The Case of the Disappearing ELLs: Chapter 3

“Reward results, not compliance.” 1
- Pioneer Institute White Paper on ELLs

“Criterion 3: Compliance: The school compiles a record of compliance with the terms of its charter and applicable state and federal laws and regulations.” 2
- Charter School Performance Criteria, MA DESE

“DESE’s data indicate notable demographic disparities between charter schools and their sending districts, both statewide and for the 21 Commonwealth charter schools in Boston.” 3

- Suzanne Bump, Auditor of the Commonwealth

Welcome to the third and final chapter in the Case of the Disappearing ELLs.  We ended our last post with the following questions:

• Where is the scrutiny towards the Boston charter sector, which continues to dramatically under-enroll English Language Learners despite legislative attempts that require them to recruit and retain a student demographic similar to the sending district?  
• Who is in charge of enforcing the accountability measures of state laws related to charters and ELLs? 
    

In this post, we’ll take a look at how MA DESE and State Education leaders have addressed these issues.

Lowering ELL Enrollment Standards for Charter Schools
DESE’s Feb. 2016 Charter School Enrollment Data Annual Report (which, by state law, the Commissioner must present to the Legislature every year) goes to great lengths to highlight the Board’s oversight of enrollment at MA Charter schools: “One of the Department’s key strategic priorities with respect to charter schools is to utilize enhanced tools and oversight processes to support and oversee compliance with these regulations (emphasis added).”  In 2013-14 DESE showed its "resolve"on the issue of compliance with statutory enrollment requirements by adding a new set of expectations related to student access and equity, and developing CHART, the Charter Analysis and Review Tool to track MA Charter Schools’ progress related to these expectations.   
 
All of this seems well and good until one actually looks closely at whose advice helped shape the “expectations” included in CHART. The advisees included “an Access and Equity Working Group of charter school leaders convened during the 2013 – 2014 school year, in conjunction with the Massachusetts Charter Public School Association (MCPSA).” 4 So, the State agency responsible for ensuring Charter School compliance with state law wrote their new regulations with Charter School leaders and the state's main Charter School advocacy organization. Let's take a look at how that worked out.  

The first accountability measure included in CHART is called the Comparison Index.  Here’s how it is described:

“The comparison index is a statistically calculated value designed to produce a fairer and more realistic comparison measure that takes into account the charter school’s size and the actual prevalence of student subgroups within only those grade levels in common with the charter school.” (emphasis added)5

 
In case you missed it, what DESE is saying is that in order to be “fair” and “realistic” in its oversight of Charters, it needs to lower the enrollment standards for ELLs established by the wording of the State statute.6 What does this lower standard look like in Boston Commonwealth Charters? According to DESE, it means that, Charter Schools in the community with the third largest percentage of ELLs in the state (30%) should receive special accommodations that only require them to enroll, on average, 19% ELLs in their schools. But the fairness doesn’t stop there!  

As all good teachers know, when you make accommodations for a student in the classroom, you also need to adjust your assessments to meet those accommodations.  So, CHART also includes a second accommodation called the Gap Narrowing Target (GNT). The GNT is DESE’s assessment tool for tracking Charter Schools’ progress toward the new reduced ELL enrollment standards. Here’s how the GNT is described:

The Gap Narrowing Target (GNT) refers to the halfway point between the school’s baseline rate (which is the rate in the 2010-11 school year, or the first year enrollment data is collected if after 2010-11), and the current Comparison Index (the “target”). The object is to meet this halfway point by the 2016-17 school year (or in a later year if baseline is after 2010-11), giving the school six years to do so. For a school to be on schedule to meet its GNT, an incremental increase must be met annually.7

DESE’s use of the the Gap Narrowing Target to assess Charter schools effectively reduces the already adjusted Charter School ELL enrollment requirements by 50%. What does that mean for Boston’s Commonwealth Charter Schools?  According to DESE, it means that Boston’s Charter innovators have met or exceeded the standard if their schools have enrolled, on average, 8% ELL students. Thanks to this second accommodation, DESE’s evaluations of Charter School ELL enrollments are also no longer directly correlated to the student populations in the communities where they are located, as State statute dictates.

So how did Boston’s Commonwealth Charter Schools perform relative to these new lower standards? In the five-year period from 2012 - 2016, nine Boston Commonwealth Charter Schools failed to meet their Gap Narrowing Targets for ELLs at least three of the five years. Six of those schools failed to meet their GNTs for ALL FIVE YEARS.8 Only three schools met or exceeded their Comparison Index at least 3 of the five years.

