Early Childhood Assessment: Why, What, and How (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 Perspectives on Early Childhood Learning Standards and Assessment I n a perfect world, participants in the development of a set of early childhood services at either a local or system level would begin by thinking about what is needed to improve the physi- cal well-being and developmental competence of young children. They would decide what outcomes could be anticipated for children who participate in a particular well-designed program or set of services. They would subsequently concern themselves with what standards and processes would be needed to ensure that participating children would benefit from the program. The planners would select formative assessments to track children’s progress toward the standards and use this information to guide instructional adjustments. And finally, reliable and valid ­processes to assess whether children’s overall development and learning have met the expectations of the planners would be selected and employed. The results of such assessment would be used to refine the program practices with the expectation that the outcomes for children would improve even further. In the real world, this rarely happens. The underresourced complex of early childhood care and education settings in the United States is seldom able to implement the ideal sequence of steps at the local, state, or national level. The federal government, individual states, and local providers usually find themselves working at least partially backward to create workable processes 43 

44 EARLY CHILDHOOD ASSESSMENT to determine what the expectations for children and their families should be, what program standards lead to the accomplishment of those outcomes, and how to assess children’s status related to the standards as a function of program participation. That picture is changing as the early childhood field, as never before, is influenced by and actively reconfigures itself in response to the burgeoning development of state prekindergarten (pre-K) programs and accompanying expectations for documentation of children’s progress, the development of learning standards in K-12 education, the parallel development of state assessment systems, and the accompanying development of quality rating systems across the early care and education sector. This chapter describes the development of well-defined expec- tations for child outcomes—that is, early learning standards—as a function of participation in an early childhood setting of some kind, how these learning standards are being used, and how prac- titioners are able to access information about how to use them. We use the term “early learning standards,” as defined by the Early Childhood Education Assessment Consortium of the Council of Chief State School Officers, in collaboration with several early childhood organizations. Early learning standards are statements that describe expectations for the learning and development of young children across the domains of health and physical well- being, social and emotional well-being, approaches to learning, language development and symbol systems, and general knowl- edge about the world around them (Council of Chief State School Officers and Early Childhood Education Assessment Consortium, 2007). Until recently the very idea of defined expectations for what children should know and be able to do at particular times in these very early years of their lives was rejected by many in the early childhood field. Policy makers, researchers, program lead- ers, and teachers have historically depended on structural pro- gram and process standards (e.g., the qualifications of staff, group size and ratio, nature of the curriculum, provisions for parental involvement, the nature of adult and child interaction) to assess whether a program was offering a quality experience for children. These sets of program and process standards exist in forms as diverse as the minimum regulations each state requires for child 

PERSPECTIVES ON EARLY CHILDHOOD LEARNING STANDARDS 45 care settings, to requirements for operating the federal Head Start program, to regulations for state prekindergarten programs, to standards for National Association for the Education of Young Children accreditation (National Association for the Education of Young Children, 2006). Program standards can reflect the mini- mum floor under which a program cannot operate, such as in the case of the states’ child care regulations, or they can be the high- est quality requirements, as in the case of the new Accreditation Standards of the National Association for the Education of Young Children (2006). DEVELOPMENT OF EARLY LEARNING STANDARDS Decades of research on effective programs have demonstrated that children participating in programs adhering to high-quality program and process standards exhibit improved developmental and learning outcomes compared with children with no program or those experiencing a low-quality program (Ackerman and B ­ arnett, 2006; High/Scope Educational Research Foundation, 2002). Many states making an investment in prekindergarten conduct evaluations of program quality and, in some cases, assess child outcomes. These studies are in addition to the regu- lar program monitoring done to ensure that programs meet state standards, and they have increased in number as more and more states have begun to invest public money in prekindergarten ( ­ Gilliam and Zigler, 2001). Michigan, for example, has compelling longitudinal program evaluation data on the link between pro- gram quality and child outcomes in the Michigan School Readi- ness Program (High/Scope Educational Research Foundation, in press; National Institute for Early Education Research, 2005). Few other public or private programs (e.g., child care, private preschools) are subject to either quality-driven program standards or requirements for assessing child outcomes. The earliest state early learning standards were developed by states operating pre-K programs (typically for 3- and 4-year-olds or just 4-year-olds). Such standards were developed on the premise that evaluation of child outcomes could not be done without a set of early learning standards against which to measure children’s progress. Since the early 1990s, there has been an explosion of 

