2013_10-31_CEC1_WXY_AssignmentPolicyStudy FINAL with edits.pdf

7/27/2019 2013_10-31_CEC1_WXY_AssignmentPolicyStudy FINAL with edits.pdf

http://slidepdf.com/reader/full/201310-31cec1wxyassignmentpolicystudy-final-with-editspdf 1/68

George M. Janes Assoc



CONTENTS

2/ Executive findin s

3/ Assignment policy context

4/ Study area and period5/ Methodology

str ct n ngs



Understand the impacts of open enrollment policy in terms of stratification of“ ”, , ,

between the District 1 schools during the years 1999-2011.

Compare District 1 to another NYC District to better understand NYC trendsover the time period.



findin s

rom - , ere as een an ncrease o c us er ng o s u en s yrace and free lunch status, increasing each school’s dissimilarity from theother schools in the District. This increase has been most pronounceddurin the o en enrollment eriod.

rom - , ere s an ncrease n var a y e ween sc oo s nterms of student achievement as measured by Math and ELA scores.

,appears to be a sharp increase in the number of children with the District

area a en ng sc oo s e ne as up a e .

4/ When compared to other NYC schools districts that utilize zoned,

considerably and has a large proportion of public housing, thes ra ca on o sc oo s y race ree unc ac evemen scores an eincrease in “uptake” is noteworthy.



oliccontext

Vancouver Jefferson County

Boston

NYC Cape Town



oliccontext

u c c oo oc e y sso ves an merges w war sc oo s

1864/ Ward school system becomes the Board of Education district system

1911/ Woodrow Wilson endorses the School as Social Center movement

Brown v. Board of Education

1971/ Swann v. Charlotte-Mecklenburg Board of Education

2007/ Parents Involved in Communit Schools v. Seattle School District No. 1



oliccontext

As of Dec 2012, there are 37

UNZONED districts within NYC:

Lower East Side [D1] in 2007

Ocean Hill / Brownsville [D23] in 2012 1

23



and eriod

CEC1 Assignment Policy History

1991- 2003Communit School Board 1 removes zones and catchments in hases while uaranteein

placement of elementary school students in a school within 0.5 mile of their home, andinstitutes an o en enrollment olic with lotteries, usin race and ender uotas inoversubscribed schools.

2003DOE closes district offices and centralizes admissions, and the quota policy becomesunclear.

2007- 2013The DOE re uires CEC1 to move to a ure o en enrollment model without uotas.



and eriod

In SY 1999-2000:

MIDDLE SCHOOLS

0 CHARTER SCHOOLS

7,941 STUDENTS



and eriod

In SY 2011-2012:

MIDDLE SCHOOLS

2 CHARTER SCHOOLS

8,111 STUDENTS



SOURCES

Census 2000 + 2010 Demographics

Common Core of Data (NCES) EnrollmentRaceFree Lunch

NYC Department of Education EnrollmentRace

Free Lunch

epor ar Free Lunch



School aged children 5 to 13 years .

Race Asian + Pacific Islander Black + Two or MoreHispanic + Other + Two or MoreWhite

Socioeconomic Free Lunch

c evemen Math and ELA assessments

o e ementary age stu entswho attend public school within



• Data from 2001 – 2004 is unreliable, ,

and 2011.• - .

• In some cases, % for free lunch value was over 100% and,

sources

• Schools with completely missing free lunch are simply omitted from the analysis

• NYS DOE District totals for Free Lunch include data for high

• Total enrollment values were taken from two sources:



INDICES OF DISSIMILARITY

1/ The district has an overall racial and free lunch profile that changesevery year

2/ An Index of Dissimilarity measures how dissimilar the individual schools

are from the district-wide average

3/ Four races (Asian, Black, Hispanic and White) and Free Lunch studentsu

• easure y a ng e ean so u e ercen erence rom e s r c -w eaverage for each category

• -w v x u z• If the index increases, segregation is increasing

,




• Simple to explain and understand• -• Every school is scored every year, so year-to-year changes are

evident• Each ear is inde endent from other ears

• – ’ * • Some data missing, especially in the early years for race, 2002-

or ree unc , an e - sc oo year

* , , .



