Sacramento Head Start Alumni Association

A Cost-Benefit Approach to Head Start

Executive Summary

The short and long term effects of providing at-risk preschool children with quality, comprehensive, early childhood experiences are widely documented in the research literature. When children born in poverty are provided with high-quality, active learning preschool experiences, the benefits are significant, both in terms of improved quality of life for these children when they become adults and in terms of lower costs to society for welfare, grade retention, special education, social services, and crime costs. A sampling of these findings shows:

One third more at-risk children who attended a quality early childhood programs graduated from high school than those who did not attend (Schweinhart, et al, 1993; Lazar, et al, 1982; Lally, et al, 1988). This has meaningful significance to those concerned with the nation’s economic health since unemployment rates for high school graduates are 75 percent lower than those for non-graduates and graduates earn an average of 25 percent more than non-graduates (U.S. Bureau of the Census, 1992).
Children at-risk who have benefited from a quality early childhood program spend 1.3 years less in some form of special education placement (Schweinhart, et al, 1993). The average costs of special education per year (in addition to standard educational costs) are approximately $1,800 per student (Department of Education, 1993).
Children at-risk who have been enrolled in a quality early childhood program are 25 percent less likely to be retained in grade (Schweinhart, et al, 1993; Lazar, et al, 1982; Lally, et al, 1988). Since the current cost of public education is $5,200 per student, per year, this can save taxpayers a significant amount of money (Department of Education, 1993).
As adults, those attending a quality early childhood program are three times as likely to be home owners by age 27, compared to those who did not receive the program (Schweinhart, et al, 1993). As we all know, home ownership is an indicator of successful adaptation to society.
Young women who have experienced a quality early childhood program are one-third less likely to have out-of-wedlock births and 25 percent less likely to be teen mothers (Schweinhart, et al, 1993; Lazar, et al, 1982; Lally, et al, 1988). The cost to society for teenage childbearing is extremely high, with $16.6 billion of federal support alone in 1985 (Haskins, 1989).
At-risk children not afforded the opportunity to participate in a quality early childhood program are five times more likely to be arrested repeatedly by age 27 (Schweinhart, et al, 1993; Lazar, et al, 1982; Lally, et al, 1988). Furthermore, there is evidence that the nature of the crimes committed by program children are of a less serious nature (Lally et al, 1988). We all know the tremendous cost to society of crime and its consequences. It has been estimated that the annual cost to victims of crime was $92.6 billion in 1985 (Cohen, 1988). Even if an expanded Head Start could cut this by 1/20th, it will have paid for itself.


Current critics of continued expansion and funding for Head Start are quick to point out that each of these findings come from non-Head Start intervention projects and argue that the results are not supported by Head Start research. We concur that there have been few research findings of this nature that are Head Start specific, but wish to point out that there have also been no research findings to the contrary either. In fact there has been little, or no, research on the long-term outcomes for Head Start that has been discussed in the general early intervention research literature. However, the research findings on the immediate effects of the program are virtually identical to those of the model programs that have produced these findings. Is it not reasonable to believe that if the short-term effects are the same, that the long-term ones would be too?

The purpose of this paper is to outline the National Head Start Association’s position regarding criticisms related to cost issues for the Head Start program. These criticisms have been leveled by those who would drastically cut (or eliminate) funding for Head Start specifically, and children’s programs in general. It is the contention of the National Head Start Association that efforts to improve quality and expand services in Head Start begun during the Bush administration should continue because:

Head Start is cost effective.
Head Start is competitively priced.
Head Start is an investment in America.
Head Start is the RIGHT thing to do.
Furthermore, it is important for Congress and the electorate to fully understand the facts behind these criticisms before acting upon the recommendations of those with agendas other than merely balancing the federal budget. Those who would accept, without question, the conclusions of the critics are being blindly led down a euphemistic path, returning us to shamefully persecuting poor children and families, or at best ignoring them. While there are heated debates about amending the Constitution, who is not in favor of having a balanced budget? But, does this mean that we should abandon all hope for our most vulnerable children in order to save the children of the elite from inheriting our debt? We agree that tough decisions will need to be made by this Congress as well as those to follow. However, we believe that the future of our children is, at least, as important as the future of our military readiness. If children are not ready for school at age 5, then young adults at age 18 will not have the chance to attain the skills needed to become self-sufficient contributors to the economy and well-being of this nation.

