This study is an examination of Florida and Michigan's performance on the National Assessment of Educational Progress's (NAEP) standardized test, often referred to as "the nation's report card." Immediately prior to and during Florida's immense improvement on these scores from the past 15 years, the state made substantial changes to its public education system. Some of these policies have been rigorously studied and have shown a positive impact on Florida students that Michigan should emulate to improve its static performance.

Executive Summary*

Over the last 15 years, students in Florida have demonstrated remarkable improvements in average fourth- and eighth-grade reading and math scores on the National Assessment of Educational Progress, a standardized exam often referred to as “the nation’s report card.” Meanwhile, during the same period, the same NAEP test scores in Michigan have improved only slightly or not at all. This study examines both states’ results, describes the education policies that likely contributed to Florida’s success and suggests how Michigan could improve student achievement based on the Florida model.

Florida’s NAEP gains from 1992 to 2011 were the second-highest in the nation, yet they were achieved with the country’s lowest per-pupil spending increases. The contrast with Michigan is particularly stark. For example, from 1998 to 2011, Florida students’ average test scores increased by 9.1 percent in fourth-grade reading. In Michigan, these same scores increased by just 1.3 percent, and the national average increased by just 3.4 percent. In fourth-grade math, Florida students improved their scores by 11.2 percent from 1996 to 2011, while Michigan students improved by only 4.5 percent and the nation by 8.1 percent. 

Yet many people would have expected Michigan to post higher average test scores than Florida. Every year from 1990 to 2009, Michigan spent more per pupil than Florida and spent more compensating teachers. From 2000 to 2011, Michigan also had a smaller share of low-income students — that is, students whose family incomes were low enough to qualify for a federally subsidized free or reduced-price lunch.

Florida outgained Michigan in eighth-grade NAEP scores, as well. Florida’s results in eighth-grade reading improved by only 3 percent from 1998 to 2011, but Michigan’s did not improve at all, and the national average increased by less than 1 percent. In eighth-grade math, Florida boosted average scores by 5.4 percent from 1996 to 2011, while Michigan inched up by just 1.2 percent, and the national average improved by 5.2 percent. By 2009, Florida eighth-graders had passed Michigan in both subjects, though Florida fell slightly behind Michigan again in 2011.

Similar trends hold when comparing the test scores of low-income students in both states. Altogether, then, Florida’s initial test scores were lower than Michigan’s in a total of eight different grade, subject and student family income categories, but rose quickly enough to pass Michigan’s by 2009.

Immediately prior to and during Florida’s remarkable improvement, the state made substantial changes to the policies that govern its public education system. These included a new school accountability system, clear limits on social promotion, considerable expansions in the schools that parents could choose from, resources focused on literacy, and alternative routes to certification for aspiring teachers. Some of these policies have been rigorously studied and shown to have a positive impact on student achievement in Florida.

In light of Florida’s success, Michigan policymakers should consider the following reforms in this order:

1. Present voters with a proposal to remove Michigan’s constitutional prohibition on using tax credits to support the enrollment of students in private schools


2. Eliminate geographical boundaries to parents’ ability to choose from a variety of public school options, including online courses and online schools


3. Implement an easy-to-understand, A-through-F school accountability system that creates genuine rewards and consequences for schools — not districts — based on their performance


4. Expand the pool of capable teachers by increasing the ways in which aspiring teachers may become certified


5. Limit the ability of schools to socially promote third-graders who are not proficient in reading


6. Focus resources on teaching literacy.

Following the Florida model might not produce results quite as dramatic in Michigan, but even a modest portion of Florida’s gains would be significant. Many of the Sunshine State’s policies have a track record of success and provide a promising path for Michigan policymakers to follow.

Introduction

Beginning about 15 years ago, Florida enacted several new policies aimed at improving its public school system. Among these were the following: Schools were held accountable to higher standards; reading skills were emphasized, especially in the early grades; and public funding enabled more parents to choose such options as online courses, private schools, charter schools and neighboring public schools.[1]

Over the course of the following decade, Florida’s average student performance began to improve on the National Assessment of Educational Progress (generally considered the “nation’s report card”).[2] In a 2012 study from Harvard University, Florida outperformed all other states but one, Maryland, on gains in how many students tested proficient on NAEP tests in reading, mathematics and science from 1992 to 2011.[3] Florida achieved this unusual record while registering, by far, the nation’s lowest increase in per-pupil operating expenditures.[4]

The following report will look closely at the gains made by Florida students over the last decade and compare them to the performance of students in Michigan. The report will also outline the Florida reforms that coincided with these achievement gains and use this analysis to weigh recommendations for Michigan’s public education system.

Conditions of Public Education in Michigan and Florida

This section provides a comparison of the two states’ K-12 public school systems using three common measures: total per-pupil public school expenditures, teacher pay, and socioeconomic status of the K-12 student population. These factors are frequently seen as important drivers of student achievement.[5] They also provide context for the two states’ policies and test results, which are discussed later in the paper.