46 EARLY CHILDHOOD ASSESSMENT activity around the development of state learning standards, and every state now has them except North Dakota (where they exist in draft form). National early learning standards, such as those developed for Head Start and by subject-­specific professional orga- nizations, have also been created (Council of Chief State School Officers and Early Childhood Education Assessment Consortium, 2003a; U.S. Department of Health and Human Services, Admin- istration for Children and Families, 2003). A set of model early learning standards has been developed by a national committee of experts (Pre-kindergarten Standards Panel, 2002), although a 2003 study found that few states made specific reference to this docu- ment (Council of Chief State School Officers and Early Childhood Education Assessment Consortium, 2003b). Virtually every report or article about states and their devel- opment of early learning expectations begins with an expression of surprise about how quickly the development process unfolded across the nation (see Box 3-1). The development and implemen- tation of these standards reflect a significant shift in how the field has viewed the usefulness of setting expectations for young children’s learning and development. Appendix C provides more information about state early childhood standards. While acknowledging that adherence to high-quality program standards substantially increases the likelihood that participat- ing children will benefit from the program, advocates have been forceful in expressing reservations about creating these sets of expectations (Hatch, 2001; National Association for the Education of Young Children and National Association of Early Childhood Specialists in State Departments of Education, 2002). Such reser- vations include a number of concerns: • The threat of ignoring the variability of children’s develop- ment and learning and of their experiences. • Worry that early labeling of the most vulnerable children as “failures” puts their access to appropriate instruction and thus their future development at risk. • Unfairly judging programs on the basis of whether par- ticipating children meet standards, without taking into account their status at entry to the program or information about the resources available to the program. 

PERSPECTIVES ON EARLY CHILDHOOD LEARNING STANDARDS 47 BOX 3-1 The Development of Major Early Learning Standards 1989 Goal 1, “All children ready to learn,” articulated by the nation’s governors at education summit 1995 Publication of Reconsidering Children’s Early Development and Learning (Kagan, Moore, and Bredekamp, 1995) 1998 Publication of Preventing Reading Difficulties (National Research Council, 1998) Publication of Principles and Recommendations for Early Childhood Assessments (Shepad, Kagan, and Wurtz, 1998) 1999 10 states have standards for children ages 3-4 2000 Publication of From Neurons to Neighborhoods (National Research Council and Institute of Medicine, 2000) Publication of Head Start Child Outcomes Framework (U.S. Department of Health and Human Services, Administration for Children and Families, 2000) 2001 Publication of Eager to Learn (National Research Council, 2001) 2002 17 states have standards for children ages 3-4; 4 states have standards for children ages 0-3 Good Start, Grow Smart initiative (White House, 2002) launched Head Start National Reporting System launched 2007 49 states have standards for children ages 3-4; 18 states have standards for children ages 0-3 Publication of Taking Stock: Assessing and Improving Early Childhood Learning and Program Quality (National Early Childhood Accountability Task Force, 2007) States now required to report outcomes data for children with disabilities served through Part C and Part B of the Individuals with Disabilities Education Act as part of their Annual Performance Report 

48 EARLY CHILDHOOD ASSESSMENT • The risk of children being unfairly denied program partici- pation based on what they do or do not know. • The risk that responsibility for meeting the standards will shift from the adults charged with providing high-quality learning opportunities to very young children. • Whether high-quality teaching will be undermined by the pressure to meet standards, causing the curriculum to become rigid and focused on test content and the erosion of a child-centered approach to curriculum development and instructional practices. • Whether switching to child outcome standards as the sole criterion for determining the effectiveness of programs or personnel is unfair. Early childhood services continue to be underresourced, and poor child outcomes may reflect the lack of resources. • Misunderstanding of how to achieve standards frequently appears to engender more teacher-centered, didactic practices. Although these concerns cannot be dismissed, it is important to note that early learning standards were developed as a tool to improve program quality for all children. Their rapid develop- ment has resulted from a combination of policy shifts and an emerging practitioner consensus, influenced by a number of factors: • The standards-setting activity in K-12 education, which gained momentum after the 1990 establishment of the National Education Goals Panel and the subsequent pas- sage of Goals 2000 by Congress in 1994. This act and its accompanying funding led states to develop or refine K-12 standards in at least the areas of English language arts, mathematics, science, and history. • Greater understanding about the capabilities of young children. Earlier work of the National Research Council (NRC) has played a key role in informing and developing that understanding and thereby supporting the develop- ment of early learning standards. The most influential NRC document influencing the development of standards for 