INDICES OF SCHOOL ACHIEVEMENT DISPARITY

Goal of school achievement analysisTo measure disparity in educational achievement over the course of the

a enges data that are publicly available for all years within the study period

Tracking input (resources for schooling) and and output (cognitiveper ormance o s u en s




1/ The index of achievement measures how dissimilar the individualsc oo s are rom e s r c -w e propor on o s u en s per orm ng agrade level

• Analyzed student performance on grades 4 and 8 English Language Arts andMathematics assessments

• These are the only grades that participated in NYS Testing program for all

• If all schools reflect the district-wide proportion of students meeting grade-level

• If the index increases, variability in student achievement is increasing

Note: The study found that there is a significant negative correlation

tests. However, it is not reasonable to describe these assessments as ameasure o ncome pover y.



UPTAKE AND ENROLLMENT

“ ”

• It is important to consider all children within District 1, not just. .story by considering how many children within D1 do not attend D1schools and what their demographics are)

ENROLLMENT FLUCTUATIONS• It is important to consider if there are changes to enrollment that

.



DISTRICTS COMPARISONS

COMPARING TO ANOTHER DISTRICT• We need to try to “control” for:•

district that has undergone similar changes (e.g. District 13)“ ”• v v

• Uptake changes over time



/ Indices of dissimilarity

/ Achievement metrics/ Capacity and uptake

utstan ng quest ons




findin sThere have been substantial changes in District 1 duringthe study period.

Number of Students by Race % of Students Receiving Free Lunch

His anic Asian Unknown / Other

WhiteBlack * Means there is substantial missing data this year



INDICES OF DISSIMILARITY A school’s Segregation Index is calculated by averaging the

findin s difference between the District and the school’s racial and free.

Total Racial Breakdown for District 1 from 1999-2000

Asian Black Hispanic White

15% 17% 61% 6%






Asian Black Hispanic White15% 17% 61% 6%

PS 140 Nathan Straus racial breakdown from 1999-2000

8% 13% 77% 2%







PS 140 Nathan Straus racial breakdown from 1999-2000 8% 13% 77% 2%

The absolute value of the difference between the District and PS 140

Nathan Straus school

7% 4% 16% 4%





Total Racial breakdown for District 1 from 1999-2000


PS 140 Nathan Straus racial breakdown from 1999-2000 8% 13% 77% 2%

The difference between the District and PS 140 Nathan Straus school

Asian Black Hispanic White+ + + = .

The average of the difference is . (rounded). This is the Index.



INDICES OF DISSIMILARITYThe Index for a District school can change if the school’s racial

findin s breakdown does not follow the overall District racial breakdown.


Asian Black His anic WhiteTotal Racial breakdown for District 1 from 2010-2011

Asian Black His anic White DISTRICT RAC 15% 17% 61% 6%

-

20% 17% 47% 16%

-

BREAKDOWN


8% 13% 77% 2%


4% 14% 80% 2%

The difference between the District and PS 140 Nathan Straus

Asian Black Hispanic WhiteThe difference between the District and PS 140 Nathan Straus


= . = .



INDICES OF DISSIMILARITYBoth Race and Free Lunch show an increase in segregation

findin s over time.

Racial Index of Dissimilarity by Year Free Lunch Index of Dissimilarity

1st Grade race segregation

7th Grade race segregation

All Grades * Means there is substantial missing data this year



INDICES OF DISSIMILARITYWhile a two schools stand out, the overall trend is positive,

findin s meaning more there has been an increase in racial clustering.

