or upper middle class families.

You know upper middle class white kids being admitted to middle school magnets from schools like Leleck Elementary, which has a 95% farms rate?

Because those are the schools that are in the highest farms bracket, schools were there are almost no kids who don't receive free and reduced lunch.

[Anonymous]

Anonymous wrote:

Anonymous wrote:The problem with using MAP from what we've seen is that the small number of more middle class or wealthier kids at high FARMS schools will be at an advantage over the kids who are actually FARMS so you are missing the actual smart kids with a lot of potential.

I know Cogat is not perfect but it would capture more of those outlier smart kids who with the influence of strong peers and strong teachers could be really successful.

I would bet anything the local norming if you use Cogat would be unnecessary or at least nowhere as intense as it is for MAP. There would be more of an even distribution throughout the county.

The idea is that those kids don't get the exposure that their equally-able peers at low-FARMS schools get, so the local norming adjusts for that. Those more directly disadvantaged are boosted by the lower percentile threshold allowed for those receiving services, including individual FARMS designation. A FARMS student at a moderate-FARMS school might get in with a 73rd national percentile score, where their non-FARMS classmate might need to hit the 88th percentile. It ain't perfect, of course.

The idea is a mix of 2 things: they want to tilt the demographic mix to get more non-Asian PoC, and they want kids who lack a local cohort to go to a school with a cohort. FARMS is a proxy for PoC and local cohort performance. It’s not about identifying high potential students who somehow avoided learning the grade-level material in their home school but would magically learn more by skipping a year of math and joining an accelerated class.

[Anonymous]

Anonymous wrote:

Anonymous wrote:The kids we know who got in to the MS magnets from those high FARMS schools are mostly white and they are from middle class or upper middle class families.

You know upper middle class white kids being admitted to middle school magnets from schools like Leleck Elementary, which has a 95% farms rate?

Because those are the schools that are in the highest farms bracket, schools were there are almost no kids who don't receive free and reduced lunch.

...or Watkins Mill ES, which has so few white kids that they don't even show up in the school demographic count?

I'd really encourage you to look at the list of schools deemed "high FARMS" because it is much shorter than you seem to think it is. Actually mixed income schools that might have some middle class kids are in other brackets, which are much larger than the "high FARMS" bracket.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:The problem with using MAP from what we've seen is that the small number of more middle class or wealthier kids at high FARMS schools will be at an advantage over the kids who are actually FARMS so you are missing the actual smart kids with a lot of potential.

I know Cogat is not perfect but it would capture more of those outlier smart kids who with the influence of strong peers and strong teachers could be really successful.

I would bet anything the local norming if you use Cogat would be unnecessary or at least nowhere as intense as it is for MAP. There would be more of an even distribution throughout the county.

The idea is that those kids don't get the exposure that their equally-able peers at low-FARMS schools get, so the local norming adjusts for that. Those more directly disadvantaged are boosted by the lower percentile threshold allowed for those receiving services, including individual FARMS designation. A FARMS student at a moderate-FARMS school might get in with a 73rd national percentile score, where their non-FARMS classmate might need to hit the 88th percentile. It ain't perfect, of course.

The idea is a mix of 2 things: they want to tilt the demographic mix to get more non-Asian PoC, and they want kids who lack a local cohort to go to a school with a cohort. FARMS is a proxy for PoC and local cohort performance. It’s not about identifying high potential students who somehow avoided learning the grade-level material in their home school but would magically learn more by skipping a year of math and joining an accelerated class.

They may want that tilt, but you're way off on your reasoning with respect to MAP. The more exposure, whether on-grade enrichment, above-grade acceleration (especially) or merely covering modules that others must skip due to the time taken with the local cohort to get through the basics, the higher a MAP score is likely to be. Put simply, they would learn more.

Again. It isn't perfect, and it should be changed to achieve greater fidelity to ability, but it is considerably better than the results without the local norming adjustment. That is, as long as better is about identifying capability, and not whether the happenstance of one's census tract allowed for exposure.

To be clear, I don't think there's anything wrong with providing that exposure, and would encourage it to whatever extent is manageable. I do think, however, that a public-good system should strive for meaningful equal protection and avoid the reinforcement of advantage, where possible.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:They could also change the FARMS bands. I think one year they had 5 and then they changed it to 3. There's a lot of ways to manipulate the numbers to get your desired result.

The one strength I need to give them credit for is their amazing ability to manipulate the data to produce their desired results. Heaven forbid they give every student the same test (which they do) and use the data itself to determine who is best prepared for the program. No, they must come up with very complex algorithms to allocate seats away from students with outlier performance and toward students with mediocre scores.

That's because MAP performance is influenced largely by exposure, while they seek expoaure-neutral ability for magnet programs while utilizing MAP for expedience. It is far from the fidelity to capability sought (in many ways), but, when implemented correctly, local norming is a recommended practice to try to account for that disconnect, where teachers in low-FARMS schools are more routinely able to manage cohorts to provide enrichment/exposure to those able than those in high-FARMS schools. This is independent of any more socio-political aim, but it could very well serve that purpoae, as well.

