Does it match reality, and does it travel?

The last phase of validation turns outward: do the scores line up with something in the world? A confusion matrix holds the answer — sensitivity, how many true cases the instrument catches, and specificity, how many non-cases it correctly clears. Both matter, and screening tools trade one against the other deliberately. Then the harder question: does it travel? An instrument crossing a border can fail on the response format alone, before anyone reaches the content, because agreement scales are not read the same way everywhere.

Also covered: criterion validity · sensitivity and specificity · cross-cultural validity · response formats

Alongside this step

Slides and transcript — the whole video
Read the chapter — Measurement and Construct Validation

Check yourself

5 questions, marked as you go. Nothing is recorded as a grade and nothing is locked behind it — if a question catches you out, that is the useful part.

Lesson 4 · 7 steps

1Even counting the dead takes judgment5:59 2From conceptual model to measurement6:14 3Finding good indicators5:04 4Indexes and scales6:44 5Validity is not a property of an instrument7:43 6Factor analysis and reliability10:27 7Does it match reality, and does it travel?5:45

← Previous step

The Validity Toolkit · lesson 4 of 4