Total Per-Pupil Public School Expenditures

Michigan’s total per-pupil public school expenditures were consistently higher than Florida’s from 1990 to 2009 (for simplicity, school years are labeled with the calendar year in which a school year ended).[*] Michigan and Florida were quite similar in 1990, but by 1997, Michigan began to spend significantly more on average than Florida (see Graphic 1). In fact, the Great Lakes State spent at least 20 percent more per pupil in each year from 1997 to 2005, reaching a high of 42 percent more in 2003.[†]

Graphic 1: Total K-12 Public School Expenditures Per Pupil
in Michigan and Florida, 1990-2009*

Source: Author’s calculations based on “Common Core of Data,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/ccd/ (accessed March 28, 2012). Note that these figures are not adjusted for inflation.
* This study designates the school year with the calendar year in which that school year ended. Hence, the 1990 school year refers to the school year beginning in the fall of 1989 and ending in the spring of 1990.

In Graphic 2, the data from Graphic 1 is plotted against trends for the nation as a whole. As the graphic shows, since 2003, the Sunshine State has increased per-pupil spending more rapidly than it had previously.[6]

Nevertheless, in 2009 (the latest year for which data are available), Michigan still spent 12 percent more than Florida. Nationally in 2009, Michigan ranked 20th in total K-12 public school expenditures per pupil, and Florida ranked 33rd.[7]

Graphic 2: Total K-12 Public School Expenditures Per Pupil in the United States, Michigan and Florida, 1990-2009

Source: Author’s calculations based on “Common Core of Data,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/ccd/ (accessed March 28, 2012).
Note: These figures are not adjusted for inflation.

Spending on Teachers

As shown in Graphic 3, average spending[*] on K-12 teachers and other instructional employees[†] in Michigan also exceeded that of Florida. In 2009, Michigan spent $76,022 in total compensation per full-time equivalent instructional employee — 40 percent more than Florida’s average of $54,136. In fact, from 1990 to 2009, Michigan spent at least 25 percent more compensating teachers and other instructional employees every year.

Graphic 3: Salary and Benefit Expenditures Per Full-Time Equivalent K-12 Instructional Employee in Michigan and Florida, 1990-2009

Source: Author’s calculations based on “Common Core of Data,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/ccd/ (accessed March 28, 2012). Note that these figures are not adjusted for inflation.

During this period, money spent compensating teachers in Michigan was consistently above the national average, while money spent compensating teachers in Florida was consistently below the national average (see Graphic 4). Michigan ranked 15th in the nation by this measure in 2009, while Florida ranked 43rd.[8] This disparity remains even after adjusting for differences between the two states’ costs of living.[‡]

Graphic 4: Salary and Benefit Expenditures Per Full-Time Equivalent K-12 Instructional Employee in the United States, Michigan and Florida, 1990-2009

Source: Author’s calculations based on “Common Core of Data,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/ccd/ (accessed March 28, 2012).

Socioeconomic Status of K-12 Student Population

Education research has consistently demonstrated that students’ socioeconomic status is correlated with their performance on standardized tests.[*] The percentage of students who qualify for the National School Lunch Program is a common measure of a group’s socioeconomic status, since eligibility for the program is based on the household income of a student’s parents or guardians.[9] Qualifying students receive a federally subsidized free or reduced-price lunch.

In 2000, the earliest year for which standardized national data are available that include both Michigan and Florida, 44.3 percent of Florida students were NSLP-eligible, while only 29.3 percent of Michigan students were (see Graphic 5). Michigan’s portion of NSLP-eligible students increased more rapidly than Florida’s from 2000 to 2011, but the Sunshine State still maintained a higher proportion of qualifying students in 2011, the latest year for which data are available: 56.0 percent, compared to Michigan’s 45.4 percent.[†]

Graphic 5: Percentage of Students Qualifying for the National School Lunch Program in Michigan and Florida, 2000-2011

Source: Author’s calculations based on “Common Core of Data,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/ccd/ (accessed Marchn 28, 2012).

Throughout the last decade, as Graphic 6 shows, the proportion of Florida students who were NSLP-eligible was greater than not only that of Michigan, but also that of the United States as a whole. Nationally, 47.6 percent of students were NSLP-eligible in 2011.

Graphic 6: Percentage of Students Qualifying for the National School Lunch Program in the United States, Michigan and Florida, 2000-2011

Source: Author’s calculations based on “Common Core of Data,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/ccd/ (accessed March 28, 2012).

Student Achievement in Michigan and Florida

The foregoing data indicate that on average, Michigan schools spent more per pupil, compensated each teacher more and enrolled fewer children from low-income families than Florida schools did. The following analysis compares the achievement of Michigan’s students to that of Florida’s over the past 10 to 15 years.

To compare the achievement of Michigan’s and Florida’s school systems, this report uses scores on the National Assessment of Educational Progress’s math and reading tests[*] for fourth- and eighth-grade students from 1996 to 2011 — in other words, from the mid-1990s, when Florida began its reforms, until the most recent year for which the relevant NAEP data are available.[†] The report also presents the national average for these tests to provide context.[‡]

Note that NAEP provides state-level assessments in other subjects, such as economics, geography, science and writing. NAEP tests 12th-graders periodically as well. The present analysis is limited to reading and math, however, because these are the most consistently tested subjects,[§] and because since 2002, the federal government’s No Child Left Behind Act has required all states to participate in NAEP biennially in both reading and math for fourth and eighth grades.[¶]

Given the statistical relationship between a student population’s socioeconomic status and its average performance on standardized tests, this report analyzes the average score both for all students and for only those students qualifying for a free or reduced-price lunch through the National School Lunch Program.

