Introduction

Americans overwhelmingly believe that parents know best how to raise young children. This is confirmed in poll after poll. ("For Better or Worse, Results from a Newsweek Gallup poll, 1989," Newsweek, Special Issue on Children, Summer 1990, p. 18.) Experts, too, agree that the family home is the optimum setting for early childhood development.

Yet the federal government ignores this definitive data. In fact, the federal government is greatly increasing its efforts to take very young children out of their homes and put them, for part of the day or all day, into government-funded institutions. Washington is even launching a campaign to persuade parents to part with their sons and daughters at ever earlier ages.

These actions amount to a federal family policy that undermines families. And this policy is becoming more expansive, aggressive, and expensive. Examples:

Activists are seeking to increase funding beyond the $23.5 billion already authorized for child care and early education programs for 1991-1995. The total includes $4.5 billion to set up child care bureaucracies at the state level.

George Bush is asking for a nearly $700 million increase in the fiscal 1992 federal budget for Project Head Start, already the largest federally funded early childhood development program. The program would cost just over $2 billion.

The Bush Administration wants to spend $60 million in fiscal 1992 on Even Start, a Department of Education program similar to Head Start. The budget request is a 200 percent increase over fiscal 1990 and a 21 percent hike over fiscal 1991.

A congressionally appointed, taxpayer-funded panel is considering ways to conduct national academic testing of 3-year- olds.

A taxpayer-financed study for the Department of Education pleads the case for a federally directed, federally subsidized system of national child care standards and early childhood education services. The Department is listening to no opposing views.

Despite a glaring lack of scientific evidence that early childhood programs help children now or in the future, as their sponsors claim, the federal government is spending ever more taxpayers' money on them.

To be sure, for severely disadvantaged children, Head Start and other programs initially improve academic performance. But studies show that the impact does not endure for more than a year or two. Despite this, Head Start is often touted as a "magic bullet" that helps poor children succeed in school. Based on the available evidence, Head Start may deserve support as a welfare program providing nutrition and care for poor children, but not as an educational jump start.

Says J. Craig Peery, professor of human development at Brigham Young University and the author of the 1981 Project Head Start reauthorization bill: "I think we know a lot about early childhood education -- and it doesn't work." (Telephone interview, March 7, 1991.)

Potentially Harmful
Many experts feel that early education that emphasizes performance is not only not beneficial but potentially harmful. Says pediatrician and author Benjamin Spock: "When we instruct children in academic subjects, or in swimming, gymnastics, or ballet, at too early an age, we miseducate them; we put them at risk for short-term stress and long-term personality damage for no useful purpose. There is no evidence that such early instruction has lasting benefits, and considerable evidence that it can do lasting harm." (Quoted in David Elkind, Miseducation, Preschoolers at Risk (New York: Alfred A. Knopf, 1988), p. 4.)

Aside from one small study that finds significant long-term gains, even the most positive analyses of early childhood programs acknowledge few lasting effects. Indeed, the most comprehensive Head Start study compares school testing of Head Start children with those not enrolled and concludes: "by the end of the second year there are no educationally meaningful differences." (Executive Summary, The Impact of Head Start on Children, Families and Communities, Head Start Evaluation, Synthesis and Utilization Project, U.S. Department of Health and Human Services, June 1985, p. 6.)

Out-of-Home Problems
In contrast, research shows that children in out-of-home programs for extensive periods tend to be more aggressive, have more psychological problems, favor peer groups over parental authority and get into more fights than home-reared children. (Nicholas Zill, Developmental, Learning and Emotional Problems: Health of Our Nation's Children, cited in "Family-Induced Disturbances Affect One in Five Children," Education Daily, Vol. 23, No. 238, December 11, 1990, p. 3. Also, Deborah Vandell and Mary Corasaniti and others cited by J. Craig Peery in "Children At Risk: The Case Against Day Care," The Family in America, Rockford Institute, February 1991.) Young children are also more likely to contract serious illnesses in institutional settings. (Reed Bell, "Health Risks From Daycare Diseases," in Phyllis Schlafly, ed. Who Will Rock the Cradle?, Two Conferences on Childcare (Eagle Forum Education and Legal Defense Fund, 1981), pps. 115-122.)