PERSPECTIVES ON EARLY CHILDHOOD LEARNING STANDARDS 49 preschool-age children has been Eager to Learn: Educating Our Preschoolers (National Research Council, 2001). Other important influences include From Neurons to Neighbor- hoods: The Science of Early Childhood Development (National Research Council and Institute of Medicine, 2000) and Preventing Reading Difficulties in Young Children (National Research Council, 1998). • Linking of the development of early learning standards with receipt of federal funds from the Child Care and Development Fund for each state (U.S. Department of Health and Human Services, Administration for Chil- dren and Families, 2002). The requirement that all states develop voluntary early learning guidelines in language, literacy and mathematics followed the release of the 2002 early childhood initiative, Good Start, Grow Smart (White House, 2002). HEAD START CHILD OUTCOMES FRAMEWORK Head Start is a large, well-known federally funded early childhood services program, serving over 909,000 children in FY 2006. Actions taken by Head Start are highly visible and embody federal policies toward early childhood services. The follow­ing narrative provides some background for understand- ing the evolution of the Head Start National Reporting System. Development of the Framework The Head Start Child Outcomes Framework was developed in response to an unfolding set of congressional mandates beginning with the 1994 reauthorization of the Head Start Act, which man- dated the development of measures to assess services and admin- istrative and fiscal practices, to be usable for local self-assessment and peer review, to identify Head Start strengths and weaknesses, and to identify problem areas (Section 641A). The earliest response to this mandate by the Head Start Bureau was the creation of a Pyramid of Services diagram that local programs could use to support and inform continuous program improvement efforts (see Figure 3-1). The pyramid was 

50 EARLY CHILDHOOD ASSESSMENT also used in the formulation of the Family and Child Experiences Survey (FACES) (McKey and Tarullo, 1998). When Head Start was reauthorized in 1998, programs were required to include specific child outcomes in their self-assessment process. This requirement led in 2000 to the development of the Child Outcomes Framework (U.S. Department of Health and Human Services, Administration for Children and Families, 2000). The development process was informed by the participation of a committee of outside experts (the Head Start Bureau Technical Work Group on Child Outcomes), who used the Pyramid of Ser- vices as a basis for their deliberations. Bureau staff also consulted standards documents from profes- sional associations and the existing state early learning standards, of which 10 sets existed at the time. Although those sets of state standards displayed some common elements, great disparity was reflected in the ways the developmental domains were described and in which domains were included. Some included only a few domains, such as language and literacy; others reflected the five dimensions described by the National Education Goals Panel Goal 1 Technical Planning Group (Kagan, Moore, and Bredekamp, 1995) or additional content-related domains (e.g., social studies, science, mathematics, arts). As had the state leaders, the developers of the Head Start Child Outcomes Framework struggled with how to organize learning expectations for Head Start children. They settled on eight broad categories that include the domains in the Goal 1 document (Kagan, Moore, and Bredekamp, 1995), with the addi- tion of the content categories of mathematics, science, and the arts. Expectations related to social studies were included under the social emotional domain as “knowledge of families and com- FACES employs direct assessment items from several nationally normed early childhood instruments, along with teacher reports, parent reports, and obser- vation, to assess numerous cognitive and socioemotional outcomes. It follows children from their Head Start experiences through kindergarten and through the 1997 cohort into first grade (U.S. Department of Health and Human Services, Administration for Children and Families, 2006a, available: http://www.acf.hhs. gov/programs/opre/hs/faces/index.html). From Thomas Schultz via personal communication with committee member Harriet Egertson. 