Fourth-Grade Reading

From 1998 to 2011, Michigan’s average NAEP fourth-grade reading score for all students rose slightly from 216.0 to 218.9, an increase of 1.3 percent.[*] This change, however, was not statistically significant.

Over the same period, Florida’s average NAEP fourth-grade reading score leapt from 205.7 to 224.5, a 9.1 percent improvement. This increase was statistically significant and allowed Florida to easily surpass Michigan’s average score (see Graphic 7). The national average over this period showed some growth, rising from 212.8 to 220.0, a statistically significant 3.4 percent increase.[†]

Graphic 7: Average NAEP Fourth-Grade Reading Scores in the United States, Michigan and Florida, All Students, 1998-2011

Source: “NAEP Data Explorer: Main NDE,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/nationsreportcard/naepdata/ (accessed March 21, 2013).

The average NAEP fourth-grade reading score for lower-income students in Michigan increased from 200.2 in 1998 to 204.5 in 2011, a 2.2 percent increase that was not statistically significant. Florida’s average for lower-income students showed much more improvement, growing from 190.4 to 215.9 over this period — a statistically significant 13.4 percent increase (see Graphic 8). In reading, then, Florida’s lower-income fourth-graders improved at six times the rate of their peers in Michigan and scored well above them by 2011. In fact, the average score for lower-income students in Florida was only 1.4 percent below the average for all fourth-grade Michigan students in 2011. The national average for lower-income students in fourth-grade reading on the NAEP improved from 195.4 in 1998 to 206.9 in 2011, a statistically significant 5.9 percent increase.

Graphic 8: Average NAEP Fourth-Grade Reading Scores in the United States, Michigan and Florida, Students Qualifying for the National School Lunch Program, 1998-2011

Source: “NAEP Data Explorer: Main NDE,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/nationsreportcard/ naepdata/ (accessed March 21, 2013).

Fourth-Grade Mathematics

In fourth-grade math, average scores increased to a statistically significant degree for Michigan, Florida and the nation as a whole from 1996 to 2011. Florida’s gains, however, outpaced both the nation’s average and Michigan’s. In 1996, the average NAEP score in fourth-grade math for all students in Michigan was 226.3. This improved to 236.4 by 2011, a 4.5 percent increase. Over the same period, Florida’s scores grew from 215.8 to 239.8, surpassing Michigan’s and producing growth of 11.2 percent, more than two times that of the Great Lakes State (see Graphic 9). Scores increased nationally from 222.0 in 1996 to 240.1 in 2011, an 8.1 percent rise.

Graphic 9: Average NAEP Fourth-Grade Math Scores in the United States, Michigan and Florida, All Students, 1996-2011

Source: “NAEP Data Explorer: Main NDE,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/nationsreportcard/naepdata/ accessed March 21, 2013).

Average fourth-grade math scores for lower-income students in Michigan grew from 210.0 in 1996 to 223.7 in 2011. This increase represents a statistically significant 6.6 percent gain, the largest improvement seen in this study for Michigan students.

Nevertheless, the Sunshine State’s average fourth-grade math score for lower-income students improved even more — from 203.5 in 1996 to 232.3 in 2011, a statistically significant increase of 14.2 percent. The national average increased less than Florida’s, but more than Michigan’s, climbing from 206.6 to 229.2 — a statistically significant 10.9 percent improvement. Ultimately, in 2011, Florida’s average fourth-grade math score for lower-income students surpassed both Michigan’s and the nation’s (see Graphic 10).

Graphic 10: Average NAEP Fourth-Grade Math Scores in the United States, Michigan and Florida, Students Qualifying for the National School Lunch Program, 1996-2011

Source: “NAEP Data Explorer: Main NDE,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/nationsreportcard/naepdata/ (accessed March 21, 2013).

Eighth-Grade Reading

Average eighth-grade reading scores fluctuated more for both Michigan and Florida. From 2002 to 2011, Michigan’s average eighth-grade reading score on the NAEP for all students did not change in a statistically significant way, moving from 264.7 to 265.2.[*] The average for all Florida students from 1998 to 2011, meanwhile, increased from 254.5 to 262.1, a 3.0 percent increase that was statistically significant. Nationally, the average eighth-grade reading score of 263.6 in 2011 was a new high, and although it represented a statistically significant increase from 260.7 in 1998, it was up only 1.1 percent. Florida moved closer to the national average by 2011, yet still wound up lower than Michigan and the nation (see Graphic 11).

Graphic 11: Average NAEP Eighth-Grade Reading Scores in the United States, Michigan and Florida, All Students, 1998-2011

Source: “NAEP Data Explorer: Main NDE,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/nationsreportcard/naepdata/ accessed March 21, 2013).