Background

Head Start is one of the most constructive, and one of the most sensible, and also one of the most exciting programs that this nation has ever undertaken.
— President Lyndon Johnson, May 1965

Born during the "War on Poverty," Head Start best exemplified the nation’s philosophy regarding any war –– wage war on the institution, not the people. The program began with a simple objective, based on the idea that children deserve the opportunity to enter school ready to learn. This readiness was defined as: 1) having been exposed to appropriate educational experiences, 2) being healthy physically and mentally, and 3) having families that have been provided with the basic social supports needed to be actively involved in their children’s lives and development.

The questions regarding the continued expansion and, in fact, existence of Head Start as we know it, are appropriate ones. Does Head Start work? Is Head Start cost effective? The National Head Start Association, not surprisingly believes that the answer to both questions is a resounding YES. But we are not alone in that conclusion, and those who agree with us represent a broad spectrum of American politics. Bipartisan support for the program for 30 years has been its major strength in its struggle for survival. The major reasons for this support have both been economic ones, its cost-effectiveness and its value as an investment.

As the program developed during the late ‘60s and early ‘70s, the original objective of preparing children to enter school became more defined, with performance objectives and standards being developed and implemented. Issued first in 1975, after years of work dedicated to ensuring that the standards would not hamper local control of the program, the Head Start Performance Standards established specific objectives designed to achieve the goal of preparing children to enter school. In keeping with concepts of Management by Objectives, it was assumed that achieving the objectives would lead to the achievement of the goals and, therefore, serving as an investment in the nation’s future. The current Head Start objectives (DHHS, 1984) maintain this notion and set reasonable expectations for meeting the human resource needs of the United States in the years to come. They include:

The improvement of the child’s health and physical abilities, including appropriate steps to correct present physical and mental problems and to enhance every child’s access to an adequate diet. The improvement of the family’s attitude toward future health care and physical abilities.
The encouragement of self-confidence, spontaneity, curiosity, and self-discipline which will assist in the development of the child’s social and emotional health.
The enhancement of the child’s mental processes and skills with particular attention to conceptual and communications skills.
The establishment of patterns and expectations of success for the child, which will create a climate of confidence for present and future learning efforts and overall development.
An increase in the ability of the child and the family to relate to each other and to others.
The enhancement of the sense of dignity and self-worth within the child and his or her family.
It is, therefore, within the context of these goals and objectives that Head Start must be placed when trying to determine whether the program is cost effective or a worthy investment of taxpayer dollars. By the early 1980s Head Start was firmly, albeit cautiously, linked with Ronald Reagan’s "safety net." During this decade the program enjoyed popularity, as well as bipartisan support. Under President Bush this support translated to the largest expansion effort in the history of the program. Through this bipartisan effort, Head Start was recognized as the primary tool to ensure that, "by the year 2000, all children in America will start school ready to learn." Both candidates in 1992, George Bush and Bill Clinton, pledged to extend Head Start services to all eligible children. As Zigler and Muenchow (1992) put it, ". . . Republicans and Democrats were almost in the posture of competing to see who could provide the greatest boost for Head Start. Both party platforms that year called for extending services to all eligible children."

So what happened? How can a program go from being the darling of both parties in 1992, to being considered for drastic cutbacks in 1995? It would be easy to attribute it to the current national obsession with the destruction of heroes and history. No sooner does the nation get a new hero, than the popular press begins to dig up enough dirt to tarnish the shiny new image.

The Head Start program began as a summer program in 1965 with a total budget of $96,400,000 serving 561,000 children. This translated to $172 per child for an eight-week program. Today the total budget is $3.5 billion, serving around 750,000 children for an average of eight months. To paraphrase Everett Dirksen, a million here and a million there, we now have real money. The only problem with having real money is that someone always wants to take it from you. It is only right that the taxpaying public should expect to get something for its dollars, and it is our position that dollar for dollar, Head Start is one of the best investments this country has ever made.

It is the intent of this paper to provide some hard data, where available, on both the cost of Head Start and its value as an investment in the future of America’s children.

The Cost of Head Start

As the argument goes, "Head Start is too costly when compared to other child care programs." First, and foremost, Head Start is not child care. Head Start has been throughout its history, and remains today, the only truly comprehensive child and family development program with a national scope. To compare Head Start to child care is like comparing the cost and effectiveness of the U.S. Army Special Forces to the National Guard. Both are necessary, but hardly serve the same purpose.

