Emerging Issues with NCLB

Concerns About Testing Young Children in Early ChildhoodSettings

The increased use of testing at all levels has been an issue in American education, but the testing of young children is of particular concern. Standardized tests and other assessment measures are now being used in preschool, kindergarten, and primary grades to determine whether children will be admitted to preschool programs, promoted to the next grade, or retained. During the late 1980s and early 1990s, tests were used to determine whether students should be promoted from kindergarten to first grade or placed in a “transitional” first grade. Although this practice is now less popular, it persists in some school districts and states (Smith, 1999). In 2000, the National Association of Early Childhood Specialists in State Departments of Education (NAECS/SDE) was concerned about the continuing trend to deny children entry to kindergarten and first grade. They issued a position statement, “Still! Unacceptable Trends in Kindergarten Entry and Placement” (National Association of Early Childhood Specialists in State Departments of Education [NAECS/SDE], 2000). This continuing effort to advocate appropriate assessment of very young children was endorsed by the Governing Board of the National Association for the Education of Young Children (NAEYC, 2001).

By 2006 states used a wide range of types of assessments with young children entering public school. Screening tests were in use in many states for hearing and vision as well as developmental assessments and readiness tests. Many states conducted screening to identify children at risk for failing to succeed in school and/or developmental disorders or disabilities. Some states met the criteria for developmentally appropriate assessments, while others did not. For example, California required observation and portfolio materials in preschool assessments. On the other hand, Georgia students were tested for first-grade readiness at the end of the kindergarten year to determine grade placement (Education Commission of the States, 2006).

The announcement by President Bush in 2003 that all Head Start students would be given a national standardized test assessment raised new concerns. At issue were validity and reliability of tests for preschool children (Nagle, 2000) and whether such “high-stakes” testing should be used to evaluate the quality of Head Start programs (Shepard, et al., 1998). Policy makers had to address these and other concerns about appropriate assessment of young children in their decisions about how to evaluate preschool programs that receive federal funding (McMaken, 2003).

In February 2003 a large group of early childhood experts wrote their congressional representatives to express their concerns about the impending test. They made the following points:

• The test is too narrow.
• The test may reduce the comprehensive services that ensure the success of Head Start.
• The test is shifting resources away from other needs within Head Start.
• Testing should be used to strengthen teaching practices, not evaluate a program, and should in no way be linked to program funding (Fair Test, 2003; NAEYC, 2004).

In September 2003 the new test, the National Reporting System (NRS) (U.S. Department of Health and Human Services [HHS] Head Start Bureau, 2003), was administered by the Head Start Bureau in the Department of Health and Human Services (HHS) Administration for Children and Families to more than 400,000 children ages 4 and 5 and continues to be administered each year. In 2005 when Head Start funding was being considered, the Government Accountability Office (GAO) issued a report on the NRS. The report said that the NRS had not shown that it provides reliable information on children’s progress during the Head Start program year, especially for Spanish-speaking children. Moreover, the NRS had not shown that its results are valid measures of the learning that takes place in the program. In its recommendations, the GAO required that the Head Start Bureau establish validity and reliability for the NRS. As a result the NRS was not to be used for accountability purposes related to program funding (Crawford, 2005; Government Accountability Office [GAO], 2005). Because the Bush administration reportedly intended to use the NRS to establish accountability requirements similar to NCLB, this GAO finding essentially halted use of the test for that purpose.

Concerns About Testing Young Children with Cultural and Language Differences

A concurrent concern related to current trends and practices in the assessment of young children is the question of how appropriate our tests and assessment strategies are in the terms of the diversity of young children attending early childhood programs. Socioeconomic groups are changing dramatically and rapidly in our society, with an expansion of the poorer class and a corresponding shrinking of the middle class (Raymond & McIntosh, 1992). At the same time, an increase in minority citizens has occurred as the result of the continuing influx of people from other countries, especially Southeast Asia and Central and South America. Moreover, Hispanic families are no longer concentrated in the Southwest; their growth in many parts of the country has caused new communities to have unprecedented high percentages of Hispanic children. Seventy-nine percent of young ELLs in public schools speak Spanish. In addition, approximately 460 languages are represented in schools and programs in the United States, including Spanish, Chinese, Arabic, Armenian, and Hmong (Biggar, 2005; Lopez, Salas, & Flores, 2005). Assessment of the developmental progress of children from these groups is particularly important if their learning needs are to be identified and addressed.