Average eighth-grade reading scores for lower-income students declined somewhat in Michigan. Scores dropped from 256.7 in 2002 to 243.7 in 2007, but then rebounded to 253.3 in 2011 — a 1.3 percent net decrease that was statistically insignificant.[†] Florida lower-income students made statistically significant gains, however — average scores grew from 240.7 in 1998 to 253.5 in 2011, a 5.3 percent improvement, ranking it above Michigan and the national average (see Graphic 12). The national average from 1998 to 2011 increased from 244.9 to 251.4, a 2.7 percent gain that was also statistically significant.

Graphic 12: Average NAEP Eighth-Grade Reading Scores in the United States, Michigan and Florida, Students Qualifying for the National School Lunch Program, 1998-2011

Source: “NAEP Data Explorer: Main NDE,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/nationsreportcard/naepdata/ accessed March 21, 2013).

Eighth-Grade Mathematics

With eighth-grade math students, Michigan’s performance differed from both that of Florida and the nation as a whole. From 1996 to 2011, Michigan’s score grew only slightly, starting at 276.9 and increasing to 280.2, a statistically insignificant 1.2 percent increase. Florida’s average score grew from 263.6 to 277.8, a 5.4 percent increase that was statistically significant (see Graphic 13). The national average went from 268.8 to 282.7, a statistically significant 5.2 percent increase.

Graphic 13: Average NAEP Eighth-Grade Math Scores in the United States, Michigan and Florida, All Students, 1996-2011

Source: “NAEP Data Explorer: Main NDE,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/nationsreportcard/naepdata/ (accessed March 21, 2013).

Among lower-income students on the NAEP eighth-grade math test, Michigan fared slightly better than it did among all students, though Florida still outshone the Great Lakes State. Michigan’s average math test score for lower-income eighth-grade students rose from 257.0 in 1996 to 265.6 in 2011, a statistically significant 3.3 percent increase. Florida’s average lower-income student test score in math jumped from 247.9 to 266.9, a statistically significant 7.7 percent improvement (see Graphic 14). Nationally, the average lower-income eighth-grade math score improved from 250.1 in 1996 to 269.0 in 2011, a statistically significant 7.6 percent increase.

Graphic 14: Average NAEP Eighth-Grade Math Scores in the United States, Michigan and Florida, Students Qualifying for the National School Lunch Program, 1996-2011

Source: “NAEP Data Explorer: Main NDE,” (National Center for Education Statistics; United States Department of Education), http://nces.ed.gov/nationsreportcard/naepdata/ (accessed March 21, 2013).

Comparing the Conditions and the Results

Given the conditions of education in Florida and Michigan, one might reasonably have expected Michigan students to outperform those in Florida. Michigan spent more money per pupil, paid its teachers more on average and had a smaller percentage of lower-income students. Michigan schools could theoretically devote more resources to student learning, attract and retain better instructional personnel, and — all other things being equal — expect better average exam scores.[*]

Yet Florida’s average scores for all fourth- and eighth-grade students improved at a faster rate over the period in both reading and math than they did nationally and in Michigan.[†] This was true for Florida’s lower-income students as well.

Further, evidence suggests that Florida’s growth in average test scores was due to students at all levels improving. In other words, Florida’s improvements cannot be solely explained by increases in test scores among just the lowest-performing students. In all four of the subject tests analyzed above, a larger share of Florida students were deemed “advanced” and “proficient” and a smaller share were “below basic” in 2011 than in 1998 or 1996 (depending on which data were available for each subject test).[10] Florida’s rising tide seems to have lifted all boats.

Additionally, in each of the eight areas reviewed above, Florida trailed Michigan’s average scores more than a decade ago, but surpassed Michigan’s in 2009. Although Florida did not retain that advantage in every category in 2011, Florida still topped Michigan in six of the eight (see Graphic 15).

Graphic 15: Florida vs. Michigan on Fourth- and Eighth-Grade
NAEP Math and Reading Scores, 1996-2011

Source: Author’s summary of findings in “Student Achievement in Michigan and Florida.”
*The first recent year in which the two states had comparable tests was 1996, 1998 or 2002. Fourth- and eighth-grade math results were comparable in 1996; fourth-grade reading results were comparable in 1998; and eighth-grade reading results were comparable in 2002.

On the whole, Florida’s performance is striking.[‡] If spending and socioeconomic factors cannot explain it, perhaps Florida’s success lies in the way the money was spent. In other words, the difference between the two states may lie in differences between their education policies.

The Florida Reforms

School Accountability

Florida implemented a new school accountability system in 1998 called the “A-Plus Program.”[11] This system assigns letter grades, A through F, to individual schools to signal how well their students are performing on the Florida Comprehensive Assessment Test, a state-sponsored standardized test.[12] Students in grades three through 10 take FCAT tests every year in reading and math and periodically in writing and science.[13]

Schools that performed well received financial rewards directly from the state.[*] Florida schools that improve by a letter grade or maintain an A were rewarded with an additional $75 per student.[14] Although schools that received a D or an F were provided additional funding,[†] schools that consistently received poor grades faced consequences. For example, if a school receives an F or three D’s in a row, parents of children in that school are given the right to transfer their children to another public school of their choice, such as a charter public school, an in-district public school or an out-of-district public school.[‡]