Cue the diingenuous CogAT-is-more-gameable-than-MAP poster...

I mean, the curriculum is supposed to be the same across all schools in primary grades. If kids are not being exposed to the same curricular material, and that is influencing students’ standardized test performance, I think the solution is to work on fixing that inequity rather than manipulate scores based on low expectations for students at higher FARMS schools.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:The problem with using MAP from what we've seen is that the small number of more middle class or wealthier kids at high FARMS schools will be at an advantage over the kids who are actually FARMS so you are missing the actual smart kids with a lot of potential.

I know Cogat is not perfect but it would capture more of those outlier smart kids who with the influence of strong peers and strong teachers could be really successful.

I would bet anything the local norming if you use Cogat would be unnecessary or at least nowhere as intense as it is for MAP. There would be more of an even distribution throughout the county.

The idea is that those kids don't get the exposure that their equally-able peers at low-FARMS schools get, so the local norming adjusts for that. Those more directly disadvantaged are boosted by the lower percentile threshold allowed for those receiving services, including individual FARMS designation. A FARMS student at a moderate-FARMS school might get in with a 73rd national percentile score, where their non-FARMS classmate might need to hit the 88th percentile. It ain't perfect, of course.

The idea is a mix of 2 things: they want to tilt the demographic mix to get more non-Asian PoC, and they want kids who lack a local cohort to go to a school with a cohort. FARMS is a proxy for PoC and local cohort performance. It’s not about identifying high potential students who somehow avoided learning the grade-level material in their home school but would magically learn more by skipping a year of math and joining an accelerated class.

That's a strange and twusted view on this. It would be more accurate to say they want to ensure that all residents have a shot at these programs not just ones from a few wealthy schools.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:They could also change the FARMS bands. I think one year they had 5 and then they changed it to 3. There's a lot of ways to manipulate the numbers to get your desired result.

The one strength I need to give them credit for is their amazing ability to manipulate the data to produce their desired results. Heaven forbid they give every student the same test (which they do) and use the data itself to determine who is best prepared for the program. No, they must come up with very complex algorithms to allocate seats away from students with outlier performance and toward students with mediocre scores.

That's because MAP performance is influenced largely by exposure, while they seek expoaure-neutral ability for magnet programs while utilizing MAP for expedience. It is far from the fidelity to capability sought (in many ways), but, when implemented correctly, local norming is a recommended practice to try to account for that disconnect, where teachers in low-FARMS schools are more routinely able to manage cohorts to provide enrichment/exposure to those able than those in high-FARMS schools. This is independent of any more socio-political aim, but it could very well serve that purpoae, as well.

Cue the diingenuous CogAT-is-more-gameable-than-MAP poster...

I mean, the curriculum is supposed to be the same across all schools in primary grades. If kids are not being exposed to the same curricular material, and that is influencing students’ standardized test performance, I think the solution is to work on fixing that inequity rather than manipulate scores based on low expectations for students at higher FARMS schools.

The curriculum is the same. I think one difference is the amount that many of the wealthier families spend on tutoring and outside enrichment, which tilts things in their favor. Adjusting the scores to reflect differences in privilege seems like a reasonable concession to fairness.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:They could also change the FARMS bands. I think one year they had 5 and then they changed it to 3. There's a lot of ways to manipulate the numbers to get your desired result.

The one strength I need to give them credit for is their amazing ability to manipulate the data to produce their desired results. Heaven forbid they give every student the same test (which they do) and use the data itself to determine who is best prepared for the program. No, they must come up with very complex algorithms to allocate seats away from students with outlier performance and toward students with mediocre scores.

That's because MAP performance is influenced largely by exposure, while they seek expoaure-neutral ability for magnet programs while utilizing MAP for expedience. It is far from the fidelity to capability sought (in many ways), but, when implemented correctly, local norming is a recommended practice to try to account for that disconnect, where teachers in low-FARMS schools are more routinely able to manage cohorts to provide enrichment/exposure to those able than those in high-FARMS schools. This is independent of any more socio-political aim, but it could very well serve that purpoae, as well.

Cue the diingenuous CogAT-is-more-gameable-than-MAP poster...

I mean, the curriculum is supposed to be the same across all schools in primary grades. If kids are not being exposed to the same curricular material, and that is influencing students’ standardized test performance, I think the solution is to work on fixing that inequity rather than manipulate scores based on low expectations for students at higher FARMS schools.

I have had kids at true high FARMS schools and medium FARMS, and that idea seems naive. The curriculum is the same, but the FARMS kids answers questions based only on lesson material. MC/UMC kids answer questions accurately but with a wealth of background knowledge, which the teacher then explains to the whole class. So the FARMS kid meets standards but the other classroom winds up a brief tangent about other concepts and knowledge, so the exposure is absolutely different.

And that’s without getting into issues that still exist in things like EC offerings after school - it’s not economical for companies to offer classes etc if no one at the school can afford to join them. And classroom libraries differ widely based on SES status.

TLDR: money matters, so kids in high FARMS schools are at a disadvantage that can’t easily be remedied.