Standards for Educational and Psychological Testing (2014)
2 Reliability/Precision and Errors of Measurement – Standards for Educational and Psychological Testing (2014)

Standards for Educational and Psychological Testing (2014) 2.9: Precision evidence for long and short versions

What does 2.9 require?

When both long and short versions exist, the developer should report precision evidence for scores on each version, preferably from separate administrations to independent samples. Predictions carried over from the full test can rest on assumptions that fail, and short forms often use nonrandom item subsets. Direct, independent evaluation of both versions is therefore preferred where feasible.

Maintained by Gerard Blokdyk

Other controls in 2 Reliability/Precision and Errors of Measurement – Standards for Educational and Psychological Testing (2014)

Query this from an agent

The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.