Validity vs Reliability
How to tell Validity and Reliability apart when a board question makes them look alike.
Validity asks whether the tool measures the intended truth correctly; reliability asks whether repeated measurements are consistent and reproducible even if biased.
Side by side
Validity means a test measures what it is intended to measure and is accurate relative to truth.
Reliability means repeated measurements are consistent and reproducible even if they are not accurate.
Validity means a test measures what it is intended to measure and is accurate relative to truth.
Reliability means repeated measurements are consistent and reproducible even if they are not accurate.
Accuracy: agreement with a gold standard or true value; reduced by systematic bias.
Precision: repeat or inter-rater measurements agree; reduced by random error; can be reliable yet biased.
The repair move
Before naming the answer, force the stem through the separator: Validity asks whether the tool measures the intended truth correctly; reliability asks whether repeated measurements are consistent and reproducible even if biased.
Try a free question
CoreA new home blood pressure monitor consistently reads 8 mmHg higher than the true blood pressure (verified against a calibrated mercury sphygmomanometer). When measurements are repeated, they consistently cluster within 2 mmHg of each other (but 8 mmHg above truth). Which characteristic best describes this device?
Show answer and explanation
Answer: A. Reliable (precise) but NOT valid (accurate). Validity = accuracy = how close to the TRUE value. Reliability = precision = how close repeated measurements are to EACH OTHER. The home BP monitor described is RELIABLE (reproducible) but NOT VALID (systematic bias of +8 mmHg). Calibration would restore validity without affecting reliability. Distinguish bias (systematic error) from random error (precision).
Why the other choices are tempting
- B. Valid but not reliable would be: accurate on average but scattered widely.
- C. Both valid and reliable would be: accurate AND tightly clustered.
- D. Neither would be: inaccurate AND widely scattered.
- E. Both inaccurate and imprecise is the same as 'neither'.
Flashcards
Tap a card to flip it.
What clue supports Validity?
Choose Validity when the stem gives this pattern: Validity means a test measures what it is intended to measure and is accurate relative to truth.
What clue supports Reliability?
Choose Reliability when the stem gives this contrasting pattern: Reliability means repeated measurements are consistent and reproducible even if they are not accurate.
Quick answers
What is the fastest way to tell Validity from Reliability?
Validity asks whether the tool measures the intended truth correctly; reliability asks whether repeated measurements are consistent and reproducible even if biased.
What is the key clue for Validity?
Validity means a test measures what it is intended to measure and is accurate relative to truth.
What is the key clue for Reliability?
Reliability means repeated measurements are consistent and reproducible even if they are not accurate.