Assessment and Evaluation in Special Education
Assessment items on this exam test whether you can select, interpret, and defend assessment data-not whether you can memorize test names. Expect questions about reliability, score interpretation, eligibility decision-making, and the difference between accommodations and modifications during testing.
Purposes and types
- Screening identifies students who may need support; diagnostic assessment pinpoints skills; progress monitoring measures response to intervention; summative assessment evaluates programs.
- Norm-referenced tests compare a student to a sample; criterion-referenced tests compare performance to a standard. CBM (oral reading fluency, math computation probes) is brief, repeatable, and sensitive to growth.
Technical properties
- Reliability = consistency (test-retest, inter-rater, internal consistency). Validity = the test measures what it claims and supports the intended decision. A test can be reliable without being valid.
- SEM creates a confidence band around an obtained score; a standard score of 85 with SEM 4 gives roughly 77–93 at 95% confidence. Report the band, not the point.
- Standard scores: mean 100, SD 15. T-scores: mean 50, SD 10. Stanines: 1–9. Percentile rank is the percentage of the norm group at or below the score-not percent correct.
Eligibility and monitoring
- Georgia requires evidence of response to research-based intervention before SLD identification; the team must also rule out lack of instruction and limited English proficiency.
- Reevaluation occurs at least every three years unless parents and district agree otherwise. Initial evaluation timelines run 60 days from consent.
- FBA uses indirect (interview, rating scale) and direct (ABC observation) data to build a hypothesis; functional analysis manipulates variables.
- Testing accommodations must be listed in the IEP, must not invalidate the construct, and differ from modifications that change what is measured.
Worked example
A student earns a standard score of 82 (SEM 3) in reading comprehension. Reporting '82, below average, with a confidence band of roughly 76–88' is defensible; reporting 'two grade levels behind' from a grade equivalent is not, because grade equivalents are ordinal and do not map to curriculum placement.
Traps
Treating grade- and age-equivalents as interval data; confusing percentile rank with percent correct; assuming one score determines eligibility; letting a test accommodation (reading the reading test aloud) invalidate the score; and confusing screening data with comprehensive evaluation data.