Racial Dissimilarity Index by School by Year



INDICES OF DISSIMILARITYWith the two outlier schools (NEST and PS 184 Shuang Wen)

findin s removed from the data, the trend is still positive and the slope of

Racial Dissimilarity Index by School by Year



INDICES OF DISSIMILARITYThe trend in segregation is not equal by race.

findin s

Dissimilarity Index by Race, All Grades

White and H

index row

Black and A

Asian

Black

Hispa



INDICES OF DISSIMILARITY As a percentage the white population index is increasing more

findin s than others in most schools.

SCHOOLS WITH THE HIGHEST WHITE INDEX

NEST (City wide G/T) 28%THE EAST VILLAGE COMMUNITY SCHOOL 25%

NEIGHBORHOOD SCHOOL 24%

CHILDREN'S WORKSHOP SCHOOL 24%

White Index by School by Year

EARTH SCHOOL 21%UNIVERSITY NEIGHBORHOOD MIDDLE SCHOOL 17%



INDICES OF DISSIMILARITYRemoving NEST and PS 184 Shuang Wen increases the slope

findin s of the line.SCHOOLS WITH THE HIGHEST WHITE INDEX

NEIGHBORHOOD SCHOOL 24%

CHILDREN'S WORKSHOP SCHOOL 24%

White Index by School by Year

UNIVERSITY NEIGHBORHOOD MIDDLE SCHOOL 17%

HENRY STREET SCHOOL FOR INTERNATIONAL STUDIES 15%



INDICES OF DISSIMILARITYNEST has a large impact on the numbers, and it is a City-wide

findin s school, so it was removed to understand its impact on the overall.

Racial Dissimilarity Index by Year, Excluding NEST Free Lunch Dissimilarity Index by Year, Excluding NES

PS184

Opens

* Means there is substantial missing data this year



ALL SCHOOLS IN DISTRICT 1 BY PERCENT WHPercentages more than one standard deviation from mean in red

findin s




findin s ere as een s gn can c ange n s r c sc oo s, no on y asbecome more diverse, but there are also many more schools in 2011than there were in 1999.

2/ With this change has also come more clustering of students by race and’ree unc s a us ncreas ng eac sc oo s ss m ar y rom e o ers n

the District.

period. Early increases are tied to the openings of PS184 and NEST,w c a rac e popu a ons eren an e s r c -w e mean. a erincreases were more broad-based. The increase in clustering is most

easily seen in the White population and to a lesser degree the Hispanic.

e ypo es s a open enro men as exacer a e segrega on canbe supported by this work.



SCHOOL ACHIEVEMENT METRICS

findin s

Achievement Variability Index by Year, Grade 4



SCHOOL ACHIEVEMENT METRICS

findin s

Achievement Variability Index by Year, Grade 8





findin s

FINDINGS

ere are s gn can ncreases n e s ra ca on e ween sc oo s e weehigher and lower performing schools (based on State-administeredassessments in English Language Arts and Mathematics) during the study

.

ncrease s ra ca on o serve or o gra e an gra e s u en s

3/ The observed stratification between schools cannot necessarily be causally.

and makeup of standardized tests or changing district demographic.





findin s

NEXT STEP

schools across the city for each of the District 1 schools. Then model theu v u

assignment policy, controlling for a variety of demographic/socioeconomicfactors.

NYC CAPACITY AND UPTAKE



NYC CAPACITY AND UPTAKEThere has been an increase in uptake of NYC schools during

findin s this time period as the number of children overall has fallen. The

about the same.

33% not in 24% not in

NYC DOE Schools

[330,092 students]

NYC DOE Schools

[203,805 students]

up a e

[658,821 students]

up a e

[639,557 students]

988,913 children aged 5 - 13 843,362 children aged 5 - 13

NYC CAPACITY AND UPTAKE



NYC CAPACITY AND UPTAKEWhite students make up over ½ of the students not in NYC DOE

findin s schools in 2010.