Does that mean that even these new dramatically reduced enrollment standards for ELLs in Charter schools are “unrealistic?”  Not if you ask KIPP Boston, MATCH, and Bridge Boston Charter Schools.  As of 2016, all three have used their resource privilege as Charter schools to attain numbers of ELLs equal or exceeding BPS’ average of 30%. Perhaps it’s time for these three schools to share their “innovative approaches” to ELLs with their Boston Charter colleagues?

Using Demographics as an Excuse for Low Enrollment

As we noted in our first post on ELLs, Massachusetts Secretary of Education Jim Peyser is well aware that Charters under-enroll ELLs: “The biggest demographic difference between BPS and charter schools involves students whose first language is not English. About 45 percent of BPS students come from homes where a language other than English is spoken, compared to 21 percent in Boston charter schools.” 9

But instead of encouraging Charter Schools to rise up and meet the challenge of implementing new strategies for recruiting and retaining more ELL students, the Secretary has taken to offering demographic excuses meant to justify lower standards for the state’s "education innovators": “Part of this difference reflects the areas of the city in which charter schools are located and the racial and ethnic makeup of the surrounding neighborhoods, which supply most of the students.” 10

Is this statement true? Let’s take a look. We plotted the address of each Commonwealth Charter School in the DiscoverBPS tool to identify the closest BPS schools enrolling the same grade levels. We found that Peyser’s statement was largely not true (all of our data can be viewed here). For example, in 2016 Brooke East Boston has 12% ELLs, while the BPS schools Guild Elementary and Patrick Kennedy enroll 66% and 74.5% ELLs respectively. Brooke Mattapan has 3.7% ELLs, while BPS schools Sarah Greenwood has 46.4% and Lee K-8 has 16.2%. Academy of the Pacific Rim has 6.7% ELLs, while the Roosevelt K8 has 18.6% and the Channing has 19.5%.

Based on 2016 data from DESE. Each charter school paired with closest BPS schools using DiscoverBPS.

Why would Jim Peyser, who obviously understands Charter Schools very well, make a statement like this? DESE’s adoption of the Comparison Index gives us a big clue, it’s called Unified Enrollment. BPS currently uses a home based lottery that gives preference to neighborhood families, while Boston Charters have a citywide lottery. However, the Mayor’s office and the Boston Compact (backed by The Boston Foundation) are hoping to unify the enrollment process in Boston and to change charter schools to a similar neighborhood lottery. Not surprisingly, unified enrollment systems like this one were championed by current Secretary Peyser when he was the Managing Director of the New Schools Venture Fund, an education venture fund for the Charter sector.    

But are Boston’s charter schools prepared to absorb a dramatic increase in ELL students that certainly would come with Unified Enrollment? For an answer let’s take a closer look at Brooke Roslindale. For those of you not familiar with the Brooke system, it is one of the wealthiest (“financially healthy”) and fastest growing Charter school operators in Boston11, with schools in Roslindale, Mattapan and East Boston. In May 2015, Brooke Charter Schools were award the Pozen Prize for Charter Schools, "an $80,000 prize to recognize academic excellence and quality learning environments in Greater Boston charter schools." This Spring, DESE approved a Brooke proposal to build a new high school in Boston. Roslindale is Brooke’s flagship school, and is often held up by Charter School advocates and State Education Officials alike as a model for Charter school success.   

As we mentioned earlier, the Department of Elementary and Secondary Education (DESE) has a tool called CHART available on their website to evaluate the progress of Charter School Recruitment and Retention plans.

 

In 2016, Brooke Roslindale has 2.5% ELLs, while the statewide average is 9% and BPS average is 30.3%. In our research, we found the closest neighboring BPS schools enrolling the similar age groups were the Charles Sumner at 37.8% ELLs and the Philbrick with 17% ELLs. Clearly, DESE’s 2016 Gap Narrowing Target of 9.7% for Brooke Roslindale is a number that is only relative to Brooke Roslindale itself and not neighboring schools or the district. As you can see from the chart below, Brooke Roslindale has never met its Gap Narrowing Target or its Comparison Index in the past 5 years. Also, it has not made yearly incremental increases in its ELL populations as the GNT accountability measures dictate.  For more info, click here for a simple comparison of all three Brooke schools using the CHART ELL enrollment performance criteria.  

Screenshot from DESE CHART tool

Under a unified lottery, would Brooke Roslindale and all of the other Boston charter schools with a dramatically lower rate of ELLs suddenly be able to accommodate a more complex student demographic? Given a five-year track record in which 45% of Boston’s Commonwealth Charters were either unable or unwilling to consistently meet the dramatically lower enrollment criteria set out by the GNT, and only 15% of Boston’s Commonwealth Charters we able to meet or exceed their Comparison Index,  you don’t need to be a Broad Foundation Fellow to answer that question. 

