© 2016 THE EDUCATION TRUST Copyright 2016 The Education Trust Using Accountability to Drive Equity Risks and Opportunities in ESSA Daria Hall March 8, 2016
© 2016 THE EDUCATION TRUST
Copyright 2016 The Education Trust
Using Accountability to Drive EquityRisks and Opportunities in ESSA
Daria Hall
March 8, 2016
© 2016 THE EDUCATION TRUST
Goals for our time together• Build a shared understanding of what
accountability means and why it’s important
• Share principles for strong, equity-focused accountability
• Identify the opportunities and risks that ESSA poses for state accountability
© 2016 THE EDUCATION TRUST
What do we mean by “accountability”?
Policies for measuring and holding schools responsible for important outcomes, rewarding schools that are serving all students well, and prompting improvement in those that aren’t.
© 2016 THE EDUCATION TRUST
To be clear: policies themselves don’t close gaps and raise achievement.
Only the hard work of educators, students, and parents can do that.
© 2016 THE EDUCATION TRUST
But well-designed accountability systems:
• Set a clear expectation that schools must raise the achievement of all of their students, not just some
• Focus attention and resources on the full range of student
groups, including those who are sometimes ignored
• Prompt action when schools don’t meet expectations for any group of students.
© 2016 THE EDUCATION TRUST
• Accountability is not about sorting and punishing.
• It’s about making sure that schools and systems are responsible for the performance of all students, and taking action when any group falls behind. – Strong local leaders need cover to do this. Others
need pressure.
© 2016 THE EDUCATION TRUST
And why does this matter?
© 2016 THE EDUCATION TRUST
The face of our population is changing --- fast
15%
96%
137%
50%
-9%
Projected Population Growth, Ages 20-24, 2010-2050
Note: Projected Population Growth, Ages 0–24, 2010-2050
Percentage Increase, Ages 0–24
Source: National Population Projections, U.S. Census Bureau. Released 2008; NCHEMS, Adding It Up, 2007.
© 2016 THE EDUCATION TRUST
But on just about every measure, the student populations that are growing the fastest are less prepared for
success in college and the workforce than their peers
© 2016 THE EDUCATION TRUST
African Amer-ican
Latino American Indian
White Asian0%
20%
40%
60%
80%
100%
34%
47%39%
75% 75%
19%
31% 26%
56% 57%
14%
29%
20%
52%
69%
12%
23%18%
48%
57%
English Reading Mathematics SciencePerc
ent o
f ACT
-tes
ted
grad
uate
s mee
ting
colle
ge
read
ines
s ben
chm
arks
Note: College readiness benchmarks are ACT-established thresholds that represent the score that a student needs to attain in order to have at least a 50% chance of receiving a B and a 75% chance of receiving a C in corresponding first-year college courses.
Source: The Condition of College & Career Readiness 2015, ACThttps://www.act.org/research/policymakers/cccr15/pdf/CCCR15-NationalReadinessRpt.pdf.
Latino, African American, and Native test takers are less likely to meet ACT college readiness benchmarks in any subject
© 2016 THE EDUCATION TRUST
African American Latino White0%
20%
40%
60%
80%
100%
39%
21%14%
Remediation rate at 4-year colleges
Perc
ent o
f stu
dent
s
Note: National estimates based on the calculated medians of data self-reported by 33 states.
Source: Complete College America, Remediation: Higher education’s bridge to nowhere, 2012, http://www.completecollege.org/docs/CCA-Remediation-final.pdf.
Students of color far more likely to enroll in remedial courses at four-year colleges
© 2016 THE EDUCATION TRUST
African American Latino White0%
20%
40%
60%
80%
100%
68%58%
47%
Remediation rate at 2-year colleges
Perc
ent o
f stu
dent
s
Note: National estimates based on the calculated medians of data self-reported by 33 states.
Source: Complete College America, Remediation: Higher education’s bridge to nowhere, 2012, http://www.completecollege.org/docs/CCA-Remediation-final.pdf.
Most students of color at two-year colleges need to take remedial courses in at least some subjects
© 2016 THE EDUCATION TRUST© 2016 THE EDUCATION TRUST
Young adults from high-income families are 7 times more likely to earn bachelor’s degrees by age 24
Source: Tom Mortenson, Bachelor’s Degree Attainment by age 24 by Family income Quartiles, 1970 to 2010, Postsecondary Education Opportunity, 2012.
20100%
10%
20%
30%
40%
50%
60%
70%
80%
90%
11%
79%
Lowest Income Quartile Highest Income Quartile
Bac
helo
r’s D
egre
e at
tain
men
t by
Age
24
© 2016 THE EDUCATION TRUST
Students Overall African American Latino White0%
20%
40%
60%
80%
100%
23%
39%
29%
16%
ASVAB Ineligibility Rates
Perc
ent o
f app
lican
ts n
ot m
eetin
g m
inim
um
acad
emic
ent
ry re
quire
men
ts, 2
004-
2009
Note: Sample is composed of young high school graduates, ages 17-20, who applied for entry into the Army between 2004 and 2009 and who took the ASVAB at a Military Entrance Processing Station, and are not a representative sample of all high school graduates.