Further, schools consistently earning an F may have to implement state-imposed reforms, including executing a state-approved “turnaround plan,” contracting with a management company to operate the school or converting to a charter school.15 All schools marked with an F for a single year or a D for three consecutive years are subject to site visits by state-appointed “community assessment teams,” which make recommendations for improvement to the district governing board that controls the school.[16]

The A-Plus Program’s grading formula has encouraged schools to focus on improving the performance of their lowest-achieving students. For most years since 1999 (the first year letter grades were assigned), grades have been a weighted average of three elements of the school’s performance on the Florida Comprehensive Assessment Test: average achievement on the FCAT in all subjects for all students (50 percent); individual learning gains on the FCAT in reading and math for all students (25 percent); and individual achievement growth on the FCAT in reading and math for students in the lowest quartile of achievement (25 percent).[§] Schools that boost the math and reading achievement of their lowest-performing students (and all schools have a lowest quartile), have the best chance of improving their overall performance, since achievement scores for these students affect all three categories.

School Choice

Beginning in the 1990s, Florida made it easier for parents to choose schools outside their school district. The Florida Department of Education boasts that the state “leads the nation in school choice options.”[17] About 780,881 students — 29 percent of all K-12 public school students — in the Sunshine State attended a public school through one of the state’s school choice programs in 2012.[18]

Under Florida’s “controlled open enrollment” policy, parents can choose to enroll their children in an in-district public school other than the one assigned to them based on where they live. Whether parents will have this intradistrict choice is ultimately up to local school boards, but 75 percent of districts do allow for this type of parental choice. About 11 percent of Florida students took advantage of this policy.[19]

Parents in Florida can also choose from a number of charter schools. Charter schools are authorized (“chartered”) by local school districts[*] and funded by the state, but often managed independently under a performance contract. The first five charter schools opened in Florida in 1996, and over the course of the next five years, the number increased dramatically, reaching 201 by 2002.[20] Between 2002 and 2012, the number of Florida charter schools more than doubled, increasing to 518,[21] while enrollment in charter schools more than quadrupled, rising from 40,465 to 179,940 — about 7 percent of Florida’s total K-12 public school enrollment.[22]

Florida also enabled students to use publicly financed vouchers to attend private schools. From 2002 to 2006, students in schools that received two consecutive F’s under the A-Plus Program were eligible to receive a voucher to pay for private school tuition.[†] Since 2000, the state-managed McKay Scholarship Program has also enabled parents of students with disabilities to obtain vouchers to send their children to a private school of their choice.[23] The McKay program served 23,011 children with special needs during the 2012-2013 school year alone.[24]

Parents whose students qualify for a free or reduced-price lunch under the National School Lunch Program can also apply for private school scholarships funded by Florida corporations under a state tax credit program. The Florida Tax Credit Scholarship Program was started in 2001 and provides corporations with a dollar-for-dollar tax credit for the money they donate to “Scholarship-Funding Organizations,” which in turn finance the private school scholarships.[25] Enrollment in this program has nearly tripled since 2007.[26] Now, the tax credit program serves nearly 51,000 low-income students.[27]

Florida also makes a number of online learning options available. Florida Virtual School, which opened in 1997 and primarily supplements course offerings available through local school districts, leads the country in the number of course enrollments for a state-run virtual school with 259,928 in 2011.[28] In 2001, districts were able to set up franchises of FLVS and enroll students full-time in online learning environments.

By 2009-2010, all students in Florida had the ability to complete all of their required coursework online through the District Virtual Instructional Program.[29] For the 2010-2011 school year, Florida had more students enrolled in online programs than did any other state.[30]

Limiting Social Promotion

In 2001, Florida curtailed the practice of “social promotion.” This is the policy of moving all students — no matter their academic proficiency — to the next grade level in order to keep them with their age cohort. Florida law now states, “No student may be assigned to a grade level based solely on age or other factors that constitute social promotion.”[31] Beginning in the 2002-2003 school year, Florida law also specifically required schools to retain third-graders who do not score at a basic proficiency level on the FCAT in reading.[*]

A Focus on Reading Proficiency

Florida Gov. Jeb Bush, through an executive order in 2001, created the Just Read, Florida! Office within the Florida Department of Education. The office is responsible for, among other things, training school reading coaches; improving reading instruction provided by regular classroom teachers; assisting school districts in creating and implementing K-12 reading instruction plans; and providing parents with information about how to assist their children in reading.[32] A new law in 2002 also mandated that all students get immediate remedial reading instruction if needed.[33] The program eventually led to the hiring of 2,000 reading coaches to provide assistance to schools.[34]

Teacher Quality

In 2002, Florida reduced barriers to teaching in the classroom by creating alternative routes to teacher certification.[35] This change enabled schools to hire high-quality teaching candidates who did not have traditional teacher training. “Educator Preparation Institutes” were established to certify teachers who were college graduates or other professionals who did not major in education.[36]

Florida also began certifying teachers trained by the nonprofit American Board for Certification of Teacher Excellence, allowed school districts to create their own certification programs and expanded the reciprocity of approved certifications from other states.[37] There are about 90 different agencies that provided alternative routes to certification, according to the National Center for Alternative Certification.[38] Figures produced by the Florida House of Representatives indicate that district-run alternative certification programs and Educator Preparation Institutes accounted for 37 percent of certifications earned in 2009.[39] Of these new teachers, 68 percent were employed the following school year, compared to 43 percent of those certified through a traditional teacher preparation program.[40]

Lessons for Michigan

No single policy Florida implemented has been shown to explain all of the state’s achievement gains on the NAEP, so Florida’s experience provides no “silver bullet” to improve test scores in Michigan. State policymakers interested in replicating Florida’s success should recognize that this achievement followed a series of reforms, and that in the absence of further research findings, the logical conclusion is to pursue all of them.

