PUBLIC POLICY VIEWPOINT
Another Step Back?
Assessing Readiness in Head Start
C. Cybele Raver and Edward F. Zigler
BTJ Editor's note: This Public Policy Viewpoint focuses on issues related to the assessment of young children and the new assessment requirements for children in Head Start programs. At the time of this Web posting (1/13/04), Head Start programs have undergone a significant event: use of a federally created test to assess the language, literacy, and math knowledge of all four-year-old children enrolled in Head Start. The Head Start Child Outcomes Framework, Program Performance Standards, and teacher qualifications requirements have influenced child care and state prekindergarten programs. As the Head Start reauthorization and this new assessment move forward, it is important to consider how testing young children may affect the future of early childhood education. We welcome your views on this topic.
Since its founding in 1965, Head Start's goal has been to help children who live in poverty prepare for school. Over the last three and a half decades, Head Start has maintained a staunch commitment to the provision of genuinely comprehensive services (Zigler & Muenchow 1992). While impressive in its breadth, this wide range of services has made it difficult for researchers to benchmark children's progress in the program. One solution has been to rely on strictly cognitive measures as a means to assess the benefits of Head Start. We criticized this approach in an earlier paper entitled "Three Steps Forward, Two Steps Back" (Raver & Zigler 1991). In that article, we pointed out that sole reliance on children's cognitive outcomes was neither in keeping with the goals of Head Start nor with many definitions of what it means to be ready to succeed in early elementary school (Goal One Technical Planning Group 1991; Administration on Children, Youth and Families 2002).
Recently Head Start has been subjected to major policy changes at the federal administrative and legislative levels. In particular, the Bush Administration instituted a new set of accountability measures that will be used to test Head Start children twice a year on language, literacy, and pre-math skills. This policy is swiftly being put into place with full implementation plans announced in both April and June 2003. The assessment system, under the National Reporting System that is part of the current law but left to the Secretary to determine, has been controversial (see for example, Wagner 2003). The measures were quickly developed by Westat, Inc., and the national assessment process is now underway. This fall, all four- and five-year-old children in Head Start (who are eligible to enroll in kindergarten next year) will undergo the first of two annual assessments. This quick pace of change proceeded despite a letter to administrators signed by some 300 professionals questioning the psychometric properties of the measures (drafted by Meisels et al. 2003).
The spring of 2003 was also the time Congress began work on reauthorizing Head Start's funding. The House version of the reauthorization bill (HR 2210) proposed substantial changes to the 38-year-old program. Most controversial was a plan to devolve Head Start to the states, but the bill also raised the issue of assessment. The bill (as introduced) emphasized children's knowledge and skills in the areas of language, literacy, and pre-math and deleted the current law's references to children's social competence, emotional development, and cultural diversity (Schumacher, Greenberg, & Mezey 2003). Why did the Bush Administration move so sharply away from Head Start's emphasis on school readiness in broadly defined terms, and toward a narrow emphasis on cognitive development as the critical factor in preparing for school?
Perhaps these moves are driven by well-meaning intentions on the part of policy makers to improve the educational chances of our nation's most disadvantaged young children. The evidence shows that while Head Start children make significant gains in preschool, they still score well below the national average on vocabulary, pre-reading, writing, and early math skills (Administration on Children, Youth and Families, U.S. Department of Health and Human Services 2003). Secretary of Health and Human Services Tommy Thompson argues that poor children deserve a better start to their educational trajectories (see U.S. Department of Health and Human Services 2003). Lawmakers on both sides of the political spectrum began to focus the debate on what it means to close the achievement gap of Head Start children with their middle-class peers. However, their pathways to that goal were quite different, as were their expectations of closing that gap merely by adding a stronger focus on literacy and math skills.
On the face of it, there is some logic to the idea that if children are less knowledgeable regarding early academics like letters and numbers, strengthening these skills should help them when they begin school. And perhaps lawmakers were persuaded by a small number of studies that suggest that some programs (but not others) have shown limited short-term improvements in older children's educational achievement when "high-stakes" achievement tests are used to increase school monitoring and accountability (Kane & Staiger 2002). In our view, however, these intentions are misguided. As we will argue, the application of a strictly cognitive focus to assessments of school readiness runs counter to what the best developmental research tells us and what past policy experience has shown. A narrow focus on benchmarking Head Start's programmatic success on early cognitive gains to the exclusion of children's emotional and social development has been tried in the past and has backfired. In this article, we briefly review these past rounds of policy debate, and consider scientific evidence regarding what disadvantaged preschoolers need to be ready for school. We then offer three concrete policy recommendations for alternatives to the steps that are in the works for Head Start accountability.