Evidence shows that standardized test scores have had a high correlation to parents’ occupations, level of education, the location of the student’s elementary school, and the family’s income bracket. Moreover, students from limited English backgrounds tend to score lower on reading and language fluency tests in English. They typically perform better on computational portions of mathematics tests (Wesson, 2001). The fairness of existing tests for children who are school disadvantaged and linguistically and culturally diverse indicates the need for alternative assessment strategies for young children (Biggar, 2005; Goodwin & Goodwin, 1993, 1997). A major issue in the 21st century is appropriate measurement and evaluation strategies that will enhance, rather than diminish, the potential for achievement.

The history of assessment of minorities who are bilingual students or learning English as a second language is one of potential bias. Children have been and continue to be tested in their nondominant language (English) or with instruments that were validated on an Anglo, middle-class sample of children. As a result, many Hispanic preschool children were and are still regularly diagnosed as developmentally delayed and placed in special education (Lopez et al., 2005). The issue of appropriate assessment of these children was addressed by court cases such as Diana v. California State Board of Education (1968) and Lau v. Nichols (1974). More recently the NCLB and the Head Start NRS have addressed the issue of testing English language learners (Crawford, 2005; David, 2005; GAO, 2005).

The impact of NCLB on testing ELLs has resulted in the development of new English language proficiency tests based on new standards adopted by each state. More important, the tests measure the reading, writing, speaking, and listening skills of ELLs (Zehr, 2006). In the summer of 2006 five states had failed to meet the Department of Education’s deadline to have tests in place. While some states designed their own tests, other states adopted tests designed by consortia or testing corporations. Nevertheless, because test development and implementation were still in the beginning stages, little was known about the validity and reliability of the tests and whether the tests met the requirements of the law.

U.S. Says Language Exam Does not Comply with Law
The state of New York was one of the states that did not meet the requirements of NCLB. In July 2006, the national Department of Education reported that New York’s methods for testing students with disabilities and students with limited English proficiency had to be corrected within a year or the state would risk losing $1.2 million in school aid.
There were many concerns with the ruling because students with limited English would have to take the regular reading test rather than the New York State English as a Second Language Achievement Test. New York state officials replied that they were in the process of addressing the problem, but they were concerned about the large number of students who would fail the regular reading test. As a result, many more schools would receive low ratings under NCLB policies.

Source: Herszenhorn, D. (2006, July 11). U.S. says language exam does not comply with law Retrieved July 12, 2006 from www.newyorktimes.com

The New York example reveals the complexity of the assessment of ELLs. The New York State test was designed to measure language acquisition, while the tests meeting NCLB measured English language skills. This was true for bilingual and ELL programs throughout the United States prior to NCLB. It would take many years to develop and validate tests that would resolve how to assess the language skills of limited-English speakers that were comparable with tests for English-speaking students.

Assessment of young children who are from families that are culturally and linguistically diverse must include many dimensions of diversity. It is not useful to proceed with assessment that is culturally fair for Hispanic or Asian populations generally. The many variations within communities and cultures must be considered, among them the educational background of the parents and the culture of the immediate community of the family. Congruence between the individual cultural perceptions of the assessors and the children being assessed, even when both are from the same culture or language population, must also be considered (Barrera, 1996). Many types of information, including the child’s background and the use of assessments, must be combined to determine a picture of the child that reflects individual, group, and family cultural characteristics (Lopez et al., 2005).

Concerns About Testing Young Children with Disabilities

The use of testing for infants and young children with disabilities cannot be avoided. Indeed, Meisels, Steele, and Quinn-Leering (1993) reflected that not all tests used are bad. Nevertheless, Greenspan, Meisels, and others (1996) believe assessments used with infants and young children have been borrowed from assessment methodology used with older children and do not represent meaningful information about their developmental achievements and capacities. Misleading test scores are being used for decisions about services, educational placements, and intervention programs. These developmental psychologists propose that assessment should be based on current understanding of development and use structured tests as one part of an integrated approach that includes observing the child’s interactions with trusted caregivers. Assessment should be based on multiple sources of information that reflect the child’s capacities and competencies and better indicate what learning environments will best provide intervention services for the child’s optimal development.

Play-based assessment is one major source of information among the multiple sources recommended. Play assessment is nonthreatening and can be done unobtrusively. Moreover, during play, children can demonstrate skills and abilities that might not be apparent in other forms of assessment. Children’s ability to initiate and carry out play schemes and use play materials can add significant information (Fewell & Rich, 1987; Segal & Webber, 1996).