33% not in 24% not in

NYC DOE Schools

[330,092 students]

NYC DOE Schools

[203,805 students]


DISTRICT 1 CAPACITY AND UPTAKE



DISTRICT 1 CAPACITY AND UPTAKESince 2000, there has been a decrease in 5 – 13 aged children

findin s in District 1, which is similar to the overall NYC trend, but there.

[4,460 students]

[1,328 students]

up a e avg.

[6,850 students]

up a e

[7,137 students]

*


“leakage” to othwhere NYC is n




DISTRICT 1 CAPACITY AND UPTAKEWhile the overall number of District 1 students has decreased,

findin s there has been an increase in number of White, Asian, and Black.

11,310 children aged 5 - 13 8,465 children aged 5 - 13 11,310 children aged 5 - 13 8,465 children aged 5 - 13

White His anic0%5 n

38%470 not

435 in District63%751 not 100%

1,809 notin District

62%in Districtin District 3,703 in District

69% ,

0%

Black Asian

611,3

39%43%

6 notin District357 not

in District

847 notin District

1,184 inDistrict57%100%77%

, no

in District

, n s r c, n s r c




DISTRICT 1 CAPACITY AND UPTAKEThe makeup of students who do not attend District 1 schools has

findin s changed: in 2010, the White and Asian population largely make.

Total Aged 5-13 who

live within District 1

Total Aged 5-13 who

live within District 1

Black 14%Hispanic 51%

Black 15%Hispanic 44%

Am. Indian 1%

Am. Indian 1%

4,460 students]

1,328 students]


OUTSTANDING QUESTIONS




findin s

1/ Is the stratification reflective of the assignment policy change? Or is thisreflective of trends associated with gentrification and larger white

Are these trends reflective of neighborhood segregation trends, which

enerall show that non-His anic white o ulations tend to cluster inpredominantly white neighborhoods? (John Logan and Brian Stults, The Persistence

Is the increased uptake being driven by the assignment policy, the

changing demographics, or the improving achievement levels?



/ Why compare District 1 + 13?

/ Indices of dissimilarity/ Achievement metrics apac ty an upta e

u u

WHY COMPARE DISTRICT 1 + 13




findin s

In 2000, Average Median Household income was $32,381 and

$39,011 in District 1 and 13, respectively.

2000

n on quare

DUMBO

NoHo Alphabet City

Brooklyn Heights

Fort Greene

Clinton Hill

BedS

Prospect Heights

Under $25K

Lower East Side$25 – 55K

–




WHY COMPARE DISTRICT 1 + 13Between 2000 – 2010, median household income rose by 57%

findin s and 55%, respectively.

2010

n on quare

DUMBO

Alphabet City

Brooklyn Heights

NoHo

Fort Greene

Clinton Hill

BedS

Prospect Heights

Under $25K

Lower East Side$25 – 55K

–




WHY COMPARE DISTRICT 1 13In District 1, 18% of residents live in NYCHA housing

findin s developments, compared to 8% in District 13.

2011




WHY COMPARE DISTRICT 1 13The percentage of white population was 32% and 23% in District

findin s 1 and 13, respectively.

2000

n on quare

DUMBO

Alphabet City

Brooklyn Heights

NoHo

Fort Greene

Clinton Hill

BedS

Prospect Heights

o o a opu a on,

White

Less than 25%

Lower East Side25% – 50%

–




Between 2000 – 2010, the total white population rose by 93%

findin s and 90%, respectively.

2010

n on quare

DUMBO

Alphabet City

Brooklyn Heights

NoHo

Fort Greene

Clinton Hill

BedS

Prospect Heights

o o a opu a on,

White

Less than 25%

Lower East Side25% – 50%

–




District 13 is notable for its decline in enrollment.findin s

Number of Students by Race, District 13 % of Students Receiving Free Lunch

Asian Hispanic UnknownDistrict 1

Black e District 13




findin s

Racial Segregation Index by Year, District 1 + 13 Free Lunch Index, District 1 + 13




findin s

Segregation Index by Race, District 13

ACHIEVEMENT METRICS



The distribution of 4th grade achievement is comparatively stablefindin sfor District 13 schools