The current cost for child care has been estimated to be $70 per week for an 8-hour day. If translated to Head Start’s average 4-hour day this would be $35 per week, or $1,400 per 40-week year for child care alone. While Head Start currently costs $3,500 per 40-week year, only 50 percent, or $1,750 per year of that cost can be attributed to the "child care" factor (DHHS, 1993). This $350 difference is easily accounted for by the increased standards of Head Start and the educational content of the program which is not present in custodial care.

Head Start is a comprehensive child development program, it is not child care. Therefore it is unfair to compare costs or effects with child care. Head Start addresses child and family needs: health, nutrition, dental, mental health, and social services, it is not a preschool program which deals primarily with early childhood education. Head Start puts a premium on parent involvement: parenting skills, literacy skills, and employment skills; and Head Start always encourages parents to participate in important program decisions.

Head Start is a child and family developmental approach, not an educational approach for children only. Even other preschool programs are not comparable in content to Head Start. School system pre-K programs offer developmentally appropriate educational programs, but fail to provide the health, mental health, social services, and parental involvement components at levels equivalent to Head Start. Even highly sought after private preschools fall short on these areas. It is interesting to note that many upper-middle class families develop anxiety disorders as they attempt to secure limited slots in preschools that can cost as much as $12,000 per year. Yet, those who claim to represent these families will argue that a $3,760 per year program for poor children is wasteful. As the following chart indicates, Head Start costs are not only competitive but, in some cases, a bargain.



Is the current $3,760 per child money well spent? Is the American taxpayer getting his/her bang for the buck? Will continued investment and expansion of Head Start pay off in noticeable results for low-income children? Each of these questions are valid ones that some will say have been answered. Depending on one’s political persuasion the answer is either yes or no. It is our position that the questions have yet to be properly asked, but that we expect the answer to be yes to all three. Furthermore, we are not alone in our assessment. The bipartisan Advisory Committee on Head Start Quality and Expansion would seem to agree with us in their recommendations to continue with expansion and focus on quality.

To start to answer these questions with the available data, let us first consider that in 1994 the average teacher salary in Head Start was $15,499 as compared to the 1992 average elementary teacher’s salary of $34,334 and the 1992 average secondary teacher’s salary of $35,750. Even the average for beginning teachers in 1992 was $23,054. This is not surpassing when we notice that the average cost per pupil in the public schools was $5,010 in 1992. If a child requires special education, an additional $1,500 minimum can be tacked on to the cost.

The cost for Head Start in 1992 was $3,415 per child and has now reached an average of $3,760 for 1994, with the increase in public school costs being similar. Continuing this comparison we have already noted that the cost per year for the Perry Preschool was $7,800 in today’s dollars. One proposal that has been floated recently has been to place poor children in state-run orphanages, but the average cost for a year in an orphanage is $15,000. Finally, if we do nothing, we should expect to pay as much as $23,000 per year to house some of these children later in jail. As the chart below indicates Head Start is a deal by comparison.



The Head Start Investment

One of the leading proponents of the argument that Head Start may not have the same impact as the model program (i.e., Perry Preschool) has been Ron Haskins. In his 1989 paper he correctly points out the weaknesses in the available data, but concludes that there ". . . is ample reason to continue its funding and to focus serious attention and additional resources on improving its performance" (page 281). Additionally, he points out the most likely reasons why the data available does not show the same levels of effects for Head Start as those found in the model programs:

Available studies have pooled data from Head Start programs without regard to quality. By definition, a program not meeting the minimum performance standards is not Head Start. Any effort to determine the effects of "Head Start" should be limited to programs meeting the standards. It is not unreasonable to expect that the effect sizes for these programs would be closer to those of the "model" programs.
At the time of most studies on the effects of Head Start most programs were only funded to serve 20 percent or fewer of the eligible children and regulations required programs to offer enrollment opportunities to those eligible children at greater risk first. All other eligible children were placed on waiting lists. It has been from this waiting list that the "controls" for most Head Start studies have been selected. Clearly, the comparison group is most likely better off to begin with than those in the program, thus reducing the measured effect of the program.
Among Head Start parents and staff the efficacy of Head Start is overwhelmingly believed to be a fact. This, however, is often not enough when dealing with public policy. While we will continue to rely heavily upon case stories, this paper presents some reflections on the empirical evidence of the program’s efficacy. It is accepted that currently there is sparse data on this subject which can be generalized and that the evidence that is available lacks sufficient strength to be convincing. However, it is the consensus of many that these problems stem from the methodology rather than weaknesses of the program.