PERSPECTIVES ON EARLY CHILDHOOD LEARNING STANDARDS 51 CHILD’S s me SOCIAL tco COMPETENCE Ou 1 2 ENHANCE STRENGTHEN children’s families as the growth and primary nurturers development. of their children. es cess 3 4 PROVIDE LINK Pro children with educational, children and families to needed health, and nutritional services. community services. 5 ENSURE well-managed programs that involve parents in decision making. FIGURE 3-1  Head Start Program performance measures conceptual frame- work. SOURCE: U.S. Department of Health and Human Services, ­Administration Alternate Fig 3-1, from downloaded source, editable vectors for Children and Families (2006). munities.” The eight general domains in the final ­ document— l ­ anguage development, literacy, mathematics, science, creative arts, social and emotional development, approaches to learning, and physical health and development—were divided further into 27 domain elements, and 100 examples of more specific indicators of children’s skills, abilities, knowledge, and behaviors considered to be important for school success (U.S. Department of Health and From S.A. Andersen via personal communication with committee member Harriet Egertson. 

52 EARLY CHILDHOOD ASSESSMENT Human Services, Administration for Children and Families, 2003). Among the 100 indicators were 13 specific, legislatively mandated domain elements or indicators in various language, literacy, and numeracy skills. Two indicators are specific to the desired out- comes for young children learning English. The framework was clearly intended to provide guidance for ongoing child assessment and program improvement efforts. Several caveats are specified in the introduction: the framework is intended to focus on children ages 3 to 5 rather than younger children and to guide local programs in selecting, developing, or adapting an assessment instrument or set of assessment tools. The framework is not intended to be an exhaustive list of everything a child should know or be able to do by the end of pre- school or to be used directly as a checklist for assessing children. There is no mention of its relationship to curriculum development. The introduction further attempts to broaden practitioner under- standing of the use of the framework: “Information on children’s progress on the Domains, Domain Elements and Indicators can be obtained from multiple sources, such as teacher observations, analysis of samples of children’s work and performance, parent reports, or direct assessment of children. Head Start assessment practices should reflect the assumption that children demonstrate progress over time in development and learning on a develop- mental continuum, in forms such as increasing frequency of a behavior or ability; increasing breadth or depth of knowledge and understanding; or increasing proficiency or independence in exer- cising a skill or ability” (U.S. Department of Health and Human Services, Administration for Children and Families, 2000). Good Start, Grow Smart Initiative The next step in the federal effort to prepare children to suc- ceed in school with improved Head Start programs came in 2002. President George W. Bush mandated the Good Start, Grow Smart initiative to help states and local communities strengthen early learning for young children. As described in the executive sum- mary of the initiative, President Bush directed the Department of Health and Human Services (HHS) to develop a strategy for assessing the standards of learning in early literacy, language, and 

PERSPECTIVES ON EARLY CHILDHOOD LEARNING STANDARDS 53 numeracy skills in every Head Start center. Every local program was required to assess all children between the ages of 3 and 5 on these indicators at the beginning, middle, and end of each year and to analyze the assessment data on the progress and accom- plishments of all enrolled children. Federal program monitoring teams were to conduct onsite reviews of each program’s imple- mentation of these requirements. HHS was also directed to design a national reporting system to collect data from every local program. This system, combined with ongoing Head Start research and onsite program monitoring reviews, was envisioned as a source of comprehensive informa- tion on local program effectiveness. Local program data would be used to target new efforts in staff training and program improve- ment to enhance the capacity of Head Start to increase children’s early literacy and school readiness. In addition, data on whether a program is successfully teaching standards of learning would be used in HHS evaluations of local Head Start agency contracts (White House, 2002). Head Start National Reporting System The Administration for Children and Families (ACF) responded to the mandate of the Good Start, Grow Smart initiative to assess children’s progress against uniform national standards by develop- ing the National Reporting System (NRS), an instrument to be used to assess all 4- and 5-year-olds in Head Start. The NRS was developed by a contractor, Westat, on an accel- erated schedule. Work began in August 2002. Westat recruited a Technical Work Group of experts in child development, assess- ment, measurement, and program evaluation as advisers and also used focus groups and other methods to gather information and plan the NRS. After a field test in spring 2003, ACF approved a 15-minute assessment battery, trained Head Start program per- sonnel as assessors, and implemented the NRS for the first time in fall 2003. The NRS in its original form assessed skills in four areas: (1) comprehension of spoken English, tested with a “language screener,” (2) vocabulary, (3) letter naming, and (4) early math- ematical skills. Westat and its advisers did not include other 

