Formative assessment guides ongoing instruction and learning.
Environmental distraction is a source of measurement error.
Metacognition requires reflection on learning strategies and awareness.
Backward design starts with outcomes, then assessment, then instruction.
Content validity concerns adequate sampling of the intended content domain.
Predictive validity concerns how well a test predicts future performance.
Inter-rater reliability is agreement between different scorers.
Cronbach's alpha estimates internal consistency.
A ceiling effect reduces the test's ability to discriminate among students.
Measurement quantifies; evaluation judges quality or value using evidence and criteria.
Criterion-referenced interpretation compares performance to specified competencies.
The test underrepresents the intended construct; it measures recall rather than application.
Authentic assessment uses real-world tasks and multiple evidence sources over time.
Percentile rank compares a student's position with the norm group.
Criterion-referenced tests measure attainment of specified standards.
A rubric includes criteria and descriptions of performance levels.
Ambiguous wording introduces variance unrelated to the intended construct.
Diagnostic pre-assessment identifies prior knowledge and readiness to guide instruction.
Consistency indicates reliability; inaccurate construct representation indicates lack of validity.
Assessment for learning is formative and used to adjust teaching.
Summative evaluation judges final overall effectiveness and outcomes.
A spiral curriculum revisits concepts with increasing depth and complexity.
Evaluation data should inform targeted revision of curriculum components.
Curriculum development typically moves from planning to content and methods, then implementation, then evaluation.
Taba's model is inductive and grassroots, starting with teacher-led diagnosis of needs.