Issue Analysis 96 – School Choice: Issues and Answers

Americans are free to make decisions about their health, education, and welfare. We can choose what to eat, how to exercise, whom to marry, where to worship, and which candidate deserves our vote. Students can use federal financial aid to attend the college of their choice – even a seminary! But most Americans have little ability to choose how or where their children will be educated during their most critical formative years.

Government decides where the vast majority of children in this country attend elementary through high school, and the fate of most children is determined by their address. Parents seeking the best school must either pay private tuition on top of the taxes they pay to support the public schools, or pay a premium to move to a neighborhood with good public schools. A lot of children do well under this system, but many do not. The children hurt the most are the most vulnerable – poor children whose families cannot afford an alternative to the public school in their neighborhood. Since public schools receive funding regardless of whether they do a good job, they face little pressure to improve. As usual, quality suffers when a substantial number of customers have no option to go elsewhere.

In an advanced Information Age economy, lifetime earnings prospects depend on the quality of education more than ever before. It should be no surprise that school choice has recently enjoyed growing support among parents and the general public.

What Makes a Good School?

To answer this question, it’s important to understand what does not make a good school. Simply spending more money does not improve performance. Since the 1970s, spending per student has steadily risen and teacher-pupil ratios have steadily fallen. Yet over that same period, standardized test scores remained unchanged. International comparisons reveal that, if anything, larger teacher-pupil are associated with better academic performance! A much-discussed Tennessee experiment found that reducing class sizes from 25 to 15 students increased achievement for kindergarten students but had no effect in higher grades.1 The experience of homeschoolers, meanwhile, suggests that teacher credentials make little difference; 25 percent of homeschooling parents are certified teachers, but their children perform no better on standardized tests than homeschooled children whose parents are not certified teachers.2

Of course, one might argue that student performance would have gotten worse if the amount of resources devoted to education had not risen. To assess the effect of inputs, we really need to look at studies that control for the effects of other factors. Eric Hanushek, an education economist at the University of Rochester, examined most of the major studies on school effectiveness and found that common panaceas like more money, smaller classes, and more teacher training rarely make a difference.

Studies Show More Inputs do not Improve Student Performance

Input Measure

Number of Studies

Studies Showing Positive Effect

Studies Showing Negative or no Effect

Teacher-Pupil Ratio




Teacher Education




Teacher Salary




Expenditure per Pupil




Source: Eric Hanushek, “Assessing the Effects of School Resources on Student Performance: An Update,” Educational Evaluation and Policy Analysis 19(2), Summer 1997, pp. 141-164.

The most important factor affecting student performance is the school’s organization and management. Effective schools generally exhibit commitment to a clearly defined academic mission, motivated leadership, and teacher participation in decisions. Schools most likely to have these characteristics are the schools that are most free from interference by higher levels of the educational bureaucracy.3 Three kinds of schools have this freedom: private schools, charter schools, and regular public schools not part of large systems or districts. If we want all children to have equal educational opportunities, all parents must have the ability to choose these schools.

Forms of School Choice

Public school choice: Open enrollment, magnet schools, and other programs within the regular public system allow parents to send their children to alternative public schools. Eighteen states offer choice among public schools throughout the state, and 11 more permit choice within or among districts. Other states impose harsh penalties on parents attempting to exercise choice; in Illinois, low-income parents who try to get their children into better public schools by “fraudulently” registering them in a school district where they do not live face a fine of $500 and a 30-day jail sentence.4

Child-centered funding: Allows parents to choose their children’s school and then funds each public and/or private school based on the number of children enrolled. Arizona adopted the first child-centered funding plan for public schools in 1998.5 One of the most extensive school systems using child-centered funding is the entire nation of New Zealand, which permits full school choice. The New Zealand government abolished all administration above the level of the individual school, and funds both government and private schools based on enrollment.6

Charter schools: A charter school is a public school that operates independently of the education bureaucracy’s rules and teachers’ union contracts. A charter school can only educate students if parents choose to send their children there, and it receives public funding based on enrollment. The school’s charter specifies the results for which it will be held accountable. If the school fails to deliver on these results, it receives no more public money. Approximately 1,200 charter schools serve more than 300,000 students in 27 states and the District of Columbia. Currently, 38 states have laws permitting charter schools.7