Even Yale psychologist Edward F. Zigler, widely known as the foremost proponent of early childhood development programs, warns that such programs may not be suitable for all children: "There is a large body of evidence indicating that there is little if anything to be gained by exposing middle-class children to early education." (Edward F. Zigler, "Formal Schooling for Four-Year-Olds? No" in Sharon L. Kagan and Edward F. Zigler, eds., Early Schooling (New Haven: Yale University Press, 1987), p. 30.) He adds: "We must listen to those families who neither want nor need their young children placed in preschool .... whenever the family situation permits it, the best place for a preschool child is often at home." (Zigler, op. cit., p. 34.)

Warning Bells
The virtual absence of proof that programs have long-term benefits and growing evidence that children in institutional settings are at risk should ring warning bells in Washington. It should make the federal government proceed with caution before funding existing and new programs for young children -- until research proves that such programs not only do not harm children but significantly help them.

Yet this evidence routinely is ignored at the Departments of Education and Health and Human Services, and sometimes even at the White House. Ignored too is the fact that the most important cultural problem facing America is dysfunctional families -- the families in which children are most often at risk.

What is puzzling is that some leading officials rhetorically admit that the well-being of American families is the key to a healthy society, even while the agencies run by these officials pay scant heed to the rhetoric. Louis W. Sullivan, Secretary of Health and Human Services, earlier this month told the annual legislative conference of the National League of Cities: "So many of our problems -- drug and alcohol abuse, the spread of AIDS, teen pregnancy, infant mortality, youth homicide, among others -- reflect the personal isolation, alienation and despair that follow from the widespread erosion of family and community." (Paul Taylor, "Family Seen as Key to Aiding Children," The Washington Post, March 11, 1991, p A-5.) Adds Minneapolis Mayor Donald M. Fraser: "We know we have a welfare system that discourages marriage.... we have a [federal] public policy that is designed to destroy families and nobody talks about it." (Ibid.)

Sullivan, a Republican, and Fraser, a Democrat, both call for the abolition of federal programs that weaken families. (Both men, however, support expansion of Head Start. Sullivan calls it a "middle ground" that shores up families without supplanting them.) This call merits very broad support. To succeed, it requires a strategy whose first steps include:

1) Ending the funding of all early childhood education programs except those for the poor.

2) Supporting research into early education for disadvantaged children to determine if there is a model that works and can be replicated. Such efforts should target only severely disadvantaged children, follow methods for which there is some evidence of success, and use a controlled experiment format.

3) Evaluating independently all existing federally subsidized early childhood education programs. Conflicts of interest should preclude bureaucracies that administer programs from evaluating their effectiveness.

4) Expanding the use of vouchers or tax credits to allow parents to choose the best education and care for their children. Parental involvement should be a paramount consideration in any program. Early childhood education may indeed be worthwhile for some disadvantaged children, but studies indicate parental involvement with children is the key developmental factor.

5) Reducing federally imposed financial burdens on the family. In 1948, a family of four of median income paid 2 percent of its income to federal taxes. This year, that family pays 24 percent in federal taxes in addition to state, local, and excise taxes. (Robert Rector, "Fourteen Myths About Families and Child Care," Harvard Journal on Legislation, Vol. 26, No. 2 (Summer 1989), p. 542.) Some ways to aid families include: increasing the personal exemption for dependents, increasing the earned income tax credit and extending it to more families, and redesigning welfare programs to encourage fatherhood, not father abandonment. Minnesota, for instance, has waived welfare rules that have been driving fathers out of the house.

The Federal Role

Despite lack of evidence that early childhood programs substantially benefit children, the U.S. government agencies actively promote these programs in several ways.

