16. Validity

This chapter is a work in progress.

There are four general types of validity:

1. Construct: validity of inferences about the higher-order constructs that represent sampling particulars. There are various types of construct validity, and ways to measure them, which include but are not limited to:

   • Content validity: experts using their judgment that something measures what it is supposed to measure

   • Convergent validity: correlations among two theoretically related constructs are strong and positive

   • Divergent validity: correlations among two theoretically not-related constructs are zero/null

   • Criterion validity: content on one test (predictor) correlates with performance on relevant criterion measures (outcome)

2. Statistical: validity of inferences about the correlation between treatment and outcome

3. Internal: validity about whether the observed relationship between A and B reflects a causal relationship between A and B

4. External: validity of inferences about whether the cause-effect relationship holds over variation in units/persons, treatment variations, outcomes, and settings (UTOS)