Home schooling: Between 700,000 and 1.5 million American children are home-schooled by their parents or tutors.8 That’s three times as many homeschoolers as in 1990, and nine times the number in 1978.9 Some of the most prominent figures in American history received most of their elementary education at home, including Thomas Edison, Andrew Carnegie, Thomas Jefferson, Woodrow Wilson, and both Theodore and Franklin D. Roosevelt.10

Vouchers: Vouchers are scholarships that let students attend a school of their parents’ choice. The principal difference between vouchers and child-centered funding is the administrative mechanism. In a voucher system, schools redeem vouchers given to them by students; under student-centered funding, schools simply tell the government how many students they have and receive funding accordingly. Publicly funded vouchers have been used for decades in two states, New Hampshire and Maine, and two cities, Cleveland, Ohio (since 1996), and Milwaukee, Wisconsin (since 1990).

Tuition tax credits/deductions: A direct reduction in taxes that fully or partially offsets expenditures by parents or others on tuition and educational materials, such as books and computers. Minnesota, Iowa, and Arizona currently offer tuition tax credits or deductions.

Tax-favored savings: Various proposals allow parents to defer or eliminate taxes on investments set aside to pay education expenses. The most common proposal would work much like the “Education IRA” that is now available to pay for postsecondary education.

Private scholarships: Since 1991, the Children’s Educational Opportunity Foundation has provided scholarships that allow children from low-income families to attend private schools. This fall 36,875 children will receive more than $140 million worth of scholarships in 37 city-based and three statewide programs. One of the most ambitious programs is in San Antonio, Texas, where business leaders have pledged $50 million over ten years to allow every low-income student in the Edgewood Independent School District to attend the school of his or her choice.11

Results of School Choice

Many forms of school choice have been subjected to rigorous academic study, and the results are simple. School choice works:

Students’ test scores are higher, and school spending and taxes are lower, in metropolitan areas where the existence of a large number of independent school districts makes it easier for parents to choose among public schools by moving to another neighborhood.12

Massachusetts has permitted parents to choose among public schools since 1991, and the district losing a student must pay up to $5,000 to the district accepting the student. School districts with the largest losses responded by changing their policies and programs in ways that brought back many students in the ensuing years.13

A nationwide survey of charter school students and parents revealed that 80 percent of students who were failing at their previous schools performed average or better in charter schools. More than half of them are now doing “good” or “excellent” work. Results improved for children in all ethnic groups and income levels.14

After four years in a state-funded voucher program in Milwaukee, low-income students performed 11 percentile points better in math and 6 points better in reading than similar students who did not receive vouchers. If the program produced such gains for all students over 12 years of schooling, it would eliminate more than half of the gap in reading performance and all of the gap in math performance between white and minority students.15

After one year, Cleveland students receiving state-funded vouchers scored higher in language, math, reading, and social studies than similar public school students. At two new schools created to serve voucher students, students scored 5.7 percentile points higher in reading and 8.6 percentile points higher in math after one year.16 In contrast, percentile scores of inner-city public school students often fall by a point or two a year.17

After one year, low-income New York City students using private scholarships to attend private schools scored 2 percentile points higher in reading and math than similar students in the public schools. Fourth and fifth graders did even better, with average math scores 5.9 percentile points higher and reading scores 4 points higher.18

Every year students remain in San Antonio’s regular public schools, their standardized test scores fall by 2 percentile points in math and 1.4 points in reading. But low-income students receiving privately funded scholarships or choosing to enroll in the city’s “multilingual” schools keep up with their nationwide peers, retaining the same percentile scores from year to year.19 Similar results occurred in Indianapolis, where the performance of low-income middle school students receiving private scholarships held steady while that of their public school peers deteriorated.20

