Which Mich. schools add most value for students?

At first blush, Godwin Heights Public Schools isn’t anything special. Only 6 percent of its juniors were considered “college ready” on their ACTs in 2012, about one-third of the state's median rate. Fourth- and eighth-grade MEAP scores were middling, at best.

On the other hand, Bloomfield Hills has the highest percentage of students deemed “college ready” by the ACT in the state – 58 percent. Its International Academy, which also draws students from surrounding districts, is often ranked as one of the top high schools in the nation.

Nevertheless, Godwin, a high-poverty school system in Grand Rapids, is Michigan’s top traditional district in a new Bride Magazine analysis, beating out Bloomfield Hills and hundreds of other schools where students score higher on standardized tests.

Is it possible that the best teaching in the state is taking place in a high-poverty district where students score at the state average? Can Bloomfield Hills learn from Godwin Heights?

Absolutely, says Amber Arellano, executive director of Education Trust-Midwest, a Royal Oak-based advocacy organization. “This would not surprise us,” Arellano said. “They do extraordinary work.”

Bridge Magazine, in collaboration with the Lansing-based research firm of Public Sector Consultants, created a ranking system measuring a school’s test scores adjusted for student family income, which is often a predictor of academic achievement. In essence, it is a ranking not of achievement, but overachievement.

Complete coverage

Top of the mark  Bridge Magazine set out to find which Michigan public schools do the best job in adding value – helping students succeed beyond expectations. This “Value-Added Matrix” identified some surprising results among both traditional and charter schools in the state.

Academic State Champs 2012  From Godwin Heights to Dearborn, from Baldwin to Troy, from Gaylord to Ann Arbor, see Bridge’s 2012 Academic State Champs.

Where’s your school?   Bridge applied its “Value-Added Matrix” to 560 schools across the state – see where yours ranked in this handy searchable database.

Newcomers, old practices  Charter schools in the tight-knit Arab communities in Dearborn dominated the top of Bridge’s statewide rankings in 2012. Educators credit their success in aiding students, many of them recent immigrants, to their familiarity with the culture and their commitment to making learning a family affair.

The charter conundrum  Charter schools in Michigan posted a remarkable presence at the top – and the bottom – of Bridge’s 2012 Academic State Champ rankings.

Finding what works  Education Trust-Midwest has long studied the central question of education policy: What works? Its executive director, Amber Arellano, tells Bridge that if Michigan helps teachers improve their skills, boosts in student learning will follow.

Read stories by MLive Media Group reporters on how schools in your area fared: http://topics.mlive.com/tag/bridge-champs/

The value-added rankings, released at a time when state leaders are discussing massive education reform, raise questions about how Michigan should distinguish success from failure among schools and teachers.

What do schools add?

To a frustrating level, school test scores are a function of the socioeconomic status of the children who walk through the doors. It’s not a coincidence that some of the school districts with highest raw test scores (Bloomfield Hills, Okemos, Forest Hills) are in wealthy communities, or that struggling school districts (such as Detroit, Flint and Saginaw) are in poor ones.

Bridge and Public Sector Consultants created a Value-Added Matrix (VAM) to determine its 2012 Academic State Champs, celebrating schools that are finding a way to push learning through the socioeconomic ceiling.

School districts traditionally thought of as high-performing do well in the analysis, but the top 10 is dominated by charter schools and high-poverty districts not typically recognized for academic success.

Star International Academy in Dearborn Heights is the top-scoring school, with a VAM score of 120 (a school scoring as expected for the income of its families would score 100). The bottom school out of 561 reviewed is also a charter school, Aisha Shule/WEB Dubois Prepatory Academy, in Detroit, with a score of 76.

The top traditional school district is Godwin Heights with a score of 116; trailing the pack is Ecorse Public Schools, with a VAM score of 82.

Bridge's VAM rankings: No. 1 to No. 560

The scoring system is similar to one introduced by the Mackinac Center for Public Policy in June. While Mackinac’s report ranked high schools, Bridge analyzed test scores from elementary, middle and high school to present a comprehensive view of school districts and K-12 charter schools. The methodology used by Mackinac and Bridge also varies slightly.

See how we did it.

While an analysis that ranks some low-test-score schools above high-performing districts seems counterintuitive, it does offer insights into the complexity of evaluating success and failure among schools with vastly different demographics.

“Unfortunately, poverty and other challenges affect children’s learning,” Arellano said. “Poverty is not destiny, but it does create a steeper hill for teachers to climb.”

Top performers on adding value

Among high school scores alone, Star International topped the VAM ranking, with Caseville Public Schools second. See the top ten.