* The Tax Credit for Dependent Care allows parents to deduct from their gross income a portion of child care expenses on their federal returns. The deductions totalled $3.895 billion in fiscal 1990 and are projected to total $4.395 billion in fiscal 1992. (Telephone interview with Terry Harrow, legislative branch director for Human Development Resources, U.S. Department of Health and Human Services, March 12, 1991.) The program has the merit of giving parents economic assistance to choose the kind of child care they want, including religious- affiliated facilities. The trouble is that the program discriminates against parents caring for children at home. In effect, the government says: "If you leave your child with someone other than one of the parents (or grandparents, or aunts, or neighbors), you get a tax break."

* Child care provisions in the 1991 Omnibus Budget Reconciliation Bill will cost an estimated $23.6 billion over five years for several programs. (Other provisions are: a Child Health Insurance Credit. This $5.2-billion program assists families in buying health insurance for children. In 1994 the maximum credit will be $483. A "Wee Tots" Supplemental Tax Credit. This $700 million program gives families with a child under one year a supplemental credit or allows them to choose the Dependent Care Tax Credit. By 1994, the "Wee Tot" maximum credit will be $403.) Among them:

1) The Earned Income Tax Credit. Projected to reduce federal tax revenues by $12.4 billion over five years, this gives a small measure of tax relief (a direct refund of up to a maximum of $953) to poor families with children.

2) Child Care and Development Block Grant. Estimated to cost $4.45 billion over five years, this gives money to states based on the number of low-income children in the state under age five. Some 75 percent of the money can be used for services to children; the other 25 percent is targeted for quality control, early childhood programs, and day care for latch-key children. Because of fuzzy wording, the money for children's services could be siphoned into campaigns for a child care bureaucracy.

3) Title IV Family Support Ways and Means Grant. Allocating $1.5 billion over five years, this gives money to poor working families for child care to keep parents off welfare rolls.

4) Standards and Training Ways and Means Grant. At a cost of $250 million, this will help states improve standards for child care, monitor compliance, and provide training for child care providers.

Head Start

The largest federally funded early childhood development program is Head Start. Begun in 1965 as part of Lyndon Johnson's War on Poverty, Head Start is a comprehensive early childhood development program providing children ages 3 to 5 from low-income families with educational, social, medical, dental, nutritional, and mental health services. (Head Start: A Child Development Program, pamphlet, U.S. Department of Health and Human Services, 1990, p. 2.) In the past quarter-century, more than 11 million children have been Head Start clients. Beginning with a summer-only enrollment of 561,000 and a budget of $96.4 million (Project Head Start Statistical Fact Sheet, January 1990, p. 2.) the program grew rapidly. By 1966 the budget was $198.9 million, enrollment rose to 733,000, and the program expanded to full-day. (Ibid.) In 1982, when seasonal programs were replaced by year-round-only services, 395,800 children were enrolled in the program, which had a budget of $911.7 million. (Ibid.) During the 1990 school year, more than 488,000 children were enrolled at a estimated average cost of $2,767 per child (Ibid., pp. 1,2.) while the budget grew to $1.386 billion. The program has a paid staff of 79,549, plus 615,000 volunteers. (Ibid., pp. 2,3.) Some 54 percent of Head Start families are headed by a single parent. Handicapped children account for 13.5 percent of enrollees. Up to 10 percent of Head Start slots can be filled by children not deemed at risk.

Head Start gives children hot meals, comprehensive health care including dental treatment, and social activities. Parents are offered positions on policy-making boards, workshops on child care and nutrition, and training for Head Start jobs. (Serving the Nation's Children and Families, Administration for Children, Youth and Families, Department of Health and Human Services, undated, p. 11.)

Modest Success
Head Start's lack of quantifiable success for long-term effects does not mean that it has not done any good; some poor children get meals and health care that they might not otherwise receive; the program has reduced the number of poor children misdiagnosed as retarded, and some studies indicate improvement in school attendance, academic performance and IQ test results for the first year or two. There is, moreover, anecdotal evidence of individual successes. Some of Head Start's operating premises -- parental involvement, self-help emphasis, local control -- are an improvement over those of typically top-heavy bureaucratic welfare agencies.