We’ll close this third and final post with something for you to ponder as you consider the evidence in the Case of the Missing ELLs.

This Spring two Boston Charter school operators, the Brooke Network and Neighborhood House, received approval from the State Board of Elementary and Secondary Education to expand their operations in Boston next year. All three Brooke Schools and Neighborhood House failed to meet their Gap Narrowing Targets for ELL enrollment for the last 5 years. 

Now that Mitchell Chester has opened up another 225 Charter seats in Boston, perhaps it’s about time that No Excuses apply to DESE and to Charter School operators, not just to their students.   

1 Charter Analysis and Review Tool - Demographics,  http://www.doe.mass.edu/charter/finance/chart/
2 MGL c. 71, § 89; 603 CMR 1.05. State law doesn’t identify a  specific number, rather it states: “a student population that is demographically comparable to similar grades in schools from which the charter school enrolls students....” 
3 Charter Analysis and Review Tool - Demographics,  http://www.doe.mass.edu/charter/finance/chart/ 
4 Charter School Enrollment Data Annual Report,  MA DESE. Feb. 2016, p. 5.
5 Charter Analysis and Review Tool - Demographics,  http://www.doe.mass.edu/charter/finance/chart/
6 MGL c. 71, § 89; 603 CMR 1.05. State law doesn’t identify a specific number, rather it states: “a student population that is demographically comparable to similar grades in schools from which the charter school enrolls students....”
7 Charter Analysis and Review Tool - Demographics,  http://www.doe.mass.edu/charter/finance/chart/
8 The six schools are: Academy of the Pacific Rim, Boston Renaissance, Brooke East Boston, Brooke Mattapan, Brooke Roslindale, Codman Academy and Neighborhood House.  All data is directly from CHART http://www.doe.mass.edu/charter/finance/chart/  

9 “Boston and the Charter School Cap” EducationNext, Winter 2014, Vol. 14, No. 1

http://educationnext.org/boston-and-the-charter-school-cap/.
10 Ibid
11 For more information on Charter school finances, see our post on Charter School “Operational Gains".

The Case of the Disappearing ELLs: Chapter 2

“...the Department urges caution in drawing conclusions or inferences regarding the comparability of subgroup populations among public schools—charter schools, traditional public schools, or other public school options.” 

Charter School Enrollment Data Annual Report,  MA DESE. Feb. 2016

“A charter school must develop a plan that includes deliberate, specific strategies the school will use to attract, to enroll, and to retain a student population that is demographically comparable to similar grades in schools from which the charter school enrolls students...MGL c. 71, § 89; 603 CMR 1.05.”

Annual Report Guidelines for Charter Schools 2015-16, MA DESE. Revised April 2016

In Chapter 1 of the Case of the Disappearing ELLs, we began with the basic facts related to English Language Learners (ELLs) in Boston Public Schools (BPS) vs. Boston Commonwealth Charter Schools. To recap, on average, 30% of BPS students are ELLs students compared to only 13% in Boston Charters, despite legislative efforts in 2010 to align these demographics. In this post, we’ll tackle the differences in the classification of ELLs and what testing reveals about the differences in ELL complexity.

Before we dig in, it’s worth a quick review of how students are identified as ELLs in the first place. Under Federal and State law, Massachusetts school districts must take necessary and appropriate steps to identify ELL students and determine appropriate educational services for their learning needs. Districts are required to administer a Home Language Survey provided by DESE to all new students Pre-K through 12th grade. Students whose survey shows that a language other than English is spoken in the home are then given a language proficiency test. A licensed ESL teacher or administrator then makes a decision about whether a student is an ELL.1 

The issue of under-representation of ELLs in BPS and Boston Charter Schools is not new. As we mentioned in our last post, in 2010 the Department of Justice (DOJ) and Office of Civil Rights (OCR) investigated BPS for ELL violations, including under-classification of ELLs. In the same year, the state Charter School law was amended requiring schools to develop and implement a Recruitment and Retention plan. In 2013-14, DESE added new expectations related to access and equity, and developed CHART, the Charter Analysis and Review Tool “which provides multi-school, multi-year demographic comparison data for each charter school and comparison schools.” 2

BPS enrollment data from DESE. View high resolution graph HERE.

While the BPS settlement with the DOJ in September 2010 resulted in a 39% single year jump in the number of students classified as ELLs attending BPS, the new DESE expectations have had a minimal effect upon what were already comparatively small percentages of ELLs in Boston Charters. 

