PARCC data is up

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:
I was trying to find data on how multiple sub-groups fared - at-risk male, black male, etc - and came across figures I hadn't seen before for at-risk whites in DC. There were 81 at-risk whites reported in the 2018 public school report card data. 28 had a median growth percentile for math (MGP). For the 28, their MGP for math was 27. For 'not at-risk white', their MGP was 61. At-risk makes a huge difference on performance - no surprise there.

Do you have this data (crosstab of race and/or ethnicity and at-risk city-wide)? I think it's probably not meaningful at the school level because of the small sample size, but I would (especially) like to see if there's a way to tease out something from the citywide data to describe the interaction of the "at-risk" designation and race.

You can get a lot of data on the OSSE website. At this point in the year, the DC Schools Report Card stuff is pretty out of date so I would not rely on it.

I would also caution you that not all PARCC scores are reported. Kids have to have been at the school for most of the year to be reported in this data set. At-risk kids are more mobile so over-represented in that non-reported group. So many schools are more at-risk than they seem from this data. Also, MSAA results are not in and at-risk kids might be overrepresented in MSAA takers. Just some things to be aware of.

Helpful to know that the number of at-risk kids is likely understated in that population - thanks. I'm more interested in blunt ways to describe the data rather than perfect instruments. And I would REALLY like to see some analysis/reporting that describes black kids that don't fall into the at-risk definition - there are more kids that fall into this category than there are white kids in the entire city, and I think the way the data is reported ignores them.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:
I was trying to find data on how multiple sub-groups fared - at-risk male, black male, etc - and came across figures I hadn't seen before for at-risk whites in DC. There were 81 at-risk whites reported in the 2018 public school report card data. 28 had a median growth percentile for math (MGP). For the 28, their MGP for math was 27. For 'not at-risk white', their MGP was 61. At-risk makes a huge difference on performance - no surprise there.

Do you have this data (crosstab of race and/or ethnicity and at-risk city-wide)? I think it's probably not meaningful at the school level because of the small sample size, but I would (especially) like to see if there's a way to tease out something from the citywide data to describe the interaction of the "at-risk" designation and race.

You can get a lot of data on the OSSE website. At this point in the year, the DC Schools Report Card stuff is pretty out of date so I would not rely on it.

I would also caution you that not all PARCC scores are reported. Kids have to have been at the school for most of the year to be reported in this data set. At-risk kids are more mobile so over-represented in that non-reported group. So many schools are more at-risk than they seem from this data. Also, MSAA results are not in and at-risk kids might be overrepresented in MSAA takers. Just some things to be aware of.

Helpful to know that the number of at-risk kids is likely understated in that population - thanks. I'm more interested in blunt ways to describe the data rather than perfect instruments. And I would REALLY like to see some analysis/reporting that describes black kids that don't fall into the at-risk definition - there are more kids that fall into this category than there are white kids in the entire city, and I think the way the data is reported ignores them.

I think you could pull that out by subtracting other races, but you would be confounded by the two or more races category. Also so many schools have n less than 10 in various categories so I dunno, I guess you could just fill in 5s or something.

OSSE has enrollment audit data, which is done in October. It might be interesting to compare that data with the smaller group of kids whose PARCC scores are reported.

[Anonymous]

Also be aware of wide variation in special needs and ELL populations. Some schools operate specialized classrooms for autism, behavior support, etc. Some schools refuse to. Some schools push out the harder kids so their at-risk population is the easier group of at-risk kids. Some schools have the challenge of mid-year entries, others refuse to share in that work.

[Anonymous]

Anonymous wrote:Also be aware of wide variation in special needs and ELL populations. Some schools operate specialized classrooms for autism, behavior support, etc. Some schools refuse to. Some schools push out the harder kids so their at-risk population is the easier group of at-risk kids. Some schools have the challenge of mid-year entries, others refuse to share in that work.

Yep - I don't think you could use that crosstab data to describe school-level results with any degree of accuracy due to small sample size, population variation between schools, etc. But I do think that there are a bunch of middle and upper middle class black kids (and parents) in DC who are looking at achievement gaps in their schools and citywide and wondering how schools are serving kids like theirs.

