Improving Efficiency for Making Screening Decisions: A Statewide Comparison of Early Literacy Curriculum‐Based Measurement Tools

Abstract

Universal screening practices play a critical role in preventing reading difficulties. Screening decisions typically rely on results from several curriculum‐based measurement (CBM) tools. In this study, data from 236 first graders were pulled as a subsample from a statewide study. Participants completed multiple early literacy CBM tools and an outcome measure. Performance differences were compared across tools and publishers to examine classification accuracy. Results show no differences in performance between nonsense word fluency tools across publishers, yet differences were found examining classification accuracy. We also report results of an exploratory analysis examining whether improvements in testing efficiency in early literacy screening are possible via multiple‐gating procedures. Improving the accuracy and efficiency of screening procedures are discussed as implications for practice.

Evaluating a gated screening approach for identifying risk in kindergarten mathematics.

“…Previous researchers have evaluated the benefits of using different screening methods in schools, including comparing gated screening to other methods. Many investigations of gated screening models have been focused on reading (e.g., Ford et al, 2018; Klingbeil et al, 2017; Paly et al, 2022; VanMeveren et al, 2020). For example, Klingbeil et al (2017) investigated the diagnostic accuracy of different methods to identify risk in Grades 3–5 reading, including one-stage screening, multivariate screening, and gated screening.…”

A preliminary evaluation of positive self‐review as an oral reading fluency intervention

“…As one of the criteria for inclusion, AIMSweb™ nonsense word fluency (NWF) probes were used to screen participants' ability to decode words by asking students to read multiple nonreal words (e.g., kus, buf, puq) for 1 min (Pearson, 2012). The AIMSweb™ NWF was found to have a strong positive correlation (r = .86) with the Dynamic Indicators of Basic Early Literacy Skills NWF (Ford et al, 2018), which demonstrated high reliability (r = .80-.90) and concurrent and predictive validity with word reading (r = .60-.90) (Fuchs et al, 2004;Good et al, 2001). AIMSweb™ NWF probes include kindergarten and first-grade level nonsense words.…”

Skill Performance Assessment for Kindergarten Reading Screening Measures: Pilot Study

“…The process typically starts with a universal screening measure for all students, which identifies most of the students as at risk on the screening measure who will also score at risk on the outcome measure (i.e., true positive). To identify a high percentage of true positive score patterns, screening measures often include many "false positive" score patterns; therefore, all students who score at risk on the screening are given an additional assessment(s), prior to providing supplemental supports for any students to rule out as many false positives as possible (Compton et al, 2006(Compton et al, , 2010Ford et al, 2018;Gilbert et al, 2012Gilbert et al, , 2013Van Norman et al, 2016). Researchers have evaluated multiple methods for gated screening, including following the initial universal screening with progress monitoring during general education instruction (e.g., Compton et al, 2006Compton et al, , 2010Gilbert et al, 2013), dynamic assessment (e.g., Compton et al, 2010;Elleman et al, 2011), assessment of prerequisite or additional skills (e.g., Ford et al, 2018;Fuchs et al, 2012), or skill/performance assessment (e.g., Duhon et al, 2004;VanDerHeyden et al, 2001).…”

“…To identify a high percentage of true positive score patterns, screening measures often include many "false positive" score patterns; therefore, all students who score at risk on the screening are given an additional assessment(s), prior to providing supplemental supports for any students to rule out as many false positives as possible (Compton et al, 2006(Compton et al, , 2010Ford et al, 2018;Gilbert et al, 2012Gilbert et al, , 2013Van Norman et al, 2016). Researchers have evaluated multiple methods for gated screening, including following the initial universal screening with progress monitoring during general education instruction (e.g., Compton et al, 2006Compton et al, , 2010Gilbert et al, 2013), dynamic assessment (e.g., Compton et al, 2010;Elleman et al, 2011), assessment of prerequisite or additional skills (e.g., Ford et al, 2018;Fuchs et al, 2012), or skill/performance assessment (e.g., Duhon et al, 2004;VanDerHeyden et al, 2001). Gated screening processes have shown promise by decreasing the number of students who appear to be false positive for risk on the universal screener (e.g., Compton et al, 2006Compton et al, , 2010Gilbert et al, 2013).…”

“…Some researchers followed the universal screening with progress monitoring for at-risk students prior to supplemental intervention, which substantially reduced false positives but extended the assessment period by 5 to 10 weeks (Compton et al, 2006(Compton et al, , 2010Gilbert et al, 2013;Pétursdóttir, 2006). In addition, gated screening procedures have been studied more frequently in first through seventh grades (Compton et al, 2006(Compton et al, , 2010Duhon et al, 2004;Elleman et al, 2011;Ford et al, 2018;Fuchs et al, 2012;Gilbert et al, 2013;Van Norman et al, 2016) than in kindergarten. Evaluating the effectiveness of gated processes in kindergarten is important, specifically because kindergarteners may engage with screening assessments differently than older students, given the novelty of these assessments, and because kindergarten is a key timeframe in which to identify students in need of additional support.…”