Good Intentions
While Head Start rightly claims modest success in aiding some disadvantaged youngsters, the program mainly thrives on good intentions. The most comprehensive study of Head Start was completed in 1984. Among its most important findings: short-term gains fade within a couple of years. (R. H. McKey, et al. , The Impact of Head Start on Children, Families and Communities, 1985, cited in Ron Haskins, "Beyond Metaphor, The Efficacy of Early Childhood Education," American Psychologist, February 1989, pp. 274-281.) Says Ron Haskins, a developmental psychologist and welfare analyst for the House Committee on Ways and Means: "Results from Head Start projects are uncertain for the special education effects, and there is virtually no evidence [that Head Start reduces] teen pregnancy, crime, welfare, or unemployment." ( Haskins, op. cit., p.279.)

No major studies of Head Start have been released since 1985, and none are underway, according to Esther Kresh, a senior researcher at Head Start. Observes Kresh, who favors more studies into Head Start's effectiveness: "There isn't anything going on now. The whole body of research was in the early days." (Telephone interview, March 8, 1991.) Kresh adds that "the real study has never been done" on Head Start. As a Head Start proponent, she believes that such a study would reveal that the program works, although not to everyone's satisfaction. For one thing, the program keeps changing. Says Kresh: "It's gone from being a child-development program to a full-family development program." (Interview, op. cit.)

No Panacea
A Head Start advisory panel last year recommended studies "to identify early and intermediate outcomes" and factors outside the program that affect children's later chances for success, such as family life and schooling. (Head Start Research and Evaluation: A Blueprint for the Future. Recommendations of the Advisory Panel for the Head Start Evaluation Design Project, Department of Health and Human Services, September 1990, p. 4.) The report affirms agency officials' belief in the success of other child development programs, but cautions that "policymakers and the general public should not be oversold that early education and intervention programs such as Head Start, even when implemented in a high-quality fashion, are some kind of panacea that succeed even in the absence of appropriate ongoing child and family support." ( Op. cit., p. 14.)

Head Start lacks a serious benefit-cost analysis, which is a method of expressing costs and benefits in dollars and comparing them. "Unfortunately, there are few benefit-cost studies of early childhood education," says Ways and Means Committee welfare analyst Haskins. "....The only thorough benefit-cost studies published to date were performed on one project -- the Perry Preschool program." (Haskins, op. cit., p. 279.) And, as Haskins emphasizes, "Perry is not Head Start." (Telephone interview, March 19, 1991.)

To sum: Hard evidence is lacking for claims by Head Start proponents. As such, before Washington spends more taxpayer money on Head Start, the program should be researched thoroughly. Without conclusive evidence of lasting benefits, Head Start should not be expanded, nor should similar programs such as Even Start.

Even Start

A little-known program, Even Start was begun in 1989. Now the U.S. Department of Education is seeking $60 million in the fiscal 1992 federal budget for Even Start, almost triple its fiscal 1991 outlay. Even Start's purpose, as stated in the Federal Register, is to assist eligible local educational agencies in "providing family-centered education projects to help parents become full partners in the education of their children, to assist children in reaching their full potential as learners, and to provide literacy training for their parents." (Federal Register, Vol. 56, No. 23 (February 4, 1991), p. 4390.) Although Head Start also has parent education, Even Start requires it. Even Start is home-based, and enrolls children aged 1 through 7; Head Start is for children ages 3 to 5 and is based in classrooms.

To enroll in Even Start, parents must reside in a poor area and be eligible for adult education assistance programs. Even Start requires parent involvement in all activities and stresses adult literacy, parenting skills, and parent-child interaction. Asked to compare Head Start and Even Start, Tisch Rennings, Even Start education program specialist at the Department of Education, told a Heritage Foundation researcher: "There really isn't any difference. It's essentially funding for the [early childhood] programs from a different source." (Telephone interview, February 26, 1991.) There are no studies to date on Even Start.

The Early Education Lobby

Taxpayers are funding a campaign to increase federal support of early childhood development programs. One of these efforts is "Excellence in Early Childhood Education: Defining Characteristics and Next-Decade Strategies," a pamphlet published by the Department of Education's Office of Educational Research and Improvement. It is written by Sharon L. Kagan, associate director of the Bush Center in Child Development and Social Policy at Yale University. She is active in child care lobbying organizations like the Child Care Action Network, the Children's Defense Fund, and the National Association for the Education of Young Children.