Source: Christina Theokas, Shut Out of the Military: Today’s high school education doesn’t mean you’re ready for today’s Army,” The Education Trust, 2010, http://www.edtrust.org/sites/edtrust.org/files/publications/files/ASVAB_4.pdf.
Students of color far more likely to fail to meet minimum academic requirements for Army enlistment
© 2016 THE EDUCATION TRUST
Source: Rising to the Challenge: College Instructors’ Views on High School Graduates’ Preparedness for College , Achieve, 2015
Over 80% of employers report that recent public high school graduates have gaps in preparation for typical jobs in their company
5%
13%
34%
31%
17%
Extremely Well Prepared Very Well Prepared Somewhat Prepared: Some GapsNot too Prepared: Large Gaps Not at all Prepared: Struggling
© 2016 THE EDUCATION TRUST
Accountability’s not a cure-all, but done well, it helps create urgency to
improve K-12 preparation for all students.
And while it’s become convenient to characterize accountability as “test and
punish” and the last 15 years as a wasteland for kids, the data suggest
something else…
© 2016 THE EDUCATION TRUST© 2015 THE EDUCATION TRUST
Since we’ve had federal requirements for annual testing, full public reporting,
and serious accountability for the results of every group of children,
achievement among black, Latino, and low-income students has improved.
© 2016 THE EDUCATION TRUST
Source:
© 2015 THE EDUCATION TRUST
Large gains for all groups of students, especially students of color
1971* 1975* 1980* 1984* 1988* 1990* 1992* 1994* 1996* 1999* 2004 2008 2012150
160
170
180
190
200
210
220
230
240
250
9 Year Olds – NAEP Reading
African American Latino White
Aver
age
Scal
e Sc
ore
National Center for Education Statistics, “The Nation's Report Card: Trends in Academic Progress 2012”*Denotes previous assessment format
© 2016 THE EDUCATION TRUST
Source:
© 2015 THE EDUCATION TRUST
Performance for all groups has risen dramatically
1973* 1978* 1982* 1986* 1990* 1992* 1994* 1996* 1999* 2004 2008 2012200
210
220
230
240
250
260
270
280
290
300
13 Year Olds – NAEP Math
African American Latino White
Aver
age
Scal
e Sc
ore
National Center for Education Statistics, “The Nation's Report Card: Trends in Academic Progress 2012”*Denotes previous assessment format
© 2016 THE EDUCATION TRUST© 2015 THE EDUCATION TRUST
Rising scores translate into big improvements in knowledge and
skills for students
© 2016 THE EDUCATION TRUST
Source:
© 2015 THE EDUCATION TRUST
Increased proficiency rates in math
2000 2013 2000 20130%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
57%
27% 23%7%
35%
47%46%
33%
7%26% 32%
60%
Grade 4 Math – by Family Income
Below Basic Basic Proficient/Advanced
Perc
enta
ge o
f Stu
dent
s
Low-Income Higher Income
National Center for Education Statistics, NAEP Data Explorer, http://nces.ed.gov/nationsreportcard/nde/
© 2016 THE EDUCATION TRUST
Source:
© 2015 THE EDUCATION TRUST
For all groups, declines in the percentage of students below the basic level
2000 2013 2000 2013 2000 2013 2000 20130%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
24%9%
65%
34%
59%
27%
61%
30%
46%
37%
31%
48%
34%
47%
32%
46%
30%
54%
4%18%
7%
26%
8%24%
Grade 4 Math – by Race/Ethnicity
Below Basic Basic Proficient/Advanced
Perc
enta
ge o
f Stu
dent
s
WhiteAfrican Amer-
ican Latino American In-dian/Alaska
Native
National Center for Education Statistics, NAEP Data Explorer, http://nces.ed.gov/nationsreportcard/nde/
© 2016 THE EDUCATION TRUST© 2015 THE EDUCATION TRUST
High school graduation rates are up for most groups
© 2016 THE EDUCATION TRUST© 2015 THE EDUCATION TRUST
Rising graduation rates, especially for black and Latino students
2003 2004 2005 2006 2007 2008 2009 2010 2011 201250%
60%
70%
80%
90%
100%
79%
85%
59%
68%66%
76%
91%
93%
68%68%
White African AmericanLatino AsianAmerican Indian/Alaska Native
Aver
aged
Fre
shm
an G
radu
ation
Rat
e
Source: U.S. Department of Education, National Center for Education Statistics, Common Core of Data, http://nces.ed.gov/ccd/data_tables.asp.