Practically speaking, however, it is difficult to imagine all of these reforms being adopted and implemented simultaneously. They were not in Florida; some were adopted years before the others.

This naturally leads to the question, Where should state policymakers start? There is no indisputable answer, but there are several guiding principles that make sense.

First, it seems reasonable to favor Florida policies shown to be effective by high-quality research. Similarly, even though we do not know exactly which reforms may have contributed most to Florida’s success, policies potentially affecting a broad number of students and schools would seem more promising than those more limited in scope.

In addition, given some uncertainty about how the Florida reforms interacted with each other, it may be that the sequence of reforms in Florida mattered. This suggests some wisdom in adopting Florida’s earliest reforms sooner rather than later.

In light of these considerations, some Florida policies would be higher priorities than others. The list of recommendations below starts with the most promising reforms based on these criteria.

1. Private School Choice

Michigan should adopt a private school choice program similar to the one used in Florida. This program could be a tuition tax credit similar to the one already proposed by Mackinac Center analysts.[41]

Private school choice programs in Florida have been shown through rigorous research to have positive impacts on student achievement. Nine studies have examined the impact of the A-Plus Program’s voucher component, and all nine find public schools in Florida improved — particularly the lowest-performing ones — as a result of Florida’s voucher program. A 2008 study also demonstrated that public schools improved as a result of the availability of vouchers for students with disabilities under the McKay Scholarship Program. A study of the Florida Tax Credit Scholarship Program yielded a similar result.[42]

Hence, these private school choice programs had a broad effect on Florida’s school system by requiring all public schools to accept a new level of accountability. The choice programs, which began in 2001, were also among the earlier reforms enacted in Florida.

Private school choice options inject a high level of competition and responsiveness into a public school system by providing highly dynamic and decentralized incentives for schools to better meet families’ educational needs. Parents are empowered to hold schools — especially for schools that are performing poorly — accountable for the actual services they provide their individual children. No other reform mentioned below places such localized pressure on schools to improve.

Unfortunately, Florida’s private school options cannot be immediately adopted in Michigan due to a 1970 amendment to the Michigan Constitution. Article 8, Section 2, of the state constitution explicitly prohibits the use of tuition vouchers or scholarships financed by tax credits for the purpose of educating students in private elementary and secondary schools.[*]

The Michigan Legislature should consider presenting the people with a proposed constitutional amendment that would overturn this prohibition.[†] The legality of well-crafted private school choice programs is no longer a concern under the federal constitution: Both vouchers and tuition tax credits have passed legal challenges in the United States Supreme Court.[43]

It is reasonable to assume that the tuition tax credit program that has been described in detail by Mackinac Center analysts would likewise pass muster with the Supreme Court. The Legislature could implement such a tax credit program following popular repeal of Article 8, Section 2.

2. Public School Choice and Virtual Schooling

Michigan should eliminate remaining barriers to public school choice, making it easier for parents to choose the learning opportunities that will best suit their children. Like private school choice, public school choice has a widespread impact by allowing individual parents to demand new levels of accountability from all public schools.

Michigan does enable parents to choose from a number of conventional public school options. Since 1991, state law has encouraged districts to enable parents to choose the school their child attends within their resident district.[44] Beginning in 1996, parents also had a limited ability to enroll their children in a nearby public school outside their district of residence (though district school boards can still reject nonresident students or limit the number of nonresidents their district schools accept).[*] In this case, Michigan resembles Florida in that both states have implemented fairly extensive parental choice systems for conventional public schools.

And like Florida, Michigan allows charter schools to operate in the state. About 8 percent of public school students in Michigan enrolled in a charter school in 2012.[45] Enrollment in these schools is up 285 percent since 1999, from 31,109 to 119,900 in 2012.[46]

Unlike Florida, however, the Great Lakes State for years effectively limited charter schooling by capping the number of charter schools that could be authorized by public universities at 150.[†] This limit was reached in 1999.[47] Consequently, while the number of charter schools in the Sunshine State grew by 240 percent, from 148 in 2001 to 503 in 2011, the number in Michigan grew by just 46 percent, from 205 in 2001 to 300 in 2011.[48]

Following the passage of legislation in late 2011, Michigan’s cap on the number of university-authorized charter schools is scheduled to end in 2015.[‡] Removing the cap is a sound decision. Research has shown that charter school students in Michigan have made demonstrable gains in student achievement.[49] It should also be noted that charter schools were adopted early in Florida.