What does past policy experience tell us?
This is not the first time that policy makers and research scientists have tried to peg evidence of Head Start's success to children's cognitive gains. During Head Start's early years, evaluators commonly found substantial gains in children's IQ scores after even brief periods of intervention. These gains were publicized as striking evidence that the programs worked. (See Zigler & Butterfield 1968; Zigler & Trickett 1978; and Zigler & Styfco 1997, for reviews.) However, when the IQ benefits were found apparently to dissipate as children progressed through elementary school, intervention efforts were quickly deemed a failure not worthy of public support.
When Head Start and other early interventions failed to show permanent gains in children's cognitive scores (as assessed by IQ), policy makers had two choices: either to capitulate to the skeptical view that early intervention is not effective, or to question whether IQ gains were the appropriate metric to have used in the first place and whether the programs were improperly evaluated. Workers in a variety of disciplines eventually convinced policy makers that intelligence alone does not guarantee academic success-that even a very bright child will do poorly in school if he or she suffers physical health or emotional problems, has trouble staying motivated, or does not interact well with teachers or peers. Consequently, researchers, policy professionals, and practitioners in the field of early childhood education seemingly resolved this issue by establishing that Head Start must continue to encompass a broader mission of school readiness that includes physical and mental health, social and emotional needs, and academic skills (Zigler & Muenchow 1992). This emphasis on both cognitive and social-emotional development was validated by specific language in the 1998 Head Start reauthorization act. Further verifying policy makers' acceptance, data on children's social and emotional development (though in limited form) began to be collected in large-scale national surveys including the Family and Child Experiences Survey (FACES, Department of Health and Human Services), the Early Childhood Longitudinal Study (ECLS-K, Department of Education), and the Head Start National Impact Study (U.S. Department of Health and Human Services). Suddenly, however, the current administration decided to reverse course.
In part, we suspect that this reversal is due to consensus that there is an unacceptably large "achievement gap" between economically disadvantaged children and their more advantaged counterparts, and that it is our responsibility as a nation to do something to reduce that gap. Yet there is major disagreement regarding the best remedies to take. Similar to ongoing debate in educational research and policy, one view is that early interventions such as Head Start are not doing a good job teaching disadvantaged preschoolers. (See, for example, the 2003 report on Head Start by the U.S. Department of Health and Human Services). From an economic and partisan perspective, the argument is that Head Start programs (like public schools) are monopolies that are inefficient and have few incentives to improve because of the lack of competition (Kane & Staiger 2002). More strictly defined standards of child performance are seen as a way to impose accountability. Literacy and math skills can be tested, and test scores can yield information about school performance to consumers (e.g., parents, government funding agencies, etc.). The hope is that market-based systems will weed out bad performers and reward higher performers, that providers will strive to improve, and that children will benefit.
The opposing view, held by many early childhood educators and advocates, sees this emphasis on accountability as a way for fiscally conservative policy makers to avoid paying for the relatively expensive solutions that are needed to enact real gains in poor children's educational attainment. This group contends that high-quality early education and care can advance disadvantaged children's learning but that it is not cheap to provide. Advocates and educators in early childhood suggest that if policy makers really wanted to close the education gap, they would make the kind of fiscal investments that are needed to provide children with the things that we know work: comprehensive, full-day services with highly trained, well-paid staff, with fewer children in each classroom, and with more time and resources to devote to learning, literacy, and social and emotional development (National Academy of Sciences 2001). They further argue that changes at the preschool level will not be enough. For Head Start children to maintain the gains they make in preschool, fiscal resources will be needed to improve the elementary schools they attend and-even more daunting-to alleviate home and community stressors that are likely to impede their future academic performance.
Our point is not to take one side of this debate or the other, but to suggest that a strictly cognitive approach to early education and assessment is likely to backfire, regardless of the position taken on best remedies for the "achievement gap" between affluent and poor preschoolers. For the sake of argument, let us consider the highly touted Texas prekindergarten program that Secretary of Health and Human Services Thompson uses as a model of success and as a purportedly strong contrast to Head Start. Using a nonexperimental research design (where investigators can inflate program effects by assigning better-performing schools to the treatment group), the evaluation of the program revealed moderate impacts on children's language scores for less than half of the participating sites (U.S. Department of Health and Human Services 2003, 21-22). At best, this translates to modest success in narrowing the educational gap between low-income Texas preschoolers and their more affluent counterparts. But, even if we believe that the Texas program included the strongest of teaching efforts tied to the best curricular choices, it could just as easily be argued that the program did not meaningfully close the gap between poor and wealthier children. Using such narrow, cognitively oriented definitions of success, not only will programs be viewed as failures but poor children will be viewed as impervious to help.