4th Grade ELA Achievement Variability, District 1 + 13 4th Grade Math Achievement Variability, District 1 +

ACHIEVEMENT METRICS



Stratification of 8th grade achievement scores actually declinesfindin sover the study period for District 13 schools

8th Grade ELA Achievement Variability, District 1 + 13 8th Grade Math Achievement Variability, District 1 +

ACHIEVEMENT METRICS



District 1 and 13 have had similar trajectories with respect to thefindin sproportion of students meeting the Grade 4 ELA and Math

Comparison of District 1 and District 13, 4th Grade Achievement Trajectories

ACHIEVEMENT METRICS



On both the 8th grade ELA and Math assessments, District 1findin sstudents outperform their District 13 counterparts and the gulf

.

Comparison of District 1 and District 13, 8th Grade

-

Comparison of District 1 and District 13, 8th Grade

-




Uptake has remained the same, even as overall numbers havefindin sdeclined.

43 not in District 41% not in District [9,723 students] [7,133 students]

up a e avg.

[13,084 students]

up a e

[10,043 students]

22,807 children aged 5 - 13 17176 children aged 5 - 13




Between 2000 – 2010, a significant number of Black families leftfindin sthe area, while White families moved in but chose not to attend

.


22,807 children aged 5 - 13 17176 children aged 5 - 13




uestions +

next steps

1/ Can we determine a causal relationship between D1’s stratification andthe change in assignment policy?

2/ Can we determine a causal relationship between D1’s uptake increase

an e c ange n ass gnmen po cy

3/ Based on the comparison to D13, does unzoned assignment policyprovide greater encouragement for uptake than zoned schools? Does this“ ”

of those neighborhoods?

4/ How does an assi nment olic have different effects in different “kinds”

of neighborhoods – gentrifying, gentrified, non-gentrified?

NEXT STEPS: FORMING AN APPROACH



uestions +

next steps

NEXT STEPS: FORMING AN APPROACH



uestions +

next steps



George M. Janes A


i



com arison


up a e avg.[11,753 students]

up a e [9,409 students]


DISTRICT 2 CAPACITY AND UPTAKEBetween 2000 to 2010 District 2 increased in number of schoolsi



Between 2000 to 2010, District 2 increased in number of schoolscom arisonfrom 39 to 108.

43% not in District16,798 students]

not n str ct853 students over the local student population

up a e 22,703 students]

104% u take 67% NYC av .[20,128 students]


DISTRICT 2 CAPACITY AND UPTAKEBetween 2000 to 2010 District 2 increased in number of schoolscom arison



Between 2000 to 2010, District 2 increased in number of schoolscom arisonfrom 39 to 108.

43% not in District16,798 students]

not n str ct853 students over the local student population

up a e 22,703 students]

104% u take 67% NYC av .[20,128 students]


School District



School District

Racial Profile

1990, District

3 Census*

1990,

School

2000, District

3 Census*

2000,

School2010, District

3 Census*

2010,

School

Race n er : : s r c : n er : s r c - : n er : s r c -

American Indian 0.2% 79 0.1% 9 0.2% 48 0.8% 52 0.2% 41 0.8%

s an ac c s an er . , . . , . , . , .

Black 10.0% 3,223 20.9% 2,098 10.1% 2,845 17.5% 1,198 10.6% 2,344 16.6%

. , . , . , . , . , .

White 9.6% 3,108 5.4% 540 9.4% 2,631 6.4% 435 13.2% 2,930 16.5%

, , , , ,

Source: 1New York City Department of City Planning , Total Population Under 18 by Mutual2 National Center for Education Statistics, Common Core of Data

2013_10-31_CEC1_WXY_AssignmentPolicyStudy FINAL with edits.pdf

Documents

Transcript of 2013_10-31_CEC1_WXY_AssignmentPolicyStudy FINAL with edits.pdf