Not that you would know this if you read DESE’s 2016 Charter School Enrollment Data Annual Report: “The percent of English language Learners (ELLs) enrolled at Boston charter schools and charter schools located in Gateway Cities has steadily increased and is approaching the enrollment found at Boston district schools and district schools." 3  

With all due respect to the authors of the report, DESE’s own data just doesn’t support that conclusion when it comes to Boston. As we noted in our previous post, just three Boston Commonwealth Charter Schools account for almost all of the growth in ELLs in Boston Charters in recent years. Their efforts and progress in this area have been real and significant. Unfortunately, their statistics are often used, as in the DESE report statement above, to mask the reasons for the equally significant lack of progress vis-a-vis ELLs at other Boston Charters. Let’s dig in and take a look. 
 
Lower Classification Rates

We’ll begin with a comparison of the percent of First Language Not English (FLNE) students who went on to be classified as ELLs in BPS vs. Boston Commonwealth Charters. 68% of FLNE students in BPS went on to be classified as ELL, while only 44% of Boston Commonwealth Charter students were classified as ELL.

Based on 2016 data from DESE. View collected data HERE. View high resolution figure HERE.

Our findings are similar to those reached by MIT researcher Elizabeth Setren in her 2015 discussion paper Special Education and English Language Learner Students in Boston Charter Schools: Impact and Classification: “In all school levels, charters remove ELL status at the time of enrollment at a substantially higher rate than traditional public schools. Ninety percent of elementary ELL applicants who enroll in traditional public schools remain ELL by the following fall, but as shown in Table 8, 20.5 percentage points fewer elementary school ELL applicants maintain their ELL classification in charters." 5

Ms. Setren's conclusions, particularly those concerning Charters' high rates of removing ELL status at the time of enrollment, bring up some important questions to which we don't have good answers at this time. Let's consider the following example. A student, identified by BPS as an ELL student enrolls in a Charter school. When that student arrives at the Charter, for whatever reason, the school removes that students' ELL status. What then happens to the additional tuition dollars the Charter school received for the purpose of providing ELL services to that student? 

Enrolling students with higher Levels of English Language Proficiency

Students who have been classified as ELLs, or even former ELLs, are administered the ACCESS exam in February of each year to rate their English language proficiency. Level 1 is the least proficient in English and Level 6 is the most proficient. 53% of the ELL students in BPS are in Levels 1 - 3, compared to 23.3% of ELLs students in charters. 47% of ELLs in BPS are in Levels 4 - 6, compared to 76.3% in Boston Commonwealth Charters. DESE’s data clearly shows that Charters are enrolling the less complex ELL student population.

Download raw data from DESE HERE. View our collected data HERE. View high resolution figure HERE.

Enrolling fewer ELLs who are in MA Schools for the 1st or 2nd year

A related and equally telling statistic is the number of ELL students in their first or second year in Massachusetts schools. In FY15, 12% of BPS students were ELLs in their first or second year in MA schools. Only three Boston Commonwealth Charters have higher percentages of 1st/2nd year ELLs than BPS in that year. The rest had significantly smaller numbers of these students. 

Raw data from DESE CHART tool. Our collected data HERE. High resolution figure HERE.

Once again, DESE’s data demonstrates that the majority of Boston Commonwealth Charters are not only enrolling fewer ELL students, but the students they are enrolling have higher levels of language proficiency and are far less likely to be a recent arrival to MA and/or to the U.S.

So, we ask these questions again: Where is the scrutiny towards the Boston charter sector, which continues to dramatically under-enroll English Language Learners despite legislative attempts that require them to recruit and retain a student demographic similar to the sending district? Who is in charge of enforcing the accountability measures of state laws related to charters and ELLs?   

In the third and final chapter coming soon, we’ll take a look at how MA DESE and State Education leaders have addressed these issues and the prospect of Unified Enrollment in the context of significant ELL enrollment disparities.

1All of this information and more can be found at DESE’s Guidance from the Massachusetts Dept. of Education and Laws http://www.doe.mass.edu/ell/guidance_laws.html.

2 Charter School Enrollment Data Annual Report,  MA DESE. Feb. 2016, p. 5.

3 Charter School Enrollment Data Annual Report,  MA DESE. Feb. 2016, p. 9 - 10.

4 For complete data sets related to ELLs that we used for this post, download CHART or go to the Selected Populations Reports on the DESE website.  

5 Setren, Elizabeth. Special Education and English Language Learner Students in Boston Charter Schools: Impact and Classification.  School Effectiveness and Inequality Initiative Discussion Paper #2015.03. p. 12.