[Anonymous]

Anonymous wrote:

Anonymous wrote:Also be aware of wide variation in special needs and ELL populations. Some schools operate specialized classrooms for autism, behavior support, etc. Some schools refuse to. Some schools push out the harder kids so their at-risk population is the easier group of at-risk kids. Some schools have the challenge of mid-year entries, others refuse to share in that work.

Yep - I don't think you could use that crosstab data to describe school-level results with any degree of accuracy due to small sample size, population variation between schools, etc. But I do think that there are a bunch of middle and upper middle class black kids (and parents) in DC who are looking at achievement gaps in their schools and citywide and wondering how schools are serving kids like theirs.

Totally agree. OSSE might give you the data if you asked.

I am really struck by the number of supposedlly HRCS that have an achievement gap despite having low at-risk and also a substantial population of non-at-risk AA kids, some who are not even low-income.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:Also be aware of wide variation in special needs and ELL populations. Some schools operate specialized classrooms for autism, behavior support, etc. Some schools refuse to. Some schools push out the harder kids so their at-risk population is the easier group of at-risk kids. Some schools have the challenge of mid-year entries, others refuse to share in that work.

Yep - I don't think you could use that crosstab data to describe school-level results with any degree of accuracy due to small sample size, population variation between schools, etc. But I do think that there are a bunch of middle and upper middle class black kids (and parents) in DC who are looking at achievement gaps in their schools and citywide and wondering how schools are serving kids like theirs.

Totally agree. OSSE might give you the data if you asked.

I am really struck by the number of supposedlly HRCS that have an achievement gap despite having low at-risk and also a substantial population of non-at-risk AA kids, some who are not even low-income.

Yep, me too. But if close to half the black kids at a school in the testing grades are at-risk and <2% of the white kids are at-risk (which is kinda roughly how my back of the envelope calculations show it would play out), that is really significant.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:Also be aware of wide variation in special needs and ELL populations. Some schools operate specialized classrooms for autism, behavior support, etc. Some schools refuse to. Some schools push out the harder kids so their at-risk population is the easier group of at-risk kids. Some schools have the challenge of mid-year entries, others refuse to share in that work.

Yep - I don't think you could use that crosstab data to describe school-level results with any degree of accuracy due to small sample size, population variation between schools, etc. But I do think that there are a bunch of middle and upper middle class black kids (and parents) in DC who are looking at achievement gaps in their schools and citywide and wondering how schools are serving kids like theirs.

Totally agree. OSSE might give you the data if you asked.

I am really struck by the number of supposedlly HRCS that have an achievement gap despite having low at-risk and also a substantial population of non-at-risk AA kids, some who are not even low-income.

Yep, me too. But if close to half the black kids at a school in the testing grades are at-risk and <2% of the white kids are at-risk (which is kinda roughly how my back of the envelope calculations show it would play out), that is really significant.

Yes. And sometimes the non-at-risk AA kids are not performing very well. Why??? Schools should be prepared to answer.

[Anonymous]

One statistic that isn't reported anywhere is the correlation between two parent households and PARCC scores. I have no idea whether there is a statistical significance but I bet there probably is---but no one has tracked it. I have observed that my children's friends who are coming from split HH have less consistency in study habits (i.e., one parent is lax while the other is more vigilant). Instead of focusing on racial/ethnic differences, it might be more informative to know about family structures.

[Anonymous]

Anonymous wrote:One statistic that isn't reported anywhere is the correlation between two parent households and PARCC scores. I have no idea whether there is a statistical significance but I bet there probably is---but no one has tracked it. I have observed that my children's friends who are coming from split HH have less consistency in study habits (i.e., one parent is lax while the other is more vigilant). Instead of focusing on racial/ethnic differences, it might be more informative to know about family structures.

One parent vs two parent household using boils down to SES because single mother headed households are the poorest in the US.

[Anonymous]

Anonymous wrote:One statistic that isn't reported anywhere is the correlation between two parent households and PARCC scores. I have no idea whether there is a statistical significance but I bet there probably is---but no one has tracked it. I have observed that my children's friends who are coming from split HH have less consistency in study habits (i.e., one parent is lax while the other is more vigilant). Instead of focusing on racial/ethnic differences, it might be more informative to know about family structures.

It is hard to talk about because defensive divorced people flip out on you.