To avoid this likely scenario, policy makers must understand that vocabulary, pre-reading, and pre-math tests only provide a rough approximation of where preschoolers stand in relation to their agemates, or where they stand relative to their own prior performance. But these tests do not capture the value of a program in supporting the multiple facets of development and learning that are undoubtedly taking place, both in those Texas classrooms and in Head Start classrooms across the country. For example, while IQ gains children make in preschool arguably fade out, graduates of quality intervention programs (including Head Start) are less likely to be retained in grade or placed in special education than similar children without good preschool experience (e.g., Barnett 1995; Currie & Thomas 1995; Reynolds et al. 2001). Clearly a wealth of learning experiences and benefits were accrued during intervention and carried through later schooling, but these were not tapped by cognitive measures. Thus a focus on cognitive outcomes without an understanding of the multiple processes that lead to school success runs the risk of disenfranchising children from learning, disenfranchising good teachers from teaching disadvantaged preschoolers, and disenfranchising voters from the view that investments in young children pay off.
What does early educational research tell us?
Policy makers must also understand that sole reliance on cognitively oriented measures is unsupported by the best scientific evidence we have about ways to support early learning. There is a bounty of scientific literature indicating that children's social and emotional skills are predictive of early achievement, with children's thinking skills and self-regulation likely to play important roles in early learning (McLelland, Morrison, & Holmes 2000; Blair 2002).
One might ask: What does self-regulation have to do with learning the basics such as preliteracy and early math? Children must be able to handle their emotions when sharing instructional materials, taking turns holding or choosing a book for story time, or getting in line. They must be able to focus their attention away from distracting sights and sounds outside the classroom window and toward the task at hand. They must be able to organize their activities and listen to and heed teachers' instructions. Emotionally supportive preschool classrooms foster children's motivation, their development of enthusiasm about school as a good place to be, and positive views of themselves as learners capable of tackling new problems and challenges. Children who are less distractible and more emotionally positive are viewed by teachers as more "teachable." In fact, a majority of teachers surveyed suggested that curiosity, enthusiasm, and ability to follow directions play a potent role in their judgment of children's "readiness" to learn (Rimm-Kaufman, Pianta, & Cox 2000).
Recent research in both areas of cognitive and emotional development has highlighted the ways in which children differ from each other in terms of "executive functioning" or "behavioral self-control." That is, while some children are good at planning, staying organized and focused when given a difficult task, and remaining attentive and calm in a classroom setting, other children have problems regulating their emotions and their attention (see Raver 2002, for a review). Decades of research suggest that (1) Children with emotional and behavioral difficulties are at greater risk for long-term academic problems, and (2) poverty-related stressors impose additional psychological strain on young children that may interfere with their ability to concentrate, pay attention, and control their feelings of sadness and frustration (Campbell, Shaw, & Gilliom 2000). Prevalence estimates suggest that between 7 percent and 25 percent of low-income children enrolled in early educational settings exhibit elevated behavioral problems (Yoshikawa & Knitzer 1997; Gross, Sambrook, & Fogg 1999). Children exposed to high levels of community and family violence also are more likely to be sad and withdrawn, with symptoms of inattentiveness and difficulty interacting prosocially with teachers and peers. In short, these problems are likely to have serious ramifications for learning. Low-income preschoolers' acquisition of preliteracy and other cognitive skills is likely to be suppressed unless the social and emotional domains of learning and development are recognized and supported.
In addition, research suggests that preschool-age children learn more and are more motivated when they are in emotionally supportive, "child-centered" classrooms, as compared to classrooms that emphasize drills, worksheets, seat-work, and "basic skills" (Stipek et al. 1998). In the recent U.S. Department of Health and Human Services report (2003) critiquing Head Start, the authors recognize the importance of teaching pre-academic content "without compromising social and emotional development." In the model Texas program that the report endorses, the evaluation included assessments of children's readiness in both cognitive and socioemotional domains. It is therefore baffling that some leaders want to eradicate social and emotional assessments from Head Start's planned evaluation efforts.
