Common Core of Data I

[Chris_Wendt]

(First in a series of editions...)

I. A Study of Public School Graduation Rates

In the course of developing a blog about the crisis in Philadelphia Public Schools, it was immediately obvious to me that a study of high school graduation rates in the U.S. was in order, especially among large school systems or districts. It was also immediately apparent that no reference material existed which provided easily obtainable, reliable and complete data on graduation rates.

The ensuing process was well worth the time and effort I invested researching and analyzing the data, however, yielding many interesting discoveries.

My first discovery was the existence of the Institute of Education Sciences, the National Center for Education Statistics, and the Elementary & Secondary Information System which allowed me to generate a master table of all school districts and schools, and enrollments by grade from the U.S. Department of Education Database. Unfortunately, that system does not provide performance data, i.e. graduation rates, for school districts. But at least I knew what and how many school districts existed, how many schools they operated, how many students they served, and where they were across the country.

Thus, at this initial step, I learned the following:

• There are 95,704 public schools…

• Operated by 14,826 school districts…

• Serving 49.3 million students (2010-11) in the 50 states + DC

• One state, Hawaii, has but a single school district for all of its schools, and that is the 9th largest district in the country

• The largest school district is New York City with about 1 million students.

• NYC has 50% greater enrollment than the #2 largest district, Los Angeles Unified

• Los Angeles has 50% greater enrollment than the #3 largest district, City of Chicago SD 299

• The 10 largest districts serve nearly 8% of America’s public school students, 3.7 million.

• The Mean Average enrollment of the 100 largest districts is over 100,000 students with a median point of 67,000 (middle of scale from 47,000 to 995,000 from San Juan County, CA to NYC); these 100 districts serve about 20% (one in five) of all public school students in the U.S.

• Four (4) of NYC’s Geographic Districts are themselves among the 100 largest districts, nationally, presenting a statistical challenge to keeping my research clean.

I had yet to obtain any reliable graduation rate data.
This will be developed over coming editions of this blog.

Stay tuned....

Chris Wendt

[rr]

Hi Chris,

I'm just curious, what's the point to be illustrated by reviewing enrollment numbers and graduation rates for all the public schools in the US? I think it might be interesting to analyze but I'm not sure I understand the purpose, especially considering that there could be areas that skew the data up or down like you've pointed out in NYC.

Not to sound rude but, is there a point to these bullets? Providing data points is fine but if there's no opinion or thesis statement to be supported by the numbers it's really just a bunch of numbers. I don't doubt the potential need for this type of data but you usually have an opinion in your posts which is often interesting to discuss and/or debate.

[Chris_Wendt]

Please consider the title of this thread ("Common Core of Data") and its apparent (non-accidental) lack of congruence with the bullets I posted. Note, too, that this edition starts with the deliberately formatted caption: I. A Study of Public School Graduation Rates...hopefully creating at least the expectation of there being II. and III. or more chapters to follow.

Finally, I did more than hint about this arising out of my developing a blog about a serious crisis in Philadelphia, which led me to this wide expansion of my perspective.

I cannot connect the dots without first laying-out the dots. Today I gave you ten dots. Far short of a reasonable amount of dots (data points) from which to draw--and defend--any conclusions.

Tomorrow I am going to relate some of my experiences garnering graduation rates. This will tie-back to some degree to the main title, "Common Core of Data", which is also a precise and official title, not something I made up.

The end game is to make this all relevant to Wantagh, to Long Island, and New York State; to you and the readers here. And yes, certainly to encourage discussion or even argument as appropriate.

Thanks for following along with me.

Deep into this quest,

Chris Wendt

[Chris_Wendt]

The Common Core concept has engendered a lot of discussion, support, opposition, and therefore, controversy for some time. Most of the conversation swirls around one aspect or feature of the Common Core, the Common Core Curriculum. Today’s installment will NOT deal with the curriculum aspect of Common Core, nor with the assessment component of that curriculum. Their days lay ahead, however. I am staying with H.S. graduation rates in the context of the Common Core of Data for now.
In the course of my learning about public high school graduation rates, I encountered an elaborate system of organized obfuscation, cheating, and outright resistance, primarily from three quarters:

• the political quarter (politicians and the legislative bodies or executive offices they inhabit)
  
  
• the educational establishment including state education bureaucrats, school district boards and administrators, and educational unions, not necessarily on the same page, but all singing out of the same hymnal
  
  
• the industrial quarter, but especially the real estate industry, the home building industry, and Industrial Development Authorities whose mission is to attract and recruit corporations to move into their state or county, or, to retain existing corporations within their boundaries.