Note: Averaged Freshman Graduation Rate is the estimated percentage of entering freshmen who graduate from high school four years later.
© 2016 THE EDUCATION TRUST© 2015 THE EDUCATION TRUST
We can’t take our foot off the accelerator now
© 2016 THE EDUCATION TRUST© 2015 THE EDUCATION TRUST
The business community has been working hard to defend standards and assessment.
Accountability is a continuation of that agenda. If tests show that students aren’t
meeting standards, something needs to be done about that.
© 2016 THE EDUCATION TRUST© 2015 THE EDUCATION TRUST
So what makes for strong accountability?
© 2016 THE EDUCATION TRUST
One way of thinking about accountability…1) What do we value most about school performance?
2) What are our expectations for performance on the things we value?
3) How do we clearly signal to parents and the public whether schools are or are not meeting those expectations?
4) What do we do if they’re not meeting the expectations?
© 2016 THE EDUCATION TRUST
The business, civil rights, and disability community works to answer some of those questions in ESEA reauthorization
© 2016 THE EDUCATION TRUST
Coalition principles for statewide accountability systems that expect and support all
students to graduate from high school ready for college and career.
© 2016 THE EDUCATION TRUST
What do we value most about school performance?
• Indicators: – Measures of student learning outcomes, including
assessments (growth and reading, math, and science proficiency), accurate high school graduation rates, and other academic measures of college/career readiness must be predominant.
– Other indicators (attendance, student surveys, school safety, etc.) may be included, but must play a secondary role.
– Evidence of English proficiency and time in program should be taken into account for English learners.
– Accurate assessment participation rates must be included
© 2016 THE EDUCATION TRUST
What are our expectations for performance on the things we value?
• Goals– States must set public statewide improvement goals
on indicators of student learning outcomes and translate these goals into improvement targets for districts and schools for students overall and for all subgroups, with greater progress expected for groups that have been behind.
© 2016 THE EDUCATION TRUST
How do we clearly signal whether schools are or are not meeting those expectations?
• Ratings– Performance against those targets must be the
predominant factor in statewide ratings systems, with other indicators making up the rest.
© 2016 THE EDUCATION TRUST
What do we do if they’re not meeting the expectations?
• Supports/Consequences– State accountability systems must ensure that steps —
including targeted interventions, supports, and ratings changes — are taken for all schools that consistently miss targets for any group of students. If these systems designate districts as the first responder for interventions and supports, states must assure effective action if those district supports and interventions don’t work.
© 2016 THE EDUCATION TRUST
ESSA: The Final Product• Not the bill we would have written, but it was the
result of compromise• BUT it is NOT a full abdication of the federal role,
as some are saying, and there ARE some key equity levers for states to take advantage of to support all groups of students
© 2016 THE EDUCATION TRUST
So what’s required?
How is that different than what’s going on now?
What should advocates watch out for?
© 2016 THE EDUCATION TRUST
• For those of you who like detail…
© 2016 THE EDUCATION TRUST
Indicators
(i.e. What do we value most about school performance?)
© 2016 THE EDUCATION TRUST
IndicatorsNCLB Waivers ESSA
• For all groups of kids
- Test performance
- Test participation rates
- Grad rates• Another
academic indicator for elementary and middle schools
• Test performance and grad rates required for all groups
• Flexibility on others. The most common ones states added were measures of college and career readiness (i.e. high school course taking, SAT/ACT), and attendance – BUT MOST NOT MEASURED FOR GROUPS
1. Academic Achievementa. Proficiency on annual assessments, as measured against goalsb. May include growth for high schools 2. Other Academic Indicatora. For high schools – graduation rate, as measured against goalsb. For non-high schools – growth or another valid, reliable, statewide academic indicator3. Progress toward English language proficiency for English learners4. Other valid, reliable, comparable and statewide measure of school quality
© 2016 THE EDUCATION TRUST
Beyond tests and grad rates, what indicators should our state use?
What will add to the picture of school performance—not mask underperformance?
© 2016 THE EDUCATION TRUST
Watch out for• Things that can’t be measured for all groups of students (i.e.
low income, black, Latino, etc.)…or for students at all• Things that all schools do well on• Things that aren’t related to the goal of college and career
readiness• Things that sound interesting but we can’t actually measure
well• Things schools don’t have control over• Things that are helpful for some purposes, but not
accountability
© 2016 THE EDUCATION TRUST
A note about opt out• Whether or no 95 percent of students overall and
in each group took the test has to be part of the state accountability system
• Most students who are not tested will automatically count as not proficient.
© 2016 THE EDUCATION TRUST
Why do test participation rates matter?
Because before we paid attention to them, schools would “opt” their lowest performers to stay home, or go on a field trip, on test day.