Plans to abandon assessment of children's social and emotional competencies in Head Start represent a grave loss of opportunity for social scientists and educators. With the emotional and behavioral data from the FACES and Head Start Impact Study, we can address questions of how changes in particular noncognitive domains are associated with changes in learning. Without the data that these assessments will provide, researchers will be unable to test the very hypotheses that may lead to teaching and curricula innovations. Finally, if measures of social and emotional development are struck from national evaluations, policy makers will be making a statement that these features are unrelated to learning and are therefore unimportant. A slew of developmental evidence, and a modicum of common sense, should tell them otherwise.
Cautions and recommendations
What will the impact of national testing of Head Start's preschoolers be? We can imagine a range of scenarios that might result from the plan to use cognitively oriented tests to assess Head Start children. One benefit might be that training and technical assistance could be targeted to centers that need the help the most. On the other hand, classrooms in areas with high levels of community and family violence are likely to have children who are less able to weather the behavioral challenges involved in test-taking, so programs serving our nation's most vulnerable families will receive the greatest share of blame and the least amount of help for children's compromised performance. In short, we may repeat past policy mistakes, with Head Start and poor children blamed for their supposed educational failures rather than rewarded and supported for their successes in the face of substantial income and educational inequality. Without being able to predict the outcome, and without being able to forestall the implementation of cognitively oriented assessments, we offer a set of cautions and recommendations.
First, we remind readers that there is no single cognitive "magic bullet" to the problems of poverty or to the achievement gap between economically disadvantaged children and their more affluent classmates. Good curriculum and hard work on the part of teachers may partially remedy that gap, and programs, teachers, parents, and children themselves are to be lauded when such successes are achieved. Certainly, comprehensive services that address families' economic self-sufficiency, housing, health, and welfare are also needed, and we know that those services are expensive. If policy makers genuinely wish to see Head Start and low-income children succeed, they must match their interest in cognitive assessment with a substantially increased investment in families, programs, and teachers so that desired gains can be realized.
Second, we caution readers that there is not clear consensus of the predictive value of cognitive assessments in guaranteeing later school performance. School success likely rests on an integrated foundation, with physical health, cognitive features, and behavioral/emotional adjustment all playing key facilitative roles in children developing positive orientations toward learning (Raver & Zigler 1997; McLelland, Morrison, & Holmes 2000; Blair 2002). Children's beliefs in themselves as capable learners, their skills in working with teachers and peers in prosocial ways, their ability to stay focused and on task, and their capacity to maintain emotional and behavioral self-control may offer important advantages in learning. We will not know the relative importance of these abilities if we do not collect the data. Thus, we urge that the twin foci on both learning and socioemotional outcomes be maintained in all Head Start evaluation and research efforts.
Third, we recommend that current teacher-rated assessments of emotional and social development be continued. But we also recommend that better methods and measures be used to provide more direct assessments (see Raver & Zigler 1997). The task is possible. Emotionally and behaviorally oriented direct assessments were developed and successfully implemented in the national evaluation of Early Head Start that included thousands of toddlers. Researchers have adequate empirical background on which to develop a comprehensive battery, through "consensus conference" on what measures provide most specificity and predictive validity on measurable change in children's emotional and behavioral adjustment. In short, researchers could standardize and validate a short set of age-appropriate measures that could be included in future years of Head Start assessments. Without such direct measurements, children's emotional and behavioral development will always be more vaguely defined and less vigorously measured than their cognitive development.
References
Administration on Children, Youth and Families. 2002. Longitudinal findings on program performance. Third progress report. Washington, DC: U.S. Department of Health and Human Services.
Administration on Children, Youth and Families. 2003. Head Start FACES 2000: A whole-child perspective on program performance. Washington, DC: U.S. Department of Health and Human Services.
Barnett, S.W. 1995. Long-term effects of early childhood programs on cognitive and social outcomes. Future of Children 5 (3): 25-50.
Blair, C. 2002. School readiness: Integrating cognition and emotion in a neurobiological conceptualization of children's functioning at school entry. American Psychologist 57: 111-27.
Campbell, S.B., D.S. Shaw, & M. Gilliom. 2000. Early externalizing behavior problems: Toddlers and preschoolers at risk for later maladjustment. Development and Psychopathology 12: 467-88.
Currie, J. & D. Thomas. 1995. Does Head Start make a difference? The American Economic Review 85: 341-64.
Goal One Technical Planning Group. 1991. The Goal One Technical Subgroup report on school readiness. National Educational Goals Panel (Ed.) Potential strategies for long-term indicator development: Reports of the technical planning subgroups. Report No. 91-0, pp. 1-18. Washington, DC: National Education Goals Panel.