"Excellence in Early Childhood Education" calls for more federal oversight, national standards, and more federal subsidies to child care professionals. The 29-page essay, part of a four-part series of "Public Policy Perspectives," calls for, among other things, a broad campaign to persuade Americans that "early care and education programs are now a permanent part of the social landscape." (Sharon Kagan, "Excellence in Early Childhood Education: Defining Characteristics and Next-Decade Strategies," Office of Educational Research and Improvement, U.S. Department of Education, July 1990, p. 21.) By publishing this essay, the Department of Education is lobbying for expansion of programs for which there is little proof of effectiveness.

Kagan consistently argues for a larger federal role in early childhood development. To do this, she wants "a single agency at the Federal level." (Kagan, op. cit., p. 4.) Her essay includes no dissent from the view that early childhood programs are effective, essential, and should be brought under a federal umbrella. Parental rights receive only lip service. The pamphlets, which cost $4,080 to print, plus a $2,500 fee for Kagan, (Telephone interview March 18, 1991, with Lewis Walker, chief of the Educational Information branch of the Office of Educational Research and Improvement, Department of Education.) were distributed to about 4,000 educators and policy makers across the country. Although the pamphlet contains the caveat that it "does not necessarily represent positions or policies of the U.S. Department of Education, and no official endorsement should be inferred," the Department's blessing is clearly implied. Indeed, the Department has issued no companion publication with opposing views.

National Test For 3-Year-Olds

A federal panel is exploring ways to test children as young as age 3 to determine "the skills and attitudes of the school-entering population." ("Congressional Panel Examines Testing Three-Year-Olds," Education Daily, December 11, 1990, p. 2.) Yet kindergarten begins for most children at age 5 or 6. Testing children at age 3 serves no purpose other than to provide statistics that can be used to marshall support for schooling at earlier ages.

The Special Study Panel on Education Indicators was mandated in the Hawkins-Stafford Elementary-Secondary School Improvement Act in 1988. Last December in a preliminary draft of a report to be released this June, the panel recommended national testing of 3-year-olds. Says panel chairman Alan Morgan, who is New Mexico's state school superintendent: "We won't break the cycle of poverty unless we intervene early." (Ibid.) He maintains that testing 3-year-olds "will allow us to determine the sources of education inequities near their roots, rather than after children have already spent time in elementary school." (Ibid.) The test, whose format is still under study but which one panelist calls "a Pampers SAT," would be expensive, requiring one-on-one sessions with "an adult trained in early-childhood development." (Ibid.)

Such a testing program could be the first step in the formation of compulsory schooling for very young children.

Why Early Childhood Education?

"Early childhood education" generally refers to developmental programs designed for 3- to 5-year-olds. Some experts prefer the term "early childhood development" because they fear children might be harmed by undue academic pressures.

Programs vary widely, from custodial care, which is basically baby-sitting, to highly structured curriculums. Children spend as little as a couple of hours a week at a care facility or as much as ten hours a day, five days a week.

The problem is not the concept of preschools. Even families with one full-time parent often enroll young children in nursery schools two or three mornings a week for social activities and to give tired parents a short break.

The problem is that federally funded programs are growing rapidly despite scant evidence that they work. Says developmental psychologist Peery: "Nobody is doing any critical thinking. It's a mantra now." (Telephone interview, March 7, 1991.)

In the mid-1970s, several model education programs were organized into a consortium study, whose findings were released in 1982. (Irving Lazar, et al., Lasting Effects of Early Education: A Report from the Consortium for Longitudinal Studies, Monographs of the Society for Research in Child Development, 47, 2-3 Serial No. 195, 1982, cited in Haskins, op. cit., p. 274-282.) With the exception of a study of the Perry Preschool Program, the analyses, taken as a whole, reported substantial initial gains in children's cognitive performances, followed by declines within a year or two, with inconclusive evidence about long-term effects. Perry Preschool, which early education proponents cite frequently as proof that such programs work, is the only program for which studies report dramatic long-term results. Neither the Perry Preschool study, however, nor the programs in the consortium were representative of most preschools. Notes welfare analyst Haskins: "These intervention programs were conducted under ideal circumstances: Skilled researchers, capable staffs, ample budgets... it seems unwise to claim that the benefits produced by such exemplary programs would necessarily be produced by ordinary preschool programs conducted in communities across the United States." (Haskins, op. cit., p. 277.) Head Start, on the other hand, he adds, "can be offered as a close approximation of what could be expected from universal preschool education for poor children." (Ibid.)