Home-schooled students score higher on standardized tests than either public or private school students, and their scores increase more rapidly as they progress from one grade to another. One in four homeschoolers is at least one grade ahead of other students of the same age, and only 5 percent are working below the grade level for their age. The longer students are home-schooled, the better their performance, even after accounting for the fact that homeschooling parents are more highly educated than the general population.21

In 1999, home-schooled high school seniors tied Rhode Island’s Seniors for the highest average scores on the ACT college admissions test.22

School choice also improves regular public schools by encouraging innovation and competition:

A new program in Florida offers state-funded vouchers to students at schools who received “failing” grades in two out of four years. Two schools lost students due to vouchers in 1999; one responded by adopting a sharper focus on reading, writing, and math. All teachers in the school spend 90 minutes each morning teaching reading. One mother who took advantage of the vouchers noted, “It took competition to make improvements…Everyone’s focus now is the kids. It’s a totally different scene. Why didn’t they do it before?”23

California’s first charter school, the San Carlos Charter Learning Center, developed curriculum and instructional approaches that have been widely copied in its district. The Los Angeles School District revised its purchasing system after a “60 Minutes” story revealed that a charter school principal purchased computers for her school at lower cost than the school district and in much less time – six days instead of a year!

In response to a flurry of charter school activity in Boston, the school district and teachers’ union launched a “pilot schools” project that offered waivers of district policies and union contract provisions. Other Massachusetts public schools have been forced by charter school competition to open kindergartens and shorten student vacations.24

When parents began applying for vouchers in Cleveland, the public school system offered applicants a chance to place their children in magnet schools and enrichment programs.25

Arizona has the nation’s most progressive charter school law and the largest number of charter schools. A survey of teachers in Arizona’s regular public schools revealed that principals are much more likely to inform parents about school programs, encourage experimentation in teaching, and involve teachers in decisionmaking when the school faces strong competition from charter schools.26

In Milwaukee, the public school system seeks to retain students by advertising that they will pay for a reading tutor for any child who does not learn to read in regular classes.

Parents Want School Choice

Poll results, surveys of parents in school choice programs, and waiting lists for choice programs all demonstrate that a substantial number of parents want school choice.

Poll results

Support for school choice is especially strong among people who have the most to gain from increased education opportunities. For example, a survey by the Joint Center for Political and Economic Studies found that 57 percent of African-Americans, 65 percent of Hispanics, and 47 percent of whites favor school choice.27 Similarly, a 1998 survey of African-Americans in Philadelphia found that 64 percent support the use of vouchers for any school, including religious schools, and 68 percent said they would opt out of the public school system if cost were of no concern.28

Parental satisfaction

Surveys consistently show that parents whose children participate in school choice programs are much more satisfied with their chosen schools:

Cleveland parents whose children received vouchers expressed much more satisfaction with their chosen schools’ academic quality, safety, discipline, and attention paid to their children than parents whose children were still in the public schools. More than 80 percent of parents planned on sending their children back to their chosen school the following year.29

Half of the parents of randomly chosen low-income students receiving privately funded scholarships in New York City give their schools a grade of “A,” compared to only 12.5 percent of parents whose children applied for, but did not receive scholarships. More than 80 percent of scholarship recipients plan to attend the same school the following year, compared to 70 percent of similar public school students.30

Surveys of low-income applicants for privately funded scholarships in Washington, DC, and Dayton, Ohio, show that parents of students in private schools are much more satisfied with their schools’ academic quality, safety, discipline, and teaching than parents of students in public schools. Half of private school parents gave their schools an “A,” but less than a fifth of public school parents did so.31

In San Antonio, Texas, low-income parents whose children received private scholarships reported much higher satisfaction with their children’s learning and school discipline than parents with children in public schools.32

Waiting lists

School choice programs typically have many more families interested in participating than can be accommodated. Such excess demand reveals that many parents are well-informed and concerned about their children’s education, and want options for their children other than their assigned public schools.