One area of distinct difference between Michigan and Florida is in the availability of online schooling options. Admittedly, the effectiveness of online schooling in primary and secondary education has not been rigorously studied (though some studies do exist).[50] Nevertheless, the argument for adopting Florida’s virtual learning reforms in Michigan is strong.

Florida began its foray into virtual learning early, starting in the mid-1990s and rapidly blossoming into a national leader. Moreover, the potential impact of online schooling is broad: When barriers to virtual schooling are removed, students can access online courses from virtually anywhere using an increasing variety of devices with an Internet connection. The ease of taking and switching courses among a wide variety of public and private providers also creates a competitive pressure that will tend to improve the quantity and quality of courses over time.

Michigan already has a statewide virtual school — Michigan Virtual School — but its enrollment levels pale in comparison to Florida’s. In the 2010-2011 school year, Florida Virtual School had nearly 260,000 course-enrollments, while Michigan Virtual School had 17,700.[51] The number of online learning opportunities for students is growing in Michigan, but unlike Florida, Michigan does not require school districts to make these opportunities available to students. Instead, Michigan’s local school boards decide how much access students will have to online instruction.[52]

Additionally, “seat-time”-based pupil accounting requirements deter many Michigan districts from expanding their online programs. As the author has argued elsewhere,[53] the Legislature should abolish this seat-time mandate.

The Legislature should also remove all geographically based barriers to a district’s receiving state aid on behalf of nonresident students — a particularly helpful reform for districts hoping to expand their online learning programs. The cap on the number of online charter schools and the number of students who can enroll in these schools should be removed as well.[54]

3. School Accountability

Florida’s A-Plus Program, which provided school assessment and accountability reforms, was also an important element of the state’s reform agenda. Indeed, the program was linked in part to Florida’s expansion of school choice opportunities.

There is some empirical evidence to suggest that Florida’s school assessment and accountability system has helped improve student achievement. A 2007 study published by the nonprofit Urban Institute showed that students in Florida schools

receiving an F subsequently increased their achievement to a statistically significant degree compared to students in other schools in both reading and math from 2002 to 2003.[*]

Florida’s A-Plus Program also has a broad impact. Every school in Florida is assigned a letter grade and receives performance-based rewards or penalties. The A-Plus Program was also an early element to the Florida reform model.

Michigan has several school assessment and accountability systems,[†] but unlike Florida’s, these are neither clear nor consequential. The guide to understanding the grading system in the Michigan School Report Card spans 29 pages. The explanation of the school accreditation methodology requires 97 pages.[55]

Further, Michigan’s assessments do not substantially differentiate schools. In 2010, 91 percent of Michigan’s general education schools (excluding alternative and special-education schools ) made “adequate yearly progress.”[‡] The same year, 97 percent of schools receiving a grade on the Michigan School Report Card got an A, B or C.[56] No schools were labeled “unaccredited” — in other words, given an F — by the Michigan Department of Education in 2011.[§]

If policymakers wish to continue Michigan’s school assessment and accountability system, they should reform it. The program has at least two major weaknesses: It is difficult to understand, and it does not provide genuine incentives — neither significant rewards nor penalties — for schools and districts to improve their performance.

Rewards could include a per-pupil foundation grant increase to the district — or better yet, extra funding provided directly to the successful school, creating a bonus similar to the one in Florida, where the award money is sent directly to the school, not the district. This school-based funding differs from the usual method of distributing education revenue across the country; typically, money is either collected by, or passed through, school district bureaucracies, which then determine how the revenue is spent.[57]

Consequences for poor school performance should include enabling parents of students in failing schools to use public funds to access alternative educational opportunities. These alternatives should include schools in other districts, charter schools, online schools and if the state constitution is amended, private schools.

Michigan would not be alone in adopting Florida’s accountability reforms. Indiana, Arizona, New Mexico, Louisiana and Utah have implemented letter-grade accountability systems similar to Florida’s.[58]

4. Alternative Teaching Certification

Next in line would be reforming alternative teaching certification. Florida did this in 2002, later than the reforms mentioned above, meaning it may have contributed less to Florida’s success. Nevertheless, alternative certification does have a broad impact on a school system, especially when the program is as extensive as Florida’s, with more than a third of new certificates coming from alternative routes in 2009.[59]

Moreover, research has found benefits to this type of reform.[60] Tim Sass of Georgia State University found evidence that Florida’s alternatively certified teachers have stronger academic qualifications than traditionally trained ones, and that alternatively certified teachers perform as well as — and in some cases, better than — traditionally certified teachers.[61]

Michigan has few genuine alternative routes to teacher certification. Of the 45,600 teachers certified by the state from 2001 to 2006, only 282 received certification through alternative means — less than 1 percent.[62] Not surprisingly, the nonprofit National Council on Teacher Quality gave Michigan’s “expanding the teaching pool” policies an F in 2009.[63] Florida received a B-minus for its policy concerning the teacher labor pool, and it received a C — the highest overall grade of all the states — for teacher quality policies.[64]

Michigan did pass legislation in late 2009 to expand alternative routes to certification.[65] Aspiring teachers were permitted to teach under a temporary “interim” teaching license.[66] While it’s too early to know whether this reform will create a more vibrant teacher labor market, there’s little reason to believe it will. Even under the reform, knowledgeable and talented aspiring teachers still must invest a significant amount of time, money and energy in obtaining the necessary requirements to obtain state-approved certification.[*]

Michigan should reform its teacher certification system to remove unnecessary obstacles to entering the teaching profession. Specifically, Michigan should give local schools, held accountable by parents, more flexibility over the people they hire. Since educational research consistently demonstrates the importance of teacher quality,[67] this is an important reform.