© 2016 THE EDUCATION TRUST
Goals
(i.e. What are our expectations for performance on the things we value?)
© 2016 THE EDUCATION TRUST
GoalsNCLB Waivers ESSA
100% proficiency for all students and each student group by 2013-2014
3 Choices:1. Cut in half the
difference between current proficiency rates and 100% in six years, overall and for each group
2. 100% proficiency by 2020
3. Another, equally ambitious measure proposed by the state
State-set long-term goals on at least tests and grad rates that expect more progress from groups of students who are farther behind
© 2016 THE EDUCATION TRUST
How do we set aggressive but achievable goals, especially on brand new tests?
© 2016 THE EDUCATION TRUST
Watch out for• Goals that don’t expect much from anyone
• i.e. “any progress” or “0.01%” is enough
• Goals that don’t expect more improvement from groups starting further behind– If all groups make the same progress, those who start
behind, stay behind
© 2016 THE EDUCATION TRUST
Ratings
(i.e. How do we clearly signal whether schools are or are not meeting those
expectations?)
© 2016 THE EDUCATION TRUST
RatingsNCLB Waivers ESSA
Schools either made AYP or didn’t, based on performance against goals for all groups of students on state tests and graduation rates
State-developed rating systems • Typically included
tests and grad rates
• Did not have to be aligned to goals
• Did not have to reflect the performance of individual student groups
State-developed rating systems that • Must be based on how
schools are performing for all groups of students on all the indicators
• If any group is consistently underperforming, that has to be reflected
© 2016 THE EDUCATION TRUST
How do we put everything together in a way that makes sense?
How do we clearly signal when any group is not meeting expectations?
© 2016 THE EDUCATION TRUST
Watch out for• So much data no one can make sense of it…or no
clear signal of what’s good enough• Only identifying the lowest of the low performers• Saying schools are doing well even if some kids
are low performing year after year
© 2016 THE EDUCATION TRUST
Massachusetts: Percent of students in Level 4 or 5 Schools
Total Enrollment White African American Hispanic FRPL ELL SWD0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2% 1% 6% 8% 6% 8% 3%
98% 99% 94% 92% 94% 92% 97%
Percent of Students in Priority Schools Percent of Students in Other Schools
Includes only schools that received accountability ratings
© 2016 THE EDUCATION TRUST
Minnesota: Percent of students attending Priority schools
Total Enrollment White African American Hispanic FRPL LEP SWD0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2% 0% 8% 4% 5% 7% 3%
98% 100% 92% 96% 95% 93% 97%
Percent of Students in Priority Schools Percent of Students in Other Schools
Includes only schools that received accountability ratings
© 2016 THE EDUCATION TRUST
Two “A” Schools: Reading
Reading Proficiency
RatesAll students: 73%White: 89%Hispanic: 48%African American: 19%Low Income: 41%Students w/IEPs: 7%
School 2
Note: This elementary school is approximately 20% African American, 10% Hispanic, and 40% Low-Income
Reading Proficiency RatesAll students: 71%White: 77%Hispanic: 74%African American: 60%Low Income: 65%Students w/IEPs: 40%
School 1
Note: This elementary school is approximately 30% African American, 40%
Hispanic, and over 50% Low-Income
© 2016 THE EDUCATION TRUST
Support/Consequences
(i.e. What do we do if they’re not meeting the expectations?)
© 2016 THE EDUCATION TRUST
Support/ConsequencesNCLB Waivers ESSA
Prescribed set of cascading interventions
Identification of, and action in, three specific types of schools:• Priority – lowest
performing overall• Focus – lowest group
performance/biggest gaps
• Reward – top performing/improving
Identification of, and locally-determined action in: • Schools in the bottom
5% of the state, schools with graduation rates lower than 67%
• Schools where one or more groups of students is consistently underperforming
© 2016 THE EDUCATION TRUST
How do we make sure kids don’t languish year after year?
How do we develop capacity in districts and schools?
© 2016 THE EDUCATION TRUST
Watch out for• No clear timeframes• Action only in the lowest of the low performers• Plan on paper but no real action• An assumption that schools can fix themselves
© 2016 THE EDUCATION TRUST
What’s next?
© 2016 THE EDUCATION TRUST
Implementation Timelines• Regulations: The U.S. Department of Ed will
publish regulations over the coming months • All ESSA requirements, except those pertaining to
accountability, go into effect in the 2016-17 school year.
• New accountability systems must be in place by 2017-18.
© 2016 THE EDUCATION TRUST
Advice for State Advocates
• Wait until regulations are done for legislative/administrative changes in most circumstances
• Begin conversations and building coalitions NOW
© 2016 THE EDUCATION TRUST
Questions?
© 2016 THE EDUCATION TRUST
For more [email protected]
Washington, D.C. Metro Detroit, MI Oakland, CA202/293-1217 734/619-8009 510/465-6444