[Anonymous]

well, you are literally judging their life choices if you are doing what's in the message chain above.

[Anonymous]

Anonymous wrote:

Anonymous wrote:One statistic that isn't reported anywhere is the correlation between two parent households and PARCC scores. I have no idea whether there is a statistical significance but I bet there probably is---but no one has tracked it. I have observed that my children's friends who are coming from split HH have less consistency in study habits (i.e., one parent is lax while the other is more vigilant). Instead of focusing on racial/ethnic differences, it might be more informative to know about family structures.

It is hard to talk about because defensive divorced people flip out on you.

As a divorced adult, it’s understandable. But the reality is if you are a divorced mother, you are likely in poverty and statistically your children won’t be as successful

[Anonymous]

Anonymous wrote:One statistic that isn't reported anywhere is the correlation between two parent households and PARCC scores. I have no idea whether there is a statistical significance but I bet there probably is---but no one has tracked it. I have observed that my children's friends who are coming from split HH have less consistency in study habits (i.e., one parent is lax while the other is more vigilant). Instead of focusing on racial/ethnic differences, it might be more informative to know about family structures.

Yes, I've seen this play out several times with friends even within high SES households. It's very hard to keep track of assignments, exams etc when there are two households involved. I've seen two parents with wildly different household rules---mom enforces study time and dad allows unlimited screen time (or vice verse) . Or this even used as a weapon in adult fights--dad allows kids' grades to tank as a way of getting back at mom. One of my good friends is going through this last scenario right now--her ex husband is using their rising 10th grader (and refusing to enforce studying) as a pawn to get back at her. Adults can be jerks.

And then you have households where only one parent is involved or primarily only one parent is involved. I imagine it's exceedingly difficult to be on top of a kid's study habits if you're doing it all.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:Also be aware of wide variation in special needs and ELL populations. Some schools operate specialized classrooms for autism, behavior support, etc. Some schools refuse to. Some schools push out the harder kids so their at-risk population is the easier group of at-risk kids. Some schools have the challenge of mid-year entries, others refuse to share in that work.

Yep - I don't think you could use that crosstab data to describe school-level results with any degree of accuracy due to small sample size, population variation between schools, etc. But I do think that there are a bunch of middle and upper middle class black kids (and parents) in DC who are looking at achievement gaps in their schools and citywide and wondering how schools are serving kids like theirs.

Totally agree. OSSE might give you the data if you asked.

I am really struck by the number of supposedlly HRCS that have an achievement gap despite having low at-risk and also a substantial population of non-at-risk AA kids, some who are not even low-income.

Yep, me too. But if close to half the black kids at a school in the testing grades are at-risk and <2% of the white kids are at-risk (which is kinda roughly how my back of the envelope calculations show it would play out), that is really significant.

Yes. And sometimes the non-at-risk AA kids are not performing very well. Why??? Schools should be prepared to answer.

At our school it's the white kids who are performing lower than the average. And most kids are white. So what does that say? The other races are doing better than expected.

[Anonymous]

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:Also be aware of wide variation in special needs and ELL populations. Some schools operate specialized classrooms for autism, behavior support, etc. Some schools refuse to. Some schools push out the harder kids so their at-risk population is the easier group of at-risk kids. Some schools have the challenge of mid-year entries, others refuse to share in that work.

Yep - I don't think you could use that crosstab data to describe school-level results with any degree of accuracy due to small sample size, population variation between schools, etc. But I do think that there are a bunch of middle and upper middle class black kids (and parents) in DC who are looking at achievement gaps in their schools and citywide and wondering how schools are serving kids like theirs.

Totally agree. OSSE might give you the data if you asked.

I am really struck by the number of supposedlly HRCS that have an achievement gap despite having low at-risk and also a substantial population of non-at-risk AA kids, some who are not even low-income.

Yep, me too. But if close to half the black kids at a school in the testing grades are at-risk and <2% of the white kids are at-risk (which is kinda roughly how my back of the envelope calculations show it would play out), that is really significant.

Yes. And sometimes the non-at-risk AA kids are not performing very well. Why??? Schools should be prepared to answer.

At our school it's the white kids who are performing lower than the average. And most kids are white. So what does that say? The other races are doing better than expected.

I have no clue. Does grade level analysis reveal any patterns?