5. Limit Social Promotion for Third-Graders

A social-promotion ban should be the next policy for Michigan to consider. In particular, Florida’s emphasis on third-grade reading has support in the education research literature. Students who fail to master third-grade reading are more likely to struggle in later grades and drop out of high school.[68]

There is some evidence that Florida’s prohibition on social promotion for third-graders who lack reading proficiency has had positive effects on student achievement. A 2004 Manhattan Institute study of the Florida policy found that compared to other students who were socially promoted to the fourth grade, third-graders who were held back made statistically significant gains in math and reading.[69] A follow-up study found that gains were greater in the second year after the retention than in the first, suggesting that third-grade retention can generate lasting positive effects.[70]

Florida’s ban on social promotion from third grade did have a relatively broad impact, affecting every third-grader and arguably improving their results in later years. Nevertheless, it did not have as far-reaching an impact as the reforms recommended above, which tended to affect the performance of entire schools and districts. In addition, this program was not implemented in Florida until the 2002-2003 school year, so it could not have been responsible for Florida’s early gains.

Michigan does not have a statewide policy regarding social promotion for any grade level. The decision about students’ grade levels rests with individual school boards, and their policies regarding the promotion of students may or may not be based on standardized achievement scores.[71]

Michigan should explore ending social promotion for third-graders who have not mastered basic reading skills. This policy would require schools to diagnose learning deficiencies early in a student’s schooling and respond accordingly — a case where an ounce of prevention may be worth a pound of cure. Such a policy also encourages parents to pay close attention to their children’s early educational performance, since very few parents want to see their children retained.

At least nine other states have had similar policies for many years.[72] News reports indicate that Tennessee and Oklahoma have enacted third-grade social-promotion bans based on Florida’s model.[73]

6. Reading Resources

The final policies Michigan should consider are Florida’s programs aimed at improving literacy for all students. These reforms began later, around the same time as alternative certification for teachers and the ban on social promotion.

It is also difficult to determine how broad an impact the programs had. There is not much research demonstrating the policies’ effectiveness. Third-graders retained in Florida based on reading skills did decrease by 40 percent from 2002 to 2007, suggesting that the targeted resources may have helped at least some groups of students.[*] By the same token, this reduction may have been the result of incentives created by the third-grade social-promotion policy itself.[74]

Michigan does not have a similar program, but the state did participate in the federal “Reading First” program. Michigan was granted $25?million or more annually from 2003 through 2010 “to promote high-quality school reading instruction for grades K-3.”[75] According to a 2008 report on the program from the U.S. Department of Education, 165 schools in Michigan participated in 2007.[76]

Florida was also granted federal funds for Reading First, and 584 schools participated.[77] Since both states participated in this federal program, however, it is unlikely to have contributed significantly to the disparity in NAEP gains between the two states.

Michigan also implemented the Michigan Literacy Progress Profile in 2001, which trained teachers to provide better reading instruction and to more accurately assess the reading abilities of students in preschool through third grade.[78] An estimated 10,000 teachers have been trained through this program.[79]

Conclusion

This study highlights the significant improvements achieved by Florida students on the NAEP exam — the “nation’s report card” — over the course of the last 15 years. In fourth- and eighth-grade math and reading, Florida’s students outgained those in Michigan, and by 2009, had surpassed them. Michigan’s performance over the same period was generally lackluster.

If, however, Michigan’s scores had improved at the same rate as Florida’s, the Great Lakes State would be at or near the top of the national rankings on the NAEP. Specifically, Michigan would have been among the top three states in average reading and math scores for fourth-graders, and it would have been among the top 10 states in average math scores for eighth-graders. Michigan’s lower-income students would have achieved similar results compared to those in other states.[80]

The student achievement gains in Florida on the NAEP are important, and they suggest two points worth considering: First, a combination of reforms may work better than any single approach; second, reforming schools is a marathon, not a sprint. In Florida, it took time for the state’s public school system to demonstrate undeniable success. Policymakers there maintained high standards regardless and even implemented new reforms in the process.

Michigan will not be able to match Florida’s gains in student achievement overnight. The Florida experience, however, provides valuable lessons for how the Great Lakes State can improve its public school system.

Acknowledgments

The authors would like to acknowledge the following people for their assistance:

• Marc Holley, evaluation unit director, Walton Family Foundation


• Dr. Matthew Ladner, senior advisor of policy and research, Foundation for Excellence in Education


• Lindsey Burke, Will Skillman fellow in education, The Heritage Foundation


• Audrey Spalding, education policy analyst, Mackinac Center


• Evan Burgess, graphics intern, Mackinac Center


• Grace Kendall, communications intern, Mackinac Center

Although these individuals helped significantly with this study, any errors in the report are the responsibility of the author alone.