Alanson Public Schools had the highest eighth-grade score. See the top ten. Godwin Heights topped the fourth-grade ranking. See the top ten

The top-ranked rural school was Mason County Eastern. See the top ten. Okemos was the top affluent school. See the top ten.

'Moneyball' approach to reform

Bridge’s Value-Added Matrix and the work of Education Trust-Midwest suggest a “Moneyball” approach to education reform.

“Think about (Godwin Heights and Bloomfield Hills) and their student populations,” Arellano explained. “In Godwin Heights, 82 percent of students qualify for free or reduced lunch, an indicator of poverty. In Bloomfield Hills, only 12 percent of students qualify for free or reduced lunch. Impoverished students typically start school much further behind than their wealthier counterparts, for lots of reasons. For example, they are exposed to fewer words as children and they typically have fewer books in the home.

“So the educators in Godwin Heights have to work much harder, in many ways, to bring their students’ knowledge up to a career- and college-ready standard than would an affluent school district,” Arellano explained.

Education Trust-Midwest research helped lead to the development of the tenure and teacher evaluation reforms passed in 2011. In 2013, the state will be building a sophisticated K-12 data system that may include value-added growth data on every student and teacher in the state. “This kind of rich data tells us how much students are learning and the impact that teachers are having even as we consider for barriers such as poverty,” Arellano said. “This would be a much more fair and nuanced picture of the quality of schools and teaching and learning.

“For example, we’ll see that some schools and educators are actually producing extraordinary student growth, given their challenges,” Arellano explained. “We should recognize such hard work and teachers’ expertise” even when raw test scores are not high.

Examples of success

That’s the case with many of the schools in the overall top 10 of Bridge’s 2012 Academic State Champs. After Star International Academy and Godwin Heights, the rest of the top 10 were: Riverside Academy in Dearborn; Crawford AuSable Schools in Grayling; Central Academy in Ann Arbor; Macon County Eastern Schools; Kingsley Area Schools; Kingston Community Schools, Okemos Public Schools and Bloomfield Hills School District. The top eight are high-poverty schools.

“I sit at my (office) window and see the empty storefronts around our little school district,” Godwin Heights Superintendent Bill Fetterhoff told Bridge last spring. “Down from us, we had a large GM stamping plant that they just finished leveling. There was a time when you could go in there and make a good living with or without a diploma.”

Fetterhoff knows his school’s raw scores aren’t high (last spring, Bridge listed Godwin Heights as one of the “Faltering 40” districts underachieving in college and career readiness.

“It’s much more than that (raw test scores),” he said. “You have to look at a growth model, to see how much impact you have. You may still have a student who is failing, but how much have they increased their capacity?

“They can perform as well as any kids,” Fetterhoff said, “but foundationally, they come in without skills that other kids have.”

About 44 percent of Godwin Heights students are Hispanic, many of whom are learning English as a second language. Most are poor.

“It boils down to more time and more work,” Fetterhoff said. “And unfortunately, you have a finite amount of time and resources.”

Both Godwin Heights and Crawford AuSable use data extensively to track how students are learning -- and to intervene quickly when a problem arises.

“It’s just like 'Moneyball',” said Crawford AuSable Superintendent Joe Powers. “Metrics are being used everywhere.”

Bridge’s Value-Added Matrix and more complex models being created for the state are “trying to compare apples to apples,” Powers said, “as opposed to looking at ‘here’s a school with great funding and extreme amount of parent participation, and here’s a school with low funding and (high poverty).’”

Senior Writer Ron French joined Bridge in 2011 after having won more than 40 national and state journalism awards since he joined the Detroit News in 1995. French has a long track record of uncovering emerging issues and changing the public policy debate through his work. In 2006, he foretold the coming crisis in the auto industry in a special report detailing how worker health-care costs threatened to bankrupt General Motors.

Facts matter. Trust matters. Journalism matters.

If you learned something from the story you're reading please consider supporting our work. Your donation allows us to keep our Michigan-focused reporting and analysis free and accessible to all. All donations are voluntary, but for as little as $1 you can become a member of Bridge Club and support freedom of the press in Michigan during a crucial election year.

Pay with VISA Pay with MasterCard Pay with American Express Donate now

Comment Form

Add new comment

Dear Reader: We value your thoughts and criticism on the articles, but insist on civility. Criticizing comments or ideas is welcome, but Bridge won’t tolerate comments that are false or defamatory or that demean, personally attack, spread hate or harmful stereotypes. Violating these standards could result in a ban.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.


