Assessment and Evaluation of Students with Exceptional Needs
Core concepts
Know the assessment vocabulary: screening, pre-referral through MTSS or RTI, eligibility evaluation, progress monitoring, and program evaluation are different purposes requiring different tools. Norm-referenced tests compare a student to a sample and yield standard scores, percentiles, and age or grade equivalents; criterion-referenced and curriculum-based measures compare a student to a criterion or to their own prior performance. Understand reliability, validity, standard error of measurement, confidence intervals, and floor and ceiling effects. Know that a standard score of 100 is the mean with a standard deviation of 15, so roughly 85 to 115 is the average range, and that subtest scatter often reflects measurement error rather than a profile. Know nondiscriminatory assessment: testing in the child's native language or mode of communication, multiple measures, and attention to cultural and linguistic loading of items. Know how to derive measurable goals from baseline data, how to use curriculum-based measurement to compute slope of improvement, and how functional behavior assessment data feed behavior intervention plans.
Traps
Percentile rank is not percent correct; an age equivalent is an ordinal marker, not a developmental promise; a five-point score change is usually not significant. Options resting on one instrument, one informant, or one setting are wrong. So are options that delay evaluation indefinitely to keep trying interventions once data show insufficient response.
Rule of thumb
Use the three-legged stool for any high-stakes decision: multiple sources, multiple settings, multiple points in time. If a leg is missing, the decision is not defensible.