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

The predictive validity of PALS-K scores has also been assessed with regard to the major outcome of interest: the Virginia state standards test for reading, the SOL test administered at the end of third grade. In this case we used discriminant analysis, which assesses the extent to which a set of variables can accurately predict group membership on a dependent measure, to determine whether PALS-K subtest scores would accurately classify students according to their SOL performance at the end of third grade. With a sample of 942 students across three school districts, discriminant analysis accurately classified 80% of the sample according to their pass-fail status on the third-grade SOL in reading. These results suggest a significant and potentially important predictive relationship between PALS-K scores and future reading performance as measured four years later by the state's end-of-third grade standards-based reading test.

Discriminant Analysis

We also used discriminant analysis to assess the degree to which the screening tool could accurately predict the classification of students into identified and nonidentified groups. The discriminant function of a measurement tool is particularly important when benchmarks are used to determine which students will receive additional instruction. Identification status was removed to determine if the discriminant function would accurately separate the cases into two groups using group centroids. For the seventh cohort of statewide implementation, discriminant analyses resulted in a significant Wilks Lambda statistic (λ = .33), indicating that PALS-K scores discriminate between identified and not-identified children beyond chance. The discriminant function correctly classified 95% of the total kindergarten sample. This analysis has been consistent since the implementation of PALS-K, with discriminant analyses accurately classifying between 93% and 97% of the kindergarten students across the first seven cohorts.

Kindergarten Teacher Survey Results

Each year the PALS office collects survey data from participating teachers, principals, and school district representatives. This voluntary survey is accessible through the online participants' PALS-K account in the spring. In Spring 2003, 1,559 kindergarten teachers who used PALS-K responded to the online survey. This represents approximately 35% of the 4,415 teachers to whom PALS-K materials were distributed. Survey information was analyzed and compiled into broad themes. Survey results indicated that, overall, kindergarten teachers consider the PALS-K tool to be an effective screening instrument for detecting early reading difficulties. As one teacher observed, "PALS is a very complete assessment. I cannot think of anything that could make this assessment any more valuable." The major themes that emerged from teacher comments fell into three broad categories: (a) instructional outcomes, (b) student outcomes, and (c) timeliness of the measure.

Several teachers noted the capacity of PALS-K to help them target instruction to students in flexible group formats based on specific skill needs, such as specific letter sounds. One teacher noted that using PALS-K helped her to get children with similar needs "working together in a small group." She added, "I get more accomplished." A number of teachers noted the instructional benefits of PALS-K focusing on specific tasks central to literacy development. One teacher reported, "I focus more on letter sounds and writing. I really feel I focus my attention on the areas tested in PALS-K and I feel my students are better prepared for first grade." Several teachers expressed relief for "finally having a clear set of expectations for alphabet and letter sound instruction in kindergarten."