In recent years, interest in early childhood education has increased for several reasons:

1) America's failing educational system. Test scores are still falling, dropout rates are high, especially among the poor, and literacy skills are declining, despite the highest level of public spending on education in American history. Some people fear that America's children will grow up unable to compete in a global economy.

2) A plethora of seemingly intractable social problems, particularly in America's inner cities. Early intervention is seen as a way to equip children with skills that will allow them to succeed in school and later in life.

3) The entry of many women into the workforce since the 1960s, which has created a demand for non-maternal child care. Just as it would be unreasonable to fashion a federal policy to induce all mothers to stay at home, it is just as unreasonable to induce all children into day care, even that which emphasizes early childhood education. As Heritage Foundation family policy analyst Rector writes in the Harvard Journal on Legislation: "We must ask whether government policy should actively discourage the home rearing of children." (Rector, op. cit., p. 541.)

Unexamined Assumptions

To justify the drive for more federal spending on programs, early childhood advocates make several claims, among them:

CLAIM #1:

Traditional child-rearing, that is, family life, is "obsolete" or a mere idealization. "Less than 10 percent of families" now fit the Ozzie and Harriet model, says Representative Patricia Schroeder, the Colorado Democrat who has just assumed chairmanship of the House Select Committee on Children, Youth and Families. (Cited in David Blankenhorn, "Ozzie and Harriet: Have Reports of Their Death Been Greatly Exaggerated?" Institute for American Values, Family Affairs, Vol. 2, No. 2-3 (Summer/Fall, 1989), p. 10.) Another typical comment is found in a report funded by a grant from the Department of Education: "[Norman] Rockwell's Portrait of American Family Is Changing: Only 4-7 percent of American families have an employed father, full-time homemaker mother, and two children." (National Commission on Working Women, cited in Resegregation of Public Schools: The Third Generation, Network of Regional Desegregation Assistance Centers, June 1989, p. 40. )

REALITY:

The citations above are very misleading. For one thing, only families with two children are counted. If a family has one child, or three or more, it is not counted as a "traditional" family, even if the family has an employed father and a full-time homemaker mother. Not counted also are families in which the mother is employed part- time, even if only during the summer, or is employed full-time for only part of the year, or families with unemployed fathers. For another thing, the term "families" is misused; the data really count "households." This includes college students, empty nesters with grown children, childless single adults, and others who are in different categories entirely.

The latest data from the U.S. Labor Department and the Census Bureau clearly show that two-thirds of the nation's young children are being raised at home, most by full-time mothers or mothers who are employed only part-time. (Marital and Family Characteristics of the Labor Force from the March 1990 Current Population Survey, U.S. Department of Labor Bureau of Labor Statistics, October 1990, p. 4.) Many women leave the labor force after the birth of a second child. Some resume part-time or full-time jobs outside the home when their children are older. More than one-third of mothers with children aged 6 to 13 are outside the labor force; half of mothers with children aged 6 to 17 either are outside the labor force or are employed part- time. (Ibid.) Of mothers employed full-time, most juggle child care responsibilities with their husbands, other family members, or neighbors. (Ibid.)

CLAIM #2:

Government programs for young children prevent social ills and boost academic scores in later life.

REALITY:

The only longitudinal study to claim dramatic, long-term results is of the Perry Preschool Program. Hundreds of other studies have failed to duplicate the Perry findings.

CLAIM #3:

A "crisis" in child care exists, so the federal government should spend more on early childhood programs.