The issue of efficiency cuts to the heart of what the Head Start program (or any attempt to provide intervention in the lives of at-risk children) is all about. Ben Franklin’s "Poor Richard’s Almanac" is full of references to this issue; a stitch in time, a penny saved, an ounce of prevention, and so forth. Now, more than 200 years later, we are struggling to understand this concept in relation to children and families at-risk. The primary thesis of this paper is that Head Start is not only effective, but that it is efficient in its "ability to produce a desired effect, product, etc. with minimum of effort, expense, or waste." The return on investment is the primary measure of this effectiveness and, we argue, this ranges from $2.50 on the dollar to $10 per dollar. Even at the low end this is a bargain when one keeps in mind we are talking about the children who will make up the future of this country. Even the most frugal political conservative cannot deny that children deserve our love and support. To paraphrase a popular country song, "when it comes to love, we don’t count the cost."

The issue of comparing Head Start results to those of the highly touted Perry Preschool has been depicted by Haskins and others as a metaphor, and thus dismissed as inadequate evidence of the efficacy of Head Start. Those using this line of attack often preface the discussion with statements regarding the similarity of short-term outcomes for children between Head Start and the Perry Preschool. They go on to argue that although there is evidence of the long-term effects for the Perry Preschool, there is little or no evidence for Head Start. However, the case is that there has been little or no attempt to find evidence for Head Start, while this was the primary purpose of the Perry Preschool. Simply because something has not been sought, it does not follow that it does not exist.

Studies have shown that a quality early childhood experience returns anywhere from $7 to $5 for every $1 invested (Schweinhart, et al, 1993). Let’s take a closer look at those numbers and then discuss the metaphor alluded to by Haskins (1989). To stay with what seems to be today’s mood of conservatism, let’s assume the $5 figure as a return on the investment for the Perry Preschool. Let us further assume that critics of Head Start’s comparison to these figures are correct and that the effect size (therefore return) is roughly half that of the Perry Preschool. Even at that we (the American taxpayer) are getting a $2.50 return on every $1 invested in Head Start. Not bad in an era where 5 percent returns are the norm. The chart below illustrates the total saving by category that society might expect from an investment in Head Start. The chart shows a side-by-side comparison of the Perry Preschool (Schweinhart, et al, 1993) and expected outcome for Head Start based on a 50 percent reduction in effect sizes.



It is important to remember that the Perry Preschool cost twice as much per year per child as Head Start and was a two-year program as opposed to a one-year program. Therefore, we should expect that the effect size should be four times that of Head Start, rather than the two times that current research shows. An appropriate research question for the future of Head Start might very well be to find out what causes a stronger effect than one would expect based on cost and length of service alone. Many would argue that the comprehensive nature of Head Start and its emphasis on parent involvement provide the answer.

However, this conservative estimate is most likely unnecessary if we really take an honest look at the metaphor of the research-oriented Perry Preschool and the real life Head Start, or any other quality early intervention. Unlike the Perry Preschool, there is no singular Head Start. Studies that have attempted to address the measurement of effect sizes with Head Start have failed to determine if the programs they studied were, indeed, Head Start. In fact, the National Head Start Association has long pointed out that quality is uneven across the Head Start community (NHSA, 1990; DHHS, 1994) and that to assume that one can average the outcomes from programs to come up with a comparable effect size is even more ludicrous than the assumption that a single year of the program at age 4 is going to inoculate a child against the crime ridden neighborhoods and poor schools that he or she will go on to attend. The major error in these assumptions was even addressed by Haskins (1989) when he rightfully pointed out that due to the inability to serve all eligible children most programs are only serving the poorest of the poor. The only studies of Head Start have used the "waiting list" children as the "controls." If Haskins is right, and we believe that he is, then the controls were already better off than the treatment group, thus the effect sizes for Head Start are likely to be even greater than they are. It has already been pointed out that the cost and duration of the Perry Preschool essentially meant that the investment was quadruple that of the average Head Start program. So, from that standpoint we can agree with Haskins and others that it is unfair to pose the metaphor and assume a direct relationship between the two programs. Our point is that, even with the difference in dosage, there are significant results for Head Start, and that there is every reason to believe that as the investment in Head Start approaches that of the Perry Preschool, the returns too will grow.

