Common Practices for Good Assessments

The National Early Childhood Accountability Task Force (PDF), a group of nationally renowned education and early childhood-development experts, was established to study and outline accountability measures that could be effective for early education programs. Pre-K Now shares their view that all assessments must be:

• Comprehensive - measuring student skills across the development spectrum (i.e., cognitive, emotional, social, linguistic, and physical),
• Ongoing - recording student gains over time rather than capturing only a snapshot of a student's developmental level, and
• Contextual - conducted in the classroom environment when students are engaging with their peers and teachers.

In their research, the NRC stressed the importance of clarity of purpose: States should be explicit about the goal of the assessment, whether to determine the progress of individual children or to gauge program effectiveness. In addition, to provide valuable, reliable findings, early education assessments must be standardized; that is, they must utilize consistent evaluation and data-collection methodologies across programs and localities. Also, assessments must involve all students, including English language learners and children with disabilities, and must be appropriate for measuring gains across diverse backgrounds, cultures, and ability levels.

Types of Assessment: Child vs. Program

Child Assessments

Individual child assessments that systematically measure early learners' gains in literacy, numeracy, and social and emotional development are critical for generating valuable data on the needs and progress of every student. Importantly, however, findings from the National Institute for Early Education Research (NIEER) and other research indicate that individual assessments involving written tests, such as those mandated by NCLB, that track yearly academic progress, are neither developmentally appropriate for nor characteristic of early childhood education. Pre-k-age children learn through play, and assessments should correspond to that dynamic. The National Early Childhood Accountability Task Force does recommend, though, that early education assessments be closely aligned to tests that measure development through grade three to give a coherent picture of each child's learning profile throughout their academic career.

Rather than employing tests similar to those mandated by NCLB, individual child assessments in early education should be broad methodologically and contextually, utilizing both direct assessment and an observation. Observational methodologies are often criterion referenced, meaning a child's performance is observed and then compared to a series of criteria describing what a pre-k-age child should know and be able to do.

Such observations are conducted in the classroom during interactions with teachers and other children and are combined with a comprehensive and ongoing analysis of children's individual work over the course of the entire pre-k year, usually called a "portfolio." Such observational assessments are used to evaluate children in the context of their typical learning environment and are more likely to provide reliable accounts of social and emotional development. The chief drawback to observational and portfolio assessments is the inability to train a large number of assessors across varied program types to interpret observations of a child's behavior and portfolio in a consistent way.

One type of individual child assessment that does not have this limitation and is easily standardized is direct assessment. This evaluation type involves the administrator (likely a trained assessor or teacher) giving each student, individually, a series of direct verbal instructions that have standardized, correct responses. The assessor then records whether or not the child performs each task correctly. Among early childhood researchers, the value of direct assessment is a topic of considerable debate. Because correct answers are easily differentiated from incorrect ones, direct assessments require only minimal subjective interpretation by the assessor, and they result in quantifiable data that can be compared across programs. Nevertheless, direct assessments pose considerable challenges. If assessors are not well trained, direct assessments can cause stress for children and produce unreliable results. For this reason, direct assessments, when used, should always be coupled with observational data.

All individual assessments should begin with screenings for mental, visual, and auditory health and to determine whether or not a child should be tested for a developmental disability. Further, pre-k programs should ensure that teachers and administrators know how to interpret and use data gleaned from all types of assessments. For instance, poor performance on an individual child assessment should not be construed to mean a student suffers from a long-term developmental disability any more than it means a student will be unable to function in a normal academic environment after receiving special attention.

Program Assessments

When considering public investments, policymakers need to know taxpayer money is well spent. Evaluations of programs can be used to inform best practices, guide investment decisions, and hold programs accountable for results.

Though individual assessments on every single child can yield information on program effectiveness, they are not always the most effective strategy for assessing overall program performance. In particular, they can be expensive and difficult to conduct in large programs. To ameliorate these problems, evaluators can perform individual child assessments using a representative sample group of students in a program. Such evaluations should employ subgroups comprising all student ethnicities and income levels and would likely yield data sufficient to indicate how all the program's students would score on individual assessments. Finally, though it is generally appropriate for teachers to conduct child assessments of their own students, objective, external experts, such as university faculty or state officials, should perform program evaluations.

Other evaluation strategies can be employed as alternatives to or, ideally, in conjunction with individual child assessments to measure program effectiveness. In classroom environment evaluations, independent analysts assess teaching practices and educational opportunities provided to children (e.g. number of books, layout of classroom space, activities of children during transition time, quality and quantity of teacher-child interactions). Additionally, both teachers and parents should be given the opportunity to rate the pre-k program to provide a broader assessment of program quality beyond scores on standardized evaluations.

Data gleaned from program assessments should always be used in the appropriate context. For instance, the results from sample group assessments or evaluations of teacher-child interactions are best used to measure the effectiveness of an individual program. Conversely, analyses of the classroom environment, spending practices, record-keeping systems, and other quality-rating mechanisms are used to guide funding and policy decisions and to determine the effectiveness of state programs as a whole. When these types of program evaluations are both ongoing and well designed, the National Early Childhood Accountability Task Force notes that, "Results can inform state decisions about funding different types of programs."