Charter schools in Michigan, for example, usually have three or four applicants for every available seat.33 A nationwide survey revealed that more than 70 percent of charter schools had waiting lists in the 1996-97 school year.34

Publicly funded school choice programs typically receive more applications from parents than they can serve. Cleveland’s state-funded voucher program received 6,244 applications for 2,000 partial scholarships in 1996, the first year they were available. Most parents – 85 percent – said they applied in order to improve the academic quality of their children’s education.35

Another demonstration of the pent-up demand for school choice comes from private scholarship programs that help children from low-income families attend non-public schools. These programs typically offer only partial scholarships; parents have to come up with a portion of the tuition. Yet wherever tried, the response is overwhelming. In New York City, 20,000 low-income public school students applied for 1,300 scholarships in the first year of the program.36 In Baltimore, 20,000 children applied for 500 scholarships awarded in 1999 – that’s 44 percent of all the children whose families had incomes low enough to qualify! Nationwide, 1.25 million children applied for 40,000 scholarships for the 1999-2000 school year.37 The actions of low-income parents speak louder than any polling results; a substantial number of them want school choice!

Mixing Church and State?

A common fear about school choice is that it leads to unconstitutional taxpayer support for religion, because many private schools are affiliated with churches. This concern is misplaced.

Several recent court decisions have determined that choice plans are constitutional if aid under the plan serves a secular purpose, is given according to criteria that neither favor nor disfavor religion, and flows to religious institutions as a result of parental or student choice. Those were the Wisconsin Supreme Court’s reasons in 1998 for upholding the constitutionality of the state-funded voucher program in Milwaukee, which permitted low-income students to choose religious schools.38 By a vote of 8 to 1, the U.S. Supreme Court declined to hear the case, allowing the Wisconsin decision to stand.

Applying the same three-pronged test, the Ohio Supreme Court in May 1999 decided that the basic structure of the Cleveland voucher program was constitutional.39 In January 1999, the Arizona Supreme Court upheld a law granting income tax credits for donations to private scholarship programs. The court reasoned that the principal purpose of the program is not to promote religion, but to aid students and parents by making private schools more accessible.40

These results should come as no surprise, since college students generate no constitutional controversy when they use federal aid to attend religiously-affiliated colleges. Some veterans have even used their G.I. Bill benefits to attend seminaries, with no constitutional problems. School choice merely gives the parents of elementary and high school students the same kinds of options that college students enjoy.

Perhaps more surprising is a federal injunction in August 1999 that halted the Cleveland program. After parental protests, the judge allowed it to continue for students who received vouchers in the previous school year. The judge’s final ruling will likely set the stage for a series of appeals that could reach the U.S. Supreme Court.

Does School Choice Threaten School Independence?

Some critics of school choice – usually conservatives – have argued that government assistance to students in charter or private schools will inevitably bring greater regulation of school management, curriculum, teacher credentials, and similar factors. Instead of improving regular public schools, such regulation might simply reduce the flexibility and diversity of charter and private schools. This is a real danger that must be dealt with in the design of choice programs.

Some forms of school choice are more vulnerable to this problem than others. Charter schools are in greatest danger, because they are still public schools, and school boards or legislators can always opt to change the rules of the charter school program. For private schools, vouchers may carry a greater danger of regulation than tuition tax credits. In voucher programs, government money flows to private schools as a result of parental decisions. Tuition tax credits, on the other hand, simply allow parents to keep more of their own money if they spend it on private school tuition. Since the money never passed through the government’s hands, it is harder for government to use tuition tax credits as a vehicle to promote greater regulation of private schools.

What About Racial Discrimination?

School choice arouses suspicions in some circles because racist southerners used school choice plans in the 1960s to evade the Supreme Court’s dictum that schools must be desegregated.41 Under such plans, a few African-American students typically chose to attend historically white schools, but no white students chose to go to the historically black schools, with the result that the system remained virtually segregated.

But private schools and charter schools are not the same as historically segregated public schools. Both Catholic and non-Catholic private schools are more racially diverse than public schools, and have been since at least the early 1980s. Students of different racial backgrounds are even more likely to sit next to each other in private school lunch rooms – especially in religious schools.42

Similarly, studies by the U.S. Department of Education and the Hudson Institute both found about half of charter school students are members of minority groups, compared to 34 percent in regular public schools.43 The U.S. DOE concluded that about 60 percent of charter schools serve the same racial mix as their surrounding district, and most of the rest have a higher percentage of minority students.44

In addition, private schools have strong incentives to avoid discrimination precisely because discrimination is a highly sensitive issue that could undermine support for a school choice program.

