The Reading Essential Skill Screener — Preschool Version (RESS-P): Studies of Reliability and Validity

Abstract

Technical characteristics of the Reading Essential Skill Screener - Preschool Version (RESS-P) were studied using four independent samples of boys and girls aged 3-5 years. A decision efficiency study (N = 91) resulted in a total predictive value (TPV) of .85 when compared with the criterion of teacher report/judgment of emerging literacy at-risk status [sensitivity = .93, specificity = .76, positive predictive power (PPP) = .79, and negative predictive power (NPP) = .92]. Item difficulty (median = .53) and item discrimination (median = .56) indicated overall adequate item selection characteristics. Principal-components analysis resulted in a four-factor extraction, slightly modifying the theoretical structure underlying the RESS-P. Inter-item consistency was r = .93 (N = 738) and r = .92 (N = 175), while 30-day test-retest reliability was r = .89 (N = 119). The RESS-P yielded concurrent validity coefficients (N = 175) of r = .89 with the Test of Early Reading Ability - 2nd edition (TERA-2), r = .80 with the Developmental Test of Visual-Motor Integration (VMI), r = .79 with the Expressive One-Word Picture Vocabulary Test- Revised (EOWPVT-R), and r = .60 and r = .50, with the Test of Visual Perceptual Skills (TVPS) Discrimination and Figure-Ground subtests, respectively. Policy and practice implications were discussed.