John Roach
Thu, 01/10/2013 - 8:48am
Quoting the article, "On the other hand, Bloomfield Hills has the highest percentage of students deemed “college ready” by the ACT in the state – 58 percent. Its International Academy is often ranked as one of the top high schools in the nation." The International Academy, though located within the geographical boundaries of the Bloomfield Hills School District, is a consortium school and is not part of the BHSD. Are the results of the IA incorrectly included in the calculation of the scores for BHSD?
Ron French
Thu, 01/10/2013 - 10:37am
Hi John, you raise a very good point. International Academy is coded as part of Bloomfield Hills school district in district-wide data at the Michigan Department of Education, even though the academy draws students from numerous districts. The academy would be separate in building-level analyses. We chose to take a holistic approach, looking at district scores in elementary, middle and high school, rather than a building-level analysis. We realize that creates a problem with Bloomfield Hills. We've now noted in our story that International Academy also draws students from numerous school districts. Thanks for your note.
Ed Haynor
Thu, 01/10/2013 - 9:56am
Certainly, many different scenarios could exist in what researches have chosen to have value. And that in itself, is troubling because regardless of what model you use its validity and reliability is always suspect. In using a “Moneyball” approach to evaluate schools would seem inconsistent in evaluating teachers, students, schools, etc., because researchers appear to be trying to establish a direct relationship between poverty and educational achievement, which would seem to skew their results and show bias by the researchers. There model appears to show that no matter how successful students score on tests, as long as those students come from rich districts, those schools would never consistently rate near the top as long as poor school districts scored fairly well themselves on tests. One might conclude that instead of poverty school districts having more value because their students are overachieving, it might be because there are intelligent parents having intelligent kids, who are living in poverty districts. There is a technical paper available at: http://www.pareonline.net/pdf/v17n17.pdf that concludes that Value-Added-Modeling (VAM) is not what it claims to be. Here are some of their concluding remarks: "We cannot at this time encourage anyone to use VAM in a high stakes endeavor. If one has to use VAM, then we suggest a two-step process to initially use statistical models to identify outliers (e.g., low-performing teachers) and then to verify these results with additional data. Using independent information that can confirm or disconfirm is helpful in many contexts. The value of this use of evaluative change results could be explored in further research efforts…."
Chuck Fellows
Thu, 01/10/2013 - 10:23am
I applaud this effort but must remind all that this study and analysis is dependent upon a data point, gathered at a specific point in time, the score on a standardized test. We have known for a very, very long time that low standardized test scores correlate with poverty. It is refreshing to see the recognition that teachers in high poverty environments have to work much, much harder with fewer resources to climb the ladder of standardized test score success. That learning should give pause to our simplistic and formulaic per pupil approach to school funding and be shared with the Michigan Council on Educator Effectiveness. It is sad to see a ranking and rating of schools and their inhabitants. Learning is not about determining the winners and losers in a contest for higher and higher test scores. It should be about creating an environment that celebrates learning for each individual within their own frame of reference. If our children have the opportunity to taste the joy of learning their education will take care of itself. As a species we did just that for millions of years but have successfully arrested that progress for many with the advent of formal education structured on mass production principles and militaristic goals of uniformity, conformity and blind obedience. I believe it was Mark Twain that said, "I have never let my schooling interfere with my education" . When are we going to learn that ranking and rating, selecting winners and losers, objectifying a subjective art severely diminishes opportunities to love lifelong learning. Our current system is "teaching" many to hate learning and no matter what level of "education" they achieve upon graduation their motivation dies. Relying upon a system that produces a "score" as the measure of success is bankrupt. Using that system of measure would have condemned Einstein to a clerks job, Franklin to a printing press, Faraday to a bookstore, Ramanujan to poverty in India, Henry Ford to watchmaking - I hope you get my point. Learning as a lifelong habit to be enjoyed in its pursuit is the goal.
Thu, 01/10/2013 - 4:07pm
You are right, sort of... ? The purpose of school is to give students the basic tools that allow them to learn (and more often not in school). Whether the the tool is to be able to communicate in a understandable cogent and logical manner, use math and science to understand data and experience, or have basic knowledge of past events to give context to current events. Unfortuately for the anti testing crowd and apparently you, these items can and do boil down to either you understand and can put these tools to use to answer a question on a test or not. Some childeren have advantages and others don't, but none of this negates the meaning or value of test results showing whether a kid has been given the tools or hasn't.
