Fire-Control Technician Achievement Test Results for Adaptive Testing Simulation

Test results for 365 fire-control technicians on a 232-item paper-and-pencil achievement test, divided into 12 subtests covering different content areas. The data was used in a real-data simulation to compare an adaptive testing strategy with conventional testing. The study was authored by Joel M. Brown.

Use Cases

• Simulating adaptive test battery performance based on real test results from 365 individuals.
• Comparing test information and length between adaptive and conventional testing strategies.
• Analyzing correlations between subtest scores from different testing methods across 12 content areas.
• Evaluating the efficiency of adaptive item selection within and between subtests.

Strengths

• Data is based on a real administration of a 232-item achievement test to 365 individuals.
• The simulation demonstrated a 50% average reduction in test length with adaptive testing, with specific examples ranging from 18% to 80% of the original item count.
• Correlations between adaptive and conventional subtest scores were .90 or higher for 11 of the 12 content areas.

Limitations

• The underlying raw data (columns, rows) is not available; only the simulation results are described.
• Column-level documentation is absent; field semantics must be inferred from the description.
• Row count for the original dataset is unknown, which may limit suitability assessment.

Provenance

Source

Joel M. Brown

Collection Method

Real-data simulation using results from a paper-and-pencil test administration.