54 EARLY CHILDHOOD ASSESSMENT domains because of the difficulty in finding high-quality instru- ments that would meet NRS requirements. Most of the items in the NRS battery were taken from existing assessment instruments that had been used in Head Start research or in local Head Start assessment programs. A Spanish-language version of the assessment was developed as well. In the first year of implementation, it was administered after the English version to children whose home language was Spanish and who passed a Spanish language screener. Thus all children were assessed in English or Spanish only if they had passed the screener for that language. The NRS aroused much concern on the part of some early childhood experts. More than 200 educators, researchers, and practitioners signed letters to Congress in early 2003 laying out their concerns about the NRS, along with some suggested ways to improve it. The letters ended with the following words: “If we can move ahead on adopting a matrix sampling design for the proposed Reporting System; if we can ensure that the System is composed of subtests that are reliable, valid, and fair; and if we can have adequate time to learn how to mount this historically largest-ever effort to test young children without creating chaos and confusion, then we will have created a system that has a chance of assisting young, at-risk children” (Meisels et al., 2003). In May 2005, the Government Accountability Office (GAO) released a report on the first year of implementation of the NRS (U.S. Government Accountability Office, 2005). In it, the GAO identified several weaknesses in the system and its implementa- tion, noting: “Currently, results from the first year of the NRS are of limited value for accountability purposes because the Head Start Bureau has not shown that the NRS meets professional standards for such uses, namely that (1) the NRS provides reli- Among the other criticisms of the NRS was dissatisfaction with the omission of any measure of socioemotional development. A socioemotional component, based on teacher observations over a 1-month period, was added to the NRS as of the fall 2006 administration. For that administration, teachers were asked to assess only children who had been in the program for at least 4 weeks. It included items asking the teacher to report on approaches to learning, cooperative classroom behavior, relations with other children, and behavior problems (U.S. Department of Health and Human Services, Administration for Children and Families, 2006b). 

PERSPECTIVES ON EARLY CHILDHOOD LEARNING STANDARDS 55 able information on children’s progress during the Head Start program year, especially for Spanish-speaking children, and (2) its results are valid measures of the learning that takes place” (U.S. Government Accountability Office, 2005, “Highlights”). The American Educational Research Association, along with a smaller group of experts, went on record with their reservations about the NRS later in 2005, when legislation was under consid- eration to suspend its implementation (American Educational Research Association, 2005; Yoshikawa and McCartney, 2005, personal communication to U.S. House of Representatives). The National Head Start Association expressed its concerns in a ­letter to the Office of Management and Budget (OMB) in 2006, after plans for continuing implementation of the NRS were submitted for OMB clearance. Believing that the burden of the reporting system on Head Start programs had been underestimated by ACF and that the results to be gained by continuing it did not justify the burden, the National Head Start Association requested that implementation of the NRS be suspended. Reactions like these were among the factors that led to the con- gressional request for this National Academies study. The reautho- rization of the Head Start program (P.L. 110-134, 2007) was signed into law in December 2007, while the current study was under way. It requires ACF to discontinue administration of the NRS in its current form, directing it to take into account the results of this National Academies report and of other scientific research in any new assessment design, development, and implementation. At the time of this writing, administration of the NRS has been terminated, and ACF is under a requirement to follow a more r ­ igorous process as it develops new assessment tools for Head Start. Other early childhood programs and funders, including state and local agencies charged with overseeing child development programs, are also working to devise assessments that can serve to improve the provision of services to children and to ensure better outcomes. This committee’s challenging task is to provide useful guidance for all these efforts.