Based on the assumption that quality is somehow related to investment, it would be reasonable to expect that continued growth in the per-child investment for Head Start will eventually bring the return rate for Head Start in line with that of the Perry Preschool project. This should indeed be one of the goals of this Congress rather than abandoning all efforts to address the need to ensure that our most vulnerable children have the opportunity to develop into competent and autonomous young adults who, to paraphrase Garmezy (1976), ". . . work well, play well, love well, and expect well." We cannot expect that each child receiving Head Start will go on to be a brain surgeon, lawyer, or national leader, but we can expect that they have the opportunity to pursue their goals.

Another problem that becomes clear when one studies the metaphor in question relates to the relative risk of the children and families served by programs. It is often very easy to have a major impact on a child or family if they are at relatively little risk. Being poor (and therefore eligible for Head Start) does not necessarily imply that a child or family is destined for failure. That is why, as Haskins reminds, that Head Start programs try to serve those children and families with multiple risk factors first. This means of course that children and families who are most likely to benefit from the program with minimal investment are less likely to be enrolled in the program. It is our contention that if appropriations ever reach the point of allowing full enrollment, then we would likely see the success rate for Head Start increase geometrically since the children finally being allowed to enroll will show more instant "success" by mainstream standards.

Conclusion

I came to live in my town after an 11 year abusive marriage, with no clothing, no housing, no money. We just packed up and left. It was my daughter’s turn to go to school . . . I got involved in every aspect of the program . . . Head Start has not only given my child a Head Start in life, but one for me as well."
Head Start Parent, Massachusetts, 1994 - Comments to The Advisory Committee on Head Start Quality and Expansion

These are the only children we have, they are our future and their future is enhanced by our investments today. During the 30 years the Head Start program has been in existence we have learned much about child and family development. We have also learned much about the resiliency of the human spirit in the face of tremendous odds. Much of this we have learned because of Head Start and the chances those social pioneers took some 30 years ago. Zigler once called Head Start the nation’s early childhood laboratory. If we are to achieve success as a nation, it is important that our nation’s children achieve success as individuals. The Head Start program, with all of its local nuances, it still the best bet for more than two-million at-risk preschoolers.

The National Head Start Association applauds any effort on the part of Congress that is likely to improve Head Start's efficiency. In a spirit of cooperation, the National Head Start Association would like to make the following recommendations regarding program issues that will further the goals of the American people:

Recommendation 1:

Fulfill the pledge first made by President Bush, and confirmed by President Clinton, to achieve full enrollment for Head Start. This will not be cheap, but nothing worthwhile is. Let's meet the challenge of Goals 2000 –– By the year 2000, all children in America will start school ready to learn. The primary objectives of Goal 1 are:

All disadvantaged and disabled children will have access to high quality and developmentally appropriate preschool programs that help prepare children for school.
Every parent in America will be a child’s first teacher and devote time each day to helping his or her preschool child learn; parents will have access to the training and support they need.
Children will receive the nutrition and health care needed to arrive at school with healthy minds and bodies; and the number of low-birthweight babies will be significantly reduced through enhanced prenatal health systems.
Head Start is the only program today actually working toward all of these objectives. To change or reduce the program now will close the door of opportunity on millions of children yet to step through a Head Start classroom door. Rather than looking to send Head Start to the states, we should examine the possibility of folding into Head Start those early childhood programs currently administered by the states. The Ohio model has done this and it is working.

Recommendation 2:

Quality issues need to be addressed in order to improve our investment. Just as with any portfolio, the American taxpayer has the right to expect that its investment managers (Congress) are doing everything within their power to increase the quality of their investment. Using the comparison figures for the Perry Preschool we can conclude that as we reach closer to the "quality" level of that project, Head Start will achieve similar effect sizes (and therefore returns on investment) to the $5 to $7 received there. It is imperative that the quality set asides continue so that we increase our return. But, without increases, there can be no quality set-aside.

