A study on a new drug has randomized groups, proper blinding, and controlled variables → High internal validity (results are trustworthy).
A study with selection bias, confounders, or poor blinding → Low internal validity (we don’t know if the drug actually works).
🚨 Contrast with external validity:
If this study was done only in healthy young males, it has high internal validity (well-controlled), but low external validity (can’t apply to older adults, women, diabetics, etc.).
If the study includes a diverse population, it has higher external validity (more generalizable).
📌 How to improve internal validity?
✔ Randomization (removes selection bias)
✔ Blinding (removes experimenter bias)
✔ Controlling confounders (ensures the effect is due to the treatment)
🌍 External Validity =
“Can we apply this to real life?”
High external validity = The study applies to a wide range of patients.
Low external validity = The study only works for a specific group.
📌 Example:
A new hypertension drug is tested only in young, healthy males → Low external validity (can’t generalize to all patients).
A study includes a diverse population (age, gender, comorbidities) → High external validity (results apply to real-world patients).
🚨 Contrast with internal validity:
A diverse study may have higher external validity but slightly lower internal validity due to more confounders.
A tightly controlled study (healthy males only) has high internal validity but low external validity.