Gross, D., A. Sambrook, & L. Fogg. 1999. Behavior problems among young urban children in low-income urban day care centers. Research in Nursing & Health 22: 15-25.
Kane, T., & D.O. Staiger. 2002. The promises and pitfalls of using imprecise school accountability measures. Journal of Economic Perspectives 16: 91-114.
McLelland, M.M., F.J. Morrison, & D.H. Holmes. 2000. Children at risk for early academic problems: The role of learning-related social skills. Early Childhood Research Quarterly 15: 307-29.
Meisels, S.J. 2003. Can Head Start pass the test? Education Week, 19 March. Retrieved November 11, 2003, from http://www.edweek.org/ew/newstory. cfm?slug=27meisels.h22.
Meisels, S.J. and colleagues. 2003. Head Start Letter, 28 February. Retrieved November 11, 2003, from http://www.fairtest.org/nattest/Head_Start_Letter.html.
National Academy of Sciences. 2001. From neurons to neighborhoods. Washington, DC: National Academy Press.
Raver, C.C. 2002. Emotions matter: Making the case for the role of young children's emotional development for early school readiness. Social Policy Report 16 (3); 3-18.
Raver, C.C., & E.F. Zigler. 1991. Three steps forward, two steps back: Head Start and the measurement of social competence. Young Children 46 (4): 3-8.
Raver, C.C., & E.F. Zigler. 1997. Social competence: An untapped dimension of Head Start's success. Early Childhood Research Quarterly 12: 363-85.
Reynolds, A.J., J. Temple, D. Robertson, & E. Mann. 2001. Long-term effects of an early childhood intervention on educational achievement and juvenile arrest. Journal of the American Medical Association 285: 2339-46.
Rimm-Kaufman, S.E., R.C. Pianta, & M.J. Cox. 2000. Teachers' judgments of problems in the transition to Kindergarten. Early Childhood Research Quarterly 15: 147-66.
Schumacher, R., M. Greenberg, & J. Mezey. 2003. Head Start reauthorization: A preliminary analysis of HR 2210 the "School Readiness Act of 2003." Washington, DC: Center for Law and Social Policy.
Stipek, D.J., R. Feiler, P. Byler, R. Ryan, S. Milburn, & J.M. Salmon. 1998. Good beginnings: What difference does the program make in preparing young children for school? Journal of Applied Developmental Psychology 19: 41-66.
U.S. Department of Health and Human Services. 2003. Strengthening Head Start: What the evidence shows. Washington, DC: Author.
Wagner, S.L. 2003. Assessment in the early childhood classroom: Asking the right questions, acting on the answers. Herr Research Center/Erikson Institute Applied Research in Child Development 4: 1-7.
Yoshikawa, H., & J. Knitzer. 1997. Lessons from the field: Head Start mental health strategies to meet changing needs. New York: National Center for Children in Poverty, Columbia University.
Zigler, E.F., & E.C. Butterfield. 1968. Motivational aspects of changes in IQ test performance of culturally deprived nursery school children. Child Development 39: 1-14.
Zigler, E.F., & S. Muenchow. 1992. Head Start: The inside story of America's most successful educational experiment. New York: Basic Books.
Zigler, E.F., & S.J. Styfco. 1997. A "Head Start" in what pursuit? IQ versus social competence as the objective of early intervention. In Intelligence, genes and success: Scientists respond to The Bell Curve, eds. B. Devlin, S.E. Fienberg, D. Resnick, & K. Roder, 283-314. New York: Springer-Verlag.
Zigler, E.F., & P.K. Trickett. 1978. IQ, social competence, and evaluation of early childhood intervention. American Psychologist 33: 789-98.
C. Cybele Raver, Ph.D., is an associate professor at the Harris School of Public Policy, University of Chicago. She also serves as director of the Harris School's Center for Human Potential and Public Policy.
Edward F. Zigler, Ph.D., is a Sterling Professor of Psychology, Emeritus at Yale University and director of the Bush Center in Child Development and Social Policy in New Haven, Connecticut. He was a member of the planning committee of Head Start.
The first author gratefully acknowledges the intellectual contributions made by the graduate students of her spring seminar on school readiness, offered at the Irving B. Harris Graduate School of Public Policy Studies, University of Chicago, in the Spring of 2003. Both authors thank Sally Styfco for her help with presenting the ideas in this manuscript.
Copyright © 2004 by the National Association for the Education of Young Children. See Permissions and Reprints online at www.naeyc.org/resources/journal.