REALITY:

Care facilities and programs for young children are expanding rapidly. Between 1960 and 1986, the capacity of formal group care centers in America increased from 141,000 to 2.1 million, while the number of centers grew from 4,400 to 39,929 in the same period. Many churches now operate full-day or part-day care centers. There even may be an oversupply of day care. National day care chains such as Kinder- Care Learning Centers, Inc. report average vacancy rates of 25 percent. (Rector, op. cit., p. 524.) The bulk of child care is being done by parents, relatives, neighbors or by the more than 1.6 million small, unlicensed day care providers. A 1988 Labor Department survey finds "no evidence in support of the contention that there is a general, national shortage of available care." (Child Care: A Workforce Issue 10, Secretary's Task Force, U.S. Department of Labor, 1988, cited by Rector in ibid.)

The Perry Preschool Experiment

The most-often cited evidence that early childhood education works are studies of the Perry Preschool Program, a project involving 123 black youths from poor families in Ypsilanti, Michigan. Begun in 1962, the study explores long-term effects of this high-quality early childhood education program on 58 children. Another 65 children were used as a control group; they did not attend preschool. The researchers began tracking the 123 youngsters when they were three and four years old.

In 1984, when the subjects were between ages 19 and 24, studies indicated long-term gains for the Perry pupils. According to the High/Scope Educational Research Foundation, which both designed the preschool curriculum and conducted the study, the Perry pupils had half the rate of teen-age pregnancy, a much lower rate of arrests and juvenile delinquency and were half as likely to depend on welfare assistance as the pupils in the control group. (John R. Berrueta-Clement, et al., Changed Lives. The Effects of the Perry Preschool Program on Youths Through Age 19 (Ypsilanti, Michigan: High/Scope Educational Research Foundation, 1984).)

Predicted Benefits
A Perry benefit-cost analysis reviewed such factors as savings in public expenditures on education, welfare and crime, and added taxes paid by later-employed Perry participants. (W.S. Barnett, The Perry Preschool Program and its Long-Term Effects: A Benefit-Cost Analysis, High/Scope Early Childhood Policy Papers, No. 2., Ypsilanti, Michigan, 1985. C.U. Weber, P.W. Foster and D.P. Weikart, An Economic Analysis of the Perry Preschool Project (Ypsilanti,Michigan:High/Scope, 1978). Both studies cited in Haskins, op. cit., p 279.) Researchers concluded that "for each dollar invested in the Perry Preschool Project, taxpayers receive a return of $1.46 by the time participants are about 20 years of age. Predicted net benefits over the lifetime of participants are even more substantial." High/Scope researchers estimate that every dollar spent on preschool education for children at risk saves $5.73 in projected future costs to taxpayers. (Ibid.)

The Perry findings have been accepted widely as proof that preschool programs, including Head Start, are good investments for taxpayers and good for America's children. Calling the Perry program "impressive," the House Ways and Means Committee's Haskins says well- designed programs probably benefit some children, at least in the short run. But he cautions: "It is unreasonable to expect that extending preschool programs of the Head Start type to more low-income children would produce benefits of the magnitude found in the Perry project." (Haskins, op. cit., p. 280.) The problem with the Perry study is that the Perry Preschool Program was far from typical. The lessons learned from it thus have very limited applicability.

Examples:

Sample size
The Perry Preschool Program used a microscopic segment of a narrowly defined population: poor, black children with IQs between 61 and 81 in households in which a parent was home during the day. (Zigler, op. cit., p. 30.) Yale psychologist Zigler warns that Perry "was not only unrepresentative of children in general; there is some doubt that it was representative of even the bulk of economically disadvantaged children.... it is even problematic as to whether the sample was representative of low-income black children." Other backers of early childhood education also say Perry's findings need perspective. Notes Head Start's Kresh: "When they talk about a difference in something happening, they're talking about 10 percent. In a sample that small, they're talking about two or three kids." (Telephone interview, March 8, 1991.)

Program Idiosyncrasies
Perry preschoolers were in a highly controlled academic environment and were instructed by highly professional educators. Perry pupils' parents were heavily involved, and instructors visited pupils' homes each week. Few programs duplicate such conditions, and even those that are similar have not reproduced Perry's successes.

