Early Literacy Screening in Kindergarten: Widespread Implementation in Virginia

Journal of Literacy Research, Winter 2004/2005 by Invernizzi, Marcia, Justice, Laura, Landrum, Timothy J, Booker, Keonya

Children's abilities across these four early literacy dimensions serve as important predictors of subsequent reading achievement (e.g., Stuart, 1995; Tunmer, Herriman, & Nesdale, 1988). Importantly, some children, such as those who have had only limited language and literacy experiences during the preschool period, will not have developed these requisite early literacy skills at the same rate as evidenced by others in their peer groups (e.g., Chaney, 1994; Dickinson & Snow, 1987; Lonigan, et al., 1999). These children enter first grade lacking adequate early literacy knowledge, which in turn places them at risk in their ability to meet the rigors of formal reading instruction in later grades. This quandary begs the need for widespread screening of early literacy in the period immediately preceding formal reading instruction. Indeed, failing to identify kindergarten children exhibiting delays in early literacy attainment is a risky venture. Research shows that reading difficulties in primary grades continue from elementary school throughout high school (Mather, Bos, & Babur, 2001). In light of these findings, it seems clear that accurate early literacy screening may well provide a powerful avenue for reducing current and future reading problems.

Perhaps the most important issue in implementing widespread early literacy screening in kindergarten is instrumentation. An early literacy screening protocol must meet several important criteria (Justice, Invernizzi, & Meier, 2002). First, the screening tool must be broad-based. It must examine children's skills across all primary dimensions of early literacy, including phonological awareness, alphabet knowledge, concept of word in text, and grapheme-phoneme correspondence. Teachers must have an idea of student development in all four domains, not just one or two. Second, the screening device must be sensitive. The instrument needs to effectively and accurately differentiate those children who require enhancement of early literacy skills from those who do not. Third, it must be easily and efficiently administered. A variety of early childhood professionals (e.g., special and regular educators, paraprofessionals) should be able to administer the screening instrument in a short period of time and with little formal training in its use. Furthermore, the quantitative results of each screening should be easily understood and instructionally transparent. Instructional consequences should be easily applied within the larger scope of the appropriate grade-level reading curriculum.

In addition to these criteria, an early literacy screening measure should meet minimum standards of technical adequacy for validity and reliability. A valid instrument measures what it purports to measure. This is typically demonstrated by comparing the results of an assessment with results obtained from other independent measures of the same constructs. Comparisons may be made with measures taken at the same point in time (concurrent validity) or at future points in time (predictive validity). A reliable instrument produces consistent and trustworthy results. For example, researchers often administer an assessment a second time within a matter of days of the initial administration (the time frame is brief to reduce the effects of learning) to determine the extent to which results are consistent (test-retest reliability). Assessment results should also be similar regardless of who administers the assessment (inter-rater reliability), and item-to-total correlations (a form of internal consistency) should be similar for all demographic groups assessed. Test developers must meet basic standards for validity and reliability as outlined by the Standards for Educational and Psychological Testing (AERA, APA, & NCME, 1999). But as researchers have frequently pointed out, there is a trade-off between validity and reliability and the tension that exists between scientifically based research standards for assessment versus the instructional needs of teachers. A broad-based, sensitive, and easily and efficiently administered instrument must find a way to solve this validity dilemma.