Numbers Aren’t the Only Issue

Some critics resist educational choice out of a belief that decisions about the education of other people’s children are too important to be left to parents. This is a truly frightening argument if we consider the implications for other, equally important, decisions that people make. Nutrition is critically important to both health and intelligence, but we do not establish “public food stores” as a tax-supported monopoly to control eating habits. People’s choice of religion plays a large role for good or ill, but history shows that establishing a tax-supported monopoly on religion is a recipe for social strife that actually reduces genuine religious belief.


The argument for educational choice is not just that choice improves academic achievement, but also that choice accommodates different parents’ diversity of views about the types of education that they want for their children. The extensive Catholic school systems in the United States, for example, were established largely because poor Catholic immigrants feared that public schools would indoctrinate their children in Protestantism.45 In present-day America, educational choice would put an end to seemingly endless arguments over sex education, textbook choice, and censorship in school libraries; parents could simply choose schools that reflect their values instead of fighting to impose their values on their neighbors’ children. The result would be an immeasurable but long-overdue increase in civility and social harmony.

1Eric A. Hanushek, “The Evidence on Class Size,” Occasional Paper Number 98-1, W. Allen Wallis Institute of Political Economy, University of Rochester, February 1998, (

2Lawrence M. Rudner, “Scholastic Achievement and Demographic Characteristics of Home School Students in 1998,” Education Policy Analysis Archives 7:8 (March 23, 1999), p. 23, (

3John Chubb and Terry Moe, Politics, Markets, and America’s Schools (Brookings Institution, 1990), pp. 152-56.

4Nina Shokraii Rees and Sarah E. Youssef, School Choice: What’s Happening in the States 1999 (Heritage Foundation, 1999), p. 34.

5School Choice: What’s Happening, p. 5.

6Maurice P. McTigue, “Charter Schools Kiwi-Style,” The Oregon Spectator (Feb./March 1999), p. 5.

7Statistics from the Center for Education Reform, (, and (

8Rudner, “Scholastic Achievement,” p. 2. Also Web site of the Home School Legal Defense Association, (

9National Home Education Research Institute, Fact Sheet IIB 1999 (

10Andrew J. Coulson, Market Education: The Unknown History (Transaction Publishers, 1999), pp. 119-20.

11Children’s Educational Opportunity Foundation of America, Children First (Winter 1999).

12Caroline Minter Hoxby, “Does Competition Among Public Schools Benefit Students and Taxpayers?” National Bureau of Economic Research Working Paper No. 4979 (December 1994).

13David J. Armour and Brett M. Peiser, “Interdistrict Choice in Massachusetts,” in Paul E. Peterson and Bryan C. Hassel (eds.), Learning from School Choice (Brookings Institution, 1998), pp. 157-86.

14Gregg Vanourek, Bruno V. Manno, Chester E. Finn, Jr., and Louann A. Bierlein, “The Educational Impact of Charter Schools,” Charter Schools in Action Project: Final Report, Part V (Hudson Institute), pp. 13-16. (Referred to henceforth as “Hudson Institute study.”)

15Jay P. Greene, Paul E. Peterson, and Jiangtao Du, “Effectiveness of School Choice: The Milwaukee Experiment,” Harvard University Program in Education Policy and Governance Occasional Paper 97-1 (March 1997), (

16Paul E. Peterson, Jay P. Greene, and William Howell, “New Findings from the Cleveland Scholarship Program,” manuscript (May 6, 1998), Harvard University Program in Education Policy and Governance, pp. 7, 9-10, available (

17Jay P. Greene, William G. Howell, and Paul E. Peterson, “Lessons from the Cleveland Scholarship Program,” manuscript (Oct. 15, 1997), Harvard University Program in Education Policy and Governance, pp. 23-25, available (

18Paul Peterson, David Myers, and William G. Howell, “An Evaluation of the New York City School Choice Scholarships Program: The First Year,” Mathematica Policy Research and Harvard University Program on Education Policy and Governance (Oct. 1998), pp. 26-27, (

19R. Kenneth Godwin, Frank R. Kemerer, and Valerie J. Martinez, “Comparing Public Choice and Private Voucher Programs in San Antonio,” in Paul E. Peterson and Bryan C. Hassel (eds.), Learning from School Choice (Brookings, 1998), pp. 286-87.

