Assessment, Observation, and Data-Based Decisions
What this domain tests
This domain asks you to select, administer, and interpret assessments for children from birth through Grade 3, and to turn observation data into instruction without over-identifying or under-identifying children.
Core rules
- Match purpose to tool: screening identifies who needs a closer look, diagnostic assessment specifies what is needed, progress monitoring tracks response over time, and summative assessment reports end-of-period status. Formative assessment happens during instruction.
- Reliability is consistency of scores across raters and occasions; validity is whether the tool measures what it claims for the intended purpose and population. A tool can be reliable and invalid.
- Observation methods: anecdotal records (brief, contextual), running records (sequential, detailed), event sampling (all instances of one behavior), time sampling (behavior at intervals), checklists and rating scales, and portfolios/work sampling for authentic progress.
- Assessment of dual language learners must consider both languages and distinguish a language difference from a disability; assess in the child's stronger language or use a qualified interpreter.
- MTSS/RTI organizes prevention into tiers, with data-based movement between tiers and fidelity of instruction checked before a referral.
Worked example
A bathroom scale reads exactly three pounds heavy every time. Reasoning: repeated identical readings show consistency, so the scale is reliable, but it does not yield the true weight, so it is not valid for that purpose. Apply the same logic to a classroom checklist completed identically by two raters: inter-rater agreement supports reliability, not validity.
Traps that cost marks
Distractors swap reliability and validity wording. A screener never diagnoses, and a single assessment never determines placement. Judging a child's ability from English-only results when the home language differs is a bias error. Finally, observation notes should record what was seen and heard, not interpretations such as "angry."