Recommendation 3:

Past practice of allocating funding to communities on a percentage of the estimated eligible children may have been adequate in the 1960s but become meaningless if we expect to have any significant impact on poverty as we move toward full enrollment. Anyone familiar with basic psychological research should understand Maslow’s hierarchy of needs and its suggestion that some children (and families) have greater need than others. A study in Europe has suggested that for some families a simple brochure is intervention, while others require a more intensive approach. Medical units in a war zone, and hospitals in general, have long practiced the art of triage by assessing the individual needs of all patients and then prioritizing treatment accordingly. It may be time to reconsider funding formulas to ensure that all communities have adequate funding to assess the needs of all eligible children in that community.

Any argument against the continuation of Head Start as a viable investment for American’s children on the basis that it is not 100 percent effective is ludicrous. No treatment for any malady is 100 percent effective, including a strong military as a prevention of attack by foreign powers. Even if the previous deficiencies of Head Start as compared to the Perry Preschool were valid, this is the best thing we have going for poor children and their families. Until something better and more effective comes along, let’s work together to make it better rather than abandoning hope and children to the ravishes of poverty.

It is apparent that those who oppose Head Start do not know Head Start. Reaction to "research" findings regarding the program never seem to come from those who have been active in actual program research. The news media and congressional opponents are quick to quote "resident scholars" when talking about the program, but seldom the original researchers. A careful review of the University of California’s on-line data base, Melvyl, shows that these "resident scholars" have yet to publish any original research on the program. Each piece written by them is an op-ed piece based on their interpretation of the research of others, while for the most part ignoring any research that does not support their handlers’ opinions. Each is about as believable as a paid witness in a trial.

One of the few attempts to look at the long-term benefits of Head Start did find that children who attended Head start were less likely to fail a grade in school than those who did not attend. Even if this were the only benefit of Head Start, the savings of the program would be $5,010 per successful child. So how do we increase the ratio of successful children if we really want to make a difference? These are the real questions that should be addressed and these are the issues that the bipartisan Advisory Committee on Head Start Quality and Expansion addressed in their 1993 report.

Finally, we are in full agreement that the economic well-being of the nation is best served by an efficient use of federal funds and that wasteful government spending needs to be eliminated. However, it is important that our government becomes "lean and clean" not "lean and mean."

References

Cohen, M.A., (1988), Pain, suffering, and jury awards: A study of the cost of crime to victims. Law and Society Review, 22, 537-555.

DHHS, (1984), The Head Start Performance Standards, Washington: Department of Health and Human Services.

DHHS, (1993), Creating a 21st Century Head Start: Final Report of the Advisory Committee on Head Start Quality and Expansion, Washington: Department of Health and Human Services.

DOE, (1993), The Conditions of Education, Washington: Department of Education, Office of Educational Research and Improvement.

Garmezy, N., (1976), Vulnerable and invulnerable children: Theory, research and intervention. Master lecture on developmental psychology. Washington: American Psychological Association.

Haskins, R., (1989), Beyond metaphor: The efficacy of early childhood education. American Psychologist, 44, 274-282.

Lally, J.R., Mangione, P.L., and Honig, A.S. (1988), The Syracuse University Family Development Research Program: Long-range impact of an early intervention with low-income children and their families. In D.R. Powell (Ed.), Parent education as early childhood intervention: Emerging directions for theory, research, and practice (pp. 79-104). Norwood, NJ: Ablex.

Lazar, I., Darlington, R., Murray, H., Royce, J., and Snipper, A., (1982), Lasting effects of early education: A report from the Consortium for Longitudinal Studies. Monographs of the Society for Research in Child Development (Serial No. 195).

NHSA (1990), Head Start: The Nation’s Pride, A Nation’s Challenge – Recommendations for Head Start in the 1990s, The Report of the Silver Ribbon Panel. Alexandria, VA: National Head Start Association.

Schweinhart, L.J., Barnes, H.V., and Weikart, D.P., (1993), Significant Benefits: The High/Scope Perry Preschool Study Through Age 27, Ypsilanti, MI: High/Scope Press.

U.S. Bureau of the Census. (1992). Money income of households, families, and persons in the United States: 1991, Current Population Reports, Series P-60, No. 180, Washington: U.S. Government Printing Office.

Zigler, E. and Muenchow, S., (1992), Head Start: The Inside Story of America’s Most Successful Educational Experiment. New York: Basic Books.

Email us
http://www.nhsa.org/govaff/costbenefit.htm