Data Collectionmedium
0:00.0

A study examines whether a new blood biomarker predicts depression risk. Consider two measurement scenarios:

Scenario 1: The test has systematic error—it reads 20% higher for ALL individuals, regardless of depression status. (Non-differential error)

Scenario 2: The test is accurate (±5%) for non-depressed people but reads 40% higher for depressed people. (Differential error—depends on outcome)

Which scenario presents a greater threat to validity?