In transdisciplinary play-based assessment, a team that includes parents observes the child at play. Each member of the team observes an area of development. During the assessment the child’s developmental level, learning styles, patterns of interaction, and other behaviors are observed (Linder, 1993).

NCLB has had an impact on curriculum and assessment of children with disabilities. While identification of children can begin very early in life, the needs of the children as they enter public education are not usually identified until first grade. However, during the last ten years, the nature and objectives of kindergarten have changed because of advances in knowledge about what young children are capable of learning and the advent of the standards-based accountability movement. Kindergarteners are taught and tested on the mastery of academic standards. This change in expectations has affected the kindergarten year for children at risk for learning disabilities. The kindergarten year formerly was used to work with at-risk children and refer them for testing at the end of the year. When they reached first grade they would be referred for identification and possible special education services. Children with disabilities or who are at risk for learning problems now need identification and services earlier than first grade. Identification of disabilities and referral for services should now be considered for the kindergarten year, even if some disabilities are difficult to identify in early childhood (Litty & Hatch, 2006).

The NCLB Act also added accountability measures to IDEA.  School districts must test at least 95 percent of students with disabilities and incorporate their test scores into school ratings. There has been strong public reaction to the inclusion of special education students in state testing and reporting. Some policy makers see this provision as an important step in every child receiving a high-quality education. Critics worry that the law is not flexible enough to meet individual needs of students with disabilities. Many teachers felt that special education students should not be expected to meet the same set of academic content standards as regular education students. These issues were yet to be resolved when the final regulations were published in August 2006 for the Individuals with Disabilities Education Improvement Act of 2004 (Education Week, n.d.; U.S. Department of Education, 2006).

Authentic and Performance Assessment

Assessment is in a period of transition. Teachers of young children are moving from more traditional strategies of assessing for knowledge and facts to assessing the students’ ability to reason and solve problems. Despite the demands for accountability for addressing early childhood standards, assessments provide a variety of methods for children to demonstrate what they understand and can do.

A broader view of assessment has incorporated a multidimensional approach to measurement, as described earlier in the sections on concerns for assessment of children from diverse populations and children with disabilities. It is now felt that too much attention has been given to the use of standardized tests, rather than a multidimensional approach that uses many sources of information. The more inclusive practice of assessment, which includes work samples, observation results, and teaching report forms, is called alternative assessment. These alternatives to standardized tests measure how students can apply the knowledge they have learned (Blum & Arter, 1996; Maeroff, 1991). Within this evolution in the purposes for assessment and interpretation of assessments is the move to authentic and performance assessments. Authentic assessments must have some connection to the real world; that is, they must have a meaningful context. They are contextual in that they emerge from the child’s accomplishments. Performance assessments permit the child to demonstrate what is understood through the performance of a task or activity (Wortham, 1998).

Performance assessment as applied through the use of portfolios provides a multifaceted view of what the young child can understand and use. Performance assessment is used because teachers in early childhood programs seek information about the child’s development and accomplishments in all domains. Performance assessment combined with other assessments provides a longitudinal record of change in development, rather than an assessment of a limited range of skills at a particular time. It is appropriately used with infants, young children, school-age children, children from diverse populations, and children with disabilities (Barrera, 1996; Meisels, 1996; Wortham, 1998).

Documentation is a form of performance assessment used by Reggio Emilia schools. First developed in Raggio Emilia Schools Italy and now widely used in the United States, documentation is a process of collecting and displaying children’s work on projects (Wurm, 2005).

This broader view of assessment in early childhood programs is echoed by the organizations that endorsed and supported the Guidelines for Appropriate Curriculum Content and Assessment in Programs Serving Children Ages 3 Through 8, a position statement of the NAEYC and the NAECS/SDE adopted in 1990 and renewed in 2000 and 2001 (NAEYC,1992; NAECS/SDE, 2000). These guidelines proposed that the purpose of assessment is to benefit individual children and to improve early childhood programs. Appropriate assessment should help enhance curriculum choices, help teachers collaborate with parents, and help ensure that the needs of children are addressed appropriately. Rather than being narrowly defined as testing, assessment should link curriculum and instruction with program objectives for young children (Hills, 1992). Authentic and performance assessments provide dynamic assessment approaches that benefit the child, parents, caregivers, and teachers.