Chuck Fellows
Sun, 01/13/2013 - 6:08pm
Matt, a paper and pencil (or computer aided) test does not and can not demonstrate that a person knows how to use "tools"; only using those tools in a way that is meaningful to the learner can. A true "test" of learning is the ability to demonstrate that learning through an actual performance of using the knowledge to demonstrate a competence, using math to produce an outcome that is meaningful to the person using the math that is recognizable to peers and impartial observers. Cooking require math. Architecture requires math. Physics, chemistry, biology, management, etc requires the use of math as knowledge to produce an outcome of value to the user and the customers of that outcome. Merely reproducing a memorization in a win - lose competition is of no real value at all, unless you can buy goods with a score. An old story: Three groups of science students were given a fine barometer and told to use it to measure the height of a building. The first group took a measurement using the barometer at the base of the building, then at the very top of the building and then used the math and formulas they memorized to calculate the approximate height of the building. The second group took the barometer to the top of the building, threw it over the side and measured the amount of elapsed time it took the barometer to crash into the ground at the base of the building. This group then took their data and knowledge of speed, acceleration and gravity to calculate the approximate height of the building. The third group found the building manager and told him he could have the fine barometer if he told them the exact height of the building, which he did by looking at his records. Which group would you have given the passing grade to?
Scott Baker
Sun, 01/13/2013 - 3:29pm
Any comparison between charter and public schools that does not take into account the percentage of special-needs students serviced and assessed is seriously flawed. Admissions policies/processes vary from school to school, creating differences in student populations that are not accounted for when simply looking at socioeconomic data. While some special-needs students take the MI-Access or MEAP-Access, many special-needs students do take the MEAP, and the greatest percentage of them are enrolled in public schools. And referring to Education Trust - Midwest as "a Royal Oak-based advocacy organization" might be the truth, but not the whole truth. Education Trust is funded by Bill Gates, Eli Broad, the Walton family, etc., the same core of ultra-rich who have used their vast wealth and influence in a concerted effort to destroy public schools and replace them with for-profit, standardized schooling for everyone. Bridge Magazine's relationship with Education Trust is troubling in light of the The Center for Michigan's stated purpose to "serve as an independent, nonpartisan, citizen-led provider of leadership programs and coalition building for improved public policy."
Wed, 01/16/2013 - 3:56pm
One cannot discuss value or value-added in school districts without taking into consideration the disparity in funding. Bloomfield Hiils' per-pupil foundation allowance is $11,000; Okemos' is $8,000. Even Godwin Heights receives $7,700 per pupil. When the rest of us get $6,966 per pupil, this is not an "apples to apples" comparison. The only "insights into the complexity of evaluating success and failure among schools with vastly different demographics" this offers is to be lucky enough to get more money from the state. View all michigan districts' per pupil foundation allowance at:https://docs.google.com/viewer?a=v&q=cache:frHb52xGDoAJ:www.senate.michi...
Tue, 02/19/2013 - 3:11pm
The problem with VAM models is simple. They don't work!!!! Here is an example in the real world why VAM is not used. I am going to compare teachers and students as if they were people working in the private sector. Teachers = Salesman A and B Student Test Scores = Sales Growth = Value Added 10% 2010/2011 Sales Data, Expected Growth for the year 10% Salesman A: Has sales of $10,000 for the year of 2010/2011 Salesman B: Has sales of $100,000 for the year of 2010/2011 2011/2012 Sales Data, Expected Growth for the year 10% Salesman A: Has an increase to $12,000 for the year of 2011/2012 (Increase of 20% over the previous year) Salesman B: Has an increase to $109,000 for the year of 2011/2012.(Increase of 9% over the previous year) Thus, in conclusion if we used VAM to evaluate the two salesman this would be the result. Salesman A, would be rated a highly effective salesman, because they added 20% value to the company and exceeded the projection of 10%. Salesman B, would be rated ineffective, becasue they only added 9% value to the company and did not hit the 10% goal. (I know your saying to youself this doesn't make sense, but that's VAM.) But here comes the kicker, Salesman A would be given a bonus and Salesman B would be put on probation and possibly fired. The same results will happen at schools if you have effective schools and effective teachers compared to ineffective schools and effective teachers. It's very hard to add value to a student that is already successfull but it's not that hard to add value to a student that is not that successful based on past test scores. I am not saying this is right or wrong, all I am saying is the Truth is the Truth!!!!
Ron French
Tue, 02/19/2013 - 3:17pm
Your analysis would be true if our VAM measured growth between years. It does not. Instead, our VAM measures how much better or worse a school's test scores are compared to other schools in the state with students of the same socioeconomic makeup in the same year.