The fourth estate, the media, have been actively complicit in their support of obfuscating graduation rates in many places.

I am happy to report at the outset that New York is probably the best among all 50 states at following federal guidelines, adopting federal standards, including nomenclature and definitions, but especially for promulgating ease of finding educational data that is consistently presented, reasonably up-to-date, and on its face, accurate and transparent. This is certainly not an across-the-board or blanket situation in NY, notable exceptions being Roosevelt (cheating on graduation rates), and NY City, where the political quarter and the educational establishment have teamed-up to embellish graduation rates to the good side, and to obfuscate pinpointing them, on the bad side. But there is just so much that lipstick can do for…an unattractive specimen.

I was able to work-through the data and resolve issues among the Geographic, Special, and central NYC districts (system). I was not able to dis-aggregate charter school graduation data from the regular public high school data, which NYC has, with permission, commingled. I did try, but there are too many charter schools, and, by definition, children can attend charter schools outside of their Geographic districts. Same goes for Special public schools, and other PS programs. NYC has tons of kids riding subways east, west, north and south every day, attending a diverse assortment of schools all over the city. So, for NYC, I developed and used what I call the NYC roll-up data to determine a realistic graduation rate for the central system, and suppressed individual Geographic and Special District graduation data from any downstream calculations I made.

Preview: NYC had a 60.9% graduation rate in 2011. More accurately, NYC’s 4-year cohort graduation rate in 2011 was 60.9%.

The 4-year cohort graduation rate is very significant terminology, and lies at the very center of resistance to transparent graduation results as they relate to the Common Core concept, but especially to the Common Core of Data (concept). As you may imagine, there are dozens of ways of manipulating the definition of “graduation rate”. The currently specified and most restrictive method is the 4-year cohort rate, which essentially compared the class which entered high school as freshmen in 2006-07 with their graduation rate as the Class of 2010-11, four years later. How many entered? How many graduated 4 years later? Bingo: 4-year cohort Graduation Rate. There are allowances for in/out transferees who can be accurately tracked in/out of other school districts or private schools, and for mortality, etc. This sounds (and is) very simple and easy. But the three quarters mentioned above have made this very, very hard in many places.

Many states have applied for waivers, some of which have been granted. Some states use two or even three different graduation measures, which they post together on the same confusing websites: 4-year cohort, 5-year cohort,”4-year graduated or retained”, and one called the Texas method, used by Texas alongside the NCLB cohort method.

Many places with large (county or big city) school districts do not publish district-level graduation rates, only individual high school rates. That keeps the good looking high schools looking attractive, and keeps the poor performing high schools from “ruining” the reputation of the good schools. A common tactic is to use the 5-year cohort rate, or the 4-year-or-retained rate, but NOT DISCLOSE WHICH ONE is being published. (I was able to determine when this was happening by noting where the number of graduates exceeded the base Grade 12 enrollment numbers in the federal database, indicating either the 5-year cohort or retained students were being counted as having graduated. Another favorite trick is publishing older rates (when better than the current rate) and simply not disclosing the cohort used. Again, careful comparison with federal enrollment data helped uncover this. If the recent past data was also bad, then a state could just drop back to a better year and publish that data as though it was the most current; for one state, that year was 2009!

To get at all this data in a reliable way, I had to comb through federal, state, local sources, and sometimes read numerous news articles to uncover the truth. In other states, notably New York and California, the data was easy to find and clear once I found it.

One of the more interesting observations from the Graduation Rate data is depicted below. Chunking enrollment into 1-million student groups from the largest district down to the 106th largest district in declining enrollment (size) order (11 million cumulative students), the graduation rate improved in nearly straight-line relationship to the number of students encompassed. The graph depicts the percentage of student who DID NOT GRADUATE ON-TIME in relationship to the number of students as the enrollment of the districts declined, in clumps of 1 million.


Obviously, size does matter, and here is proof in hard data. The number of students who did not graduate on time improved as progressively smaller districts were included in the base.

I am beginning to coalesce some opinions about this, but first I would like to publish more dots for connecting. Tomorrow, some additional observations about the actual graduation rates.

Chris Wendt
[e-mail]

[Chris_Wendt]

Although I am going to lay out additional dots, below, there is a lead-off question that we need to consider:

Is it a reasonable expectation that 100% of high school students should graduate within four years,
with academic degrees that meet rigorous federal standards?