Conflicts of Interest
Studies of the Perry Program were conducted by the same people who designed and administered the project. High/Scope researchers have a professional and economic stake in the program's success, including the marketing of early childhood education program material. Such conflicts of interest prevent High/Scope researchers, even with pure intentions, from being objective observers. In High Scope Resource magazine, High/Scope founder David Weikart tells educators: "Well folks, hold on to your hats, because your salaries are going up!... Next year, Head Start should receive approximately $700 to $800 million more in appropriations .... There are signs of increasing financial support for the early childhood field, including federal day care spending bills that provide 'wrap-around services' for Head Start, allowing Head Start to provide child care beyond its current program hours." (David P. Weikart, High Scope Resource, Fall 1990, p. 22.) Weikart calls for more lobbying: "We must be out there as advocates in the public schools across the states when they are passing their appropriations bills." (Ibid.)

Conclusion

In the absence of hard scientific data, the federal government should be very cautious before launching any programs. This is particularly true, of course, for Washington's many-faceted campaign to expand taxpayer-funded early childhood programs.

Warns psychologist Peery: "The public standard ought to be that of prohibiting government from doing anything to children unless it can be convincingly demonstrated that the children will benefit."

The answer to American cultural problems lies not in more government programs to make up for family dysfunction but policies that support families, particularly those with parents who care for their own children.

To ensure that federal efforts help America's children:

** The Executive Order on the Family should be vigorously enforced. Signed by Ronald Reagan in 1987, the order requires federal agencies to take into account their policies' impact on family formation, maintenance and general well-being" and to question whether policies "strengthen or erode the stability of the family and the marriage commitment" and "erode the authority and rights of parents in the education, nurture and supervision of their children." This Executive Order is widely ignored. The White House Office of Policy Development submits an annual report on agencies' compliance with the Executive Order on the Family and recommendations to the President. It is time for George Bush to enforce them.

** More research should be done on early education for disadvantaged children to determine if there is a model that works and can be replicated. Such efforts should target severely disadvantaged children, follow methods for which there is evidence of success, and use a controlled experiment format.

** Existing childhood education programs should be evaluated by independent researchers. Bureaucracies cannot be expected to analyze fairly the programs that they administer. Studies -- and federal policies -- should take into account the impact of and impact on families.

** Federal policies should be based on giving parents more choices. This could be achieved through vouchers and tax credits for child care and education. If schools fail to teach, it is not because they have not enrolled children soon enough.

** Families should be strengthened, not supplanted by social programs. The federal tax code should not punish two-parent families with children. The personal exemption for dependents should be increased substantially. The $600 exemption of 1948 would be worth more than $6,000 today. Welfare programs should be redesigned to encourage fatherhood, not father abandonment. Taxpayers should not be subsidizing family break-up, and government should intervene only in cases of extreme hardship or abuse.

Families are America's greatest cultural resource. Federal policies, however well-intentioned, that hinder families or supplant parental authority lead to more social problems, not fewer. Parents who sacrifice by cutting work hours, by temporarily leaving the work force, by rearranging schedules or downscaling their life styles so that one parent can care for young children at home suffer discrimination two ways: they are taxed to subsidize services they do not want or use, and they are taxed to support campaigns designed to justify those services.

Early childhood development programs must be better scrutinized for their effects on children and families before the federal government spends more taxpayers' money on them. Says former Education Secretary William J. Bennett: "Government, obviously, cannot fill a child's emotional needs. Nor can it fill his spiritual or moral needs. Government is not a father or a mother." (William J. Bennett, "Children and Culture in Modern America," remarks at University of Notre Dame, South Bend, Indiana, October 17, 1990.)

Empowering Parents
Children need their parents, and parents need more choice in how they spend their time and money. The best way to empower children is to empower their parents -- the people who care most and know what is best for their children.

Robert H. Knight, former Senior Fellow, Cultural Policy Studies

This is the fourth in a series of studies analyzing the impact of federal policies on American culture and cultural values.

Heritage Foundation Research Assistant John M. Slye contributed to this study.