20David J. Weisschrott and Sally B. Kilgore, “Evidence from the Indianapolis Voucher Program,” in Paul E. Peterson and Bryan C. Hassel (eds.), Learning from School Choice (Brookings, 1998), pp. 330-32.

21Median scores for homeschoolers are higher than median scores for other students. This result holds for all scores examined – reading, language, math, social studies, and science. See Rudner, “Scholastic Achievement,” pp. 14-19.

22Andrea Billups, “Students Schooled at Home Ace ACT,” Washington Times (Aug. 18, 1999).


24Hudson Institute study, pp. 17-18.

25Greene, Howell, and Peterson, “Lessons from Cleveland,” pp. 13-14.

26Frederick Hess, Robert Maranto, and Scott Milliman, “Coping With Competition: How School Systems Respond to School Choice,” manuscript (1999).

27Paul E. Peterson, “A Report Card on School Choice,” The State Factor (Summer 1998), p. 2.

28Gary A. Furgeson, “The Commonwealth Foundation 1998 Survey on Education Issues Among African Americans in Philadelphia,” American Viewpoint, Inc. (April 1998).

29Greene, Howell, and Peterson, “Lessons from Cleveland,” pp. 14-19.

30Peterson, Myers, and Howell, “An Evaluation of the New York City School Choice Scholarships Program,” pp. 24-15.

31Paul E. Peterson, Jay P. Greene, William G. Howell, and William McCready, “Initial Findings from an Evaluation of School Choice Programs in Washington, DC and Dayton, Ohio,” manuscript (Oct. 24, 1998); “Initial Findings from an Evaluation of School Choice Programs in Washington, DC,” manuscript (Sept. 1, 1998), Harvard University Program on Education Policy and Governance, (

32Godwin, Kemerer, and Martinez, “Comparing Public Choice and Private Voucher Programs,” pp. 280-82.

33Hudson Institute study, p. 16.

34U.S. Dept. of Education, Office of Educational Research and Improvement, A National Study of Charter Schools, 1998, Executive Summary, p. 12.

35Greene, Howell, and Peterson, “Lessons from Cleveland,” pp. 4, 10.

36Petersen, Myers, and Howell, “An Evaluation of the New York City School Choice Scholarships Program,” p. 2.

37Baltimore and nationwide statistics are from private communications with Towson State University professor Howard Baetjer, a board member of the Baltimore organization administering the scholarships.

38Jackson v. Benson, 578 N.W.2d 602 (Wisconsin 1998).

39The Court ruled that one minor provision of the plan violated the First Amendment – a section that allowed private schools with waiting lists for entry to give priority to children whose parents belong to organizations that support the school. The Court believed this provision may create incentives for parents receiving vouchers to alter their religious affiliations in order to get their children into particular schools. See Doris Simmons-Harris v. John Goff, Case 97-1117, Ohio High Court (May 27, 1999). The Court struck down the voucher program because its enactment as a rider on an appropriations bill violated the Ohio Constitution’s “single-subject” rule, and the legislature then re-enacted the program in a freestanding bill.

40Kotterman v. Killian, Arizona Supreme Court No. CV-97-0412-SA.

41Coulson, Market Education, pp. 136 and 275.

42Coulson, Market Education, pp. 276-77.

43Hudson Institute study, p. 2.

44U.S. DOE Study, Executive Summary, p. 11.

45Jerome Ellig and Jack High, “The Private Supply of Education: Some Historical Evidence,” in T. Cowen (Ed.), The Theory of Market Failure: A Critical Appraisal (George Mason University Press, 1988).