Points to ponder, before[/i][/u] answering:

• How many (what number of) students do not presently graduate from high school in four years, regardless of degree or type of diploma?
  
  
• Will increasing the academic rigor of graduation requirements for all students improve the graduation rate for those “X” number of students who do not presently graduate from high school in four years, or, will increasing academic rigor cause the number, “X” to increase as additional students, presently “at risk” of failing or of dropping-out, fail to meet the more rigorous requirements?
  
  
• Considering the dilemma of pervasive student loan debt, the paucity of college degree-related job opportunities, and what may become a shortage of eligible American workers to work in blue-collar and trade occupations, is college readiness a desirable outcome for every high school student…even those who have little or no interest in going to college?
  
  
  There is an additional point to ponder, which I am separating out and posing rhetorically:
  
  
• What is the cost-benefit of attaining a 100%, four-year cohort high school graduation rate?

This is asked rhetorically because sufficient data does not exist to answer it other than theoretically; specifically, we do not know the cost of getting from our present graduation rate situation, and we do not know the benefits, in terms of actual payback (return on investment - ROI), especially in terms of defining what is a benefit, to whom do these benefits accrue? Rhetorical questions are sometimes the most fun to answer, though, so feel free to respond.
------------------------------------------------------------------------------------------------------------------------------------------------------------
Dots for today:

• Wantagh & Seaford ranked 2,091st and 3,851st in enrollment among U.S. school districts, with 3,507 and 2,582 students, respectively in 2010-11

• Wantagh & Seaford reported 4-Year Cohort Graduation Rates for 2010-11 of 98% and 93%, respectively, meaning that 5 students and 15 students, respectively, did not graduate on-time

• Wantagh had 3 drop-outs and Seaford had 4 drop-outs from the 4-year cohort that graduated in 2011.

• The 100 largest school districts comprise a total enrollment of 10.7 million students, with an amalgamated 4-year cohort graduation rate of 73.3% for the Class of 2011, with the result that 184,350 students failed to graduate from those 100 districts with their class in 2011.

• The 100 largest school districts are located across 28 states, with Florida, California, and Texas having the greatest concentration, at 14, 12, and 19 of the largest 100 districts, respectively.

• Of the 100 largest school districts, Florida’s 14 had 2010-11 senior classes comprising 127,250 students, with a graduation rate of 70.8%, meaning 37,183 did not graduate on-time.

• Of the 100 largest school districts, California’s 12 had 2010-11 senior classes comprising 94,429 students, with a graduation rate of 74.6%, meaning 23,999 did not graduate on-time.

• Of the 100 largest school districts, Texas’ 12 had 2010-11 senior classes comprising 87,428 students, with a graduation rate of 84.7%, meaning 13,337 did not graduate on-time.

• These three big-school-district states provide further data-based evidence of the apparent inverse relationship between district size and graduation rates within jumbo school districts. Or do they?

Chris Wendt

[Chris_Wendt]

Looking at the educational landscape and the high school graduation rate performance of public school districts in America leads to some preliminary conclusions:

• The organization of most states’ schools into districts is widely inconsistent

• The graduation performance of school districts within most states, and across all states is also widely inconsistent, with the major exception that data indicate the 4-year cohort graduation rate in jumbo-sized school districts (the 106 largest US districts which collectively serve 11 million students) is inversely proportional to district enrollment among groupings of size-ordered districts comprising 1 million students, regardless of state(s) (see graph, above, this thread)

• To some degree, in addition to school district enrollment (size), states' school district organization may either have a causal relationship with low graduation rates, or, may serve as an impediment to improving low graduation rates, or both.

Considering that the 106 largest school districts serve more than one out of every five students in America, there is certainly persuasive logic to addressing graduation rates in those jumbo-sized districts as a potentially effective way of improving the national 4-year cohort graduation rate.

On the opposite end of the continuum of school district size, the nearly 13,000 small districts with enrollments between 100 and 4,999 students which collectively serve one out of every three U.S. students, efficiency-based arguments could be made for reducing the number of small districts by various means of combination, from outright merger to forming administrative structures to share costs and resources, while improving opportunities & broaden course offerings among pairings or larger groupings of them.

From a high-level perspective then, it could be considered that efficiencies (“synergies” as they are called in corporate M&A parlance) gained by combining small districts by whatever means may work, would generate free cash flows and provide available financial and human resources which could then be applied to improving flagging graduation rates in jumbo districts, while reducing local tax burdens and broadening course offerings among those smaller districts.

Thoughts?

Chris Wendt