Safety Words Inventory and Literacy Screener: Standardization and Validation

Abstract

When evaluating a school-aged patient, particularly for such issues as attention deficit disorder, or learning disabilities, health-care providers are encouraged to appraise children's school performance. Although some clinicians seek teachers' comments, recent research suggests that teachers' appraisals have limited value. As a consequence, health-care providers need brief methods, functional for busy primary care clinics, for detecting school problems. Because well-visits also include other services such as anticipatory guidance and safety and injury counseling, it would be helpful for detection tools to facilitate delivery of more than one aspect of well-child care. This study was undertaken to develop a brief method for evaluating school performance for elementaryage children and test its accuracy in identifying children with reading and other academic problems. By using common safety signs as the test stimuli, such a tool should also serve as a springboard to injury prevention counseling. Data were drawn from the standardization and validation data of an individually administered wide-range diagnostic achievement test using a geographically diverse, nationally representative sample of 934 children between 5 and 13 years of age. Children were administered the Comprehensive Inventory of Basic Skills-Revised (CIBS-R), which includes among its 10 subtests a 57-item measure of safety word recognition. Data were collected from previously administered measures of academics and intelligence. Forty-one children were administered the measure twice to assess inter-rater reliability. Logistic regression analyses were deployed using children's performance on the nine other subtests of the CIBS-R as the grouping variable, and as predictors, recognition of 57 safety words/phrases. Of these, 22 safety signs (e.g., “Keep out,” “No Trespassing,” “Danger”) were significant predictors of overall academic performance. An additional seven safety words were added due to their ecological significance (e.g., “High Voltage,” “Beware of Dog,” “Poison,” etc.). Receiver operating characteristic was used to determine cut scores based on children's ages in relation to overall academic performance. Cut scores increased with age and produced sensitivity of 78% (range, 73%-88%) and specificity of 84% (range, 77%-87%) to academic performance above and below the 25th percentile-the point where children typically fail to benefit from group instruction and become eligible for remedial reading programs (such as Title I). Children with performance below cut scores were 11 times more likely to have been retained in grade and five times more likely to children with poor performance were not receiving special education or Title I services. Interrater reliability coefficients were high and ranged from .64 to .98. Extrapolating administration times from data on the CIBS-R as a whole suggests the reduced safety words measure takes 3 to 4 minutes to administer and score. The new measure, called the Safety Word Inventory and Literacy Screener (SWILS), consists of 29 words, takes little time to administer and score, and has a high degree of sensitivity and specificity to academic deficits. Use of the SWILS in primary care should enable clinicians to provide focused safety counseling while also screening for school problems.