# P-values

## Citation, DOI & case data

**Citation:**

**Permalink:**

**rID:**

**Revisions:**

**Published:**

**System:**

**Sections:**

**Tags:**

**Institution:**

**Inclusion in quiz mode:**

**From the case:**P-values

Loading images...

## Loading Stack -

0 images remaining

This brief visual tutorial is intended to provide an intuitive understanding of P-values. Make sure to cement your knowledge of this difficult material by answering the 8 quiz questions!

## 8 case questions available

**Q: **1. Investigators plan a clinical trial to assess the ability of a new drug to lower blood pressure compared to an old drug. What is the null hypothesis?
A. new drug efficacy>old drug efficacy
B. new drug efficacy<old drug efficacy
C. new drug efficacy=old drug efficacy
D. new drug efficacy≠old drug efficacy
show answer

**A: **C. New drug efficacy=Old drug efficacy

**Q: **2. A p-value is not the probability that the null hypothesis is true because…
A. ...the p-value is calculated assuming that the null hypothesis is true.
B. ...the p-value is calculated assuming that the null hypothesis is false.
show answer

**A: **A. ...the p-value is calculated assuming that the null hypothesis is true.

**Q: **3. A p-value is the probability that the observed results were produced by random chance alone.
A. True
B. False
show answer

**A: **B. False

**Q: **4. The lower the p-value, the larger the effect size.
A. True
B. False
show answer

**A: **B. False

**Q: **5. The lower the p-value, the more incompatible trial results are with the null hypothesis.
A. True
B. False
show answer

**A: **A. True

**Q: **6. A statistically significant result (p<.05) may not be clinically significant.
A. True.
B. False
show answer

**A: **A. True

**Q: **7. A trial comparing a new to an old antihypertensive report that the new drug results in a 20 mmHg more blood pressure decrease than the old drug with a p-value of .02. Assuming that the null hypothesis is true and in the absence of bias, what is the best interpretation of this result?
A. There is a 2% chance of observing a blood pressure 20 mmHg lower in the new drug group than the old drug group.
B. There is a 2% chance of observing a blood pressure at least 20 mmHg lower in the new drug group than the old drug group.
C. There is a 2% chance of observing a blood pressure 20 mmHg lower or higher in the new drug group than the old drug group.
D. There is a 2% chance of observing a blood pressure at least 20 mmHg or more, lower or higher in the new drug group than the old drug group.
show answer

**A: **D. There is a 2% chance of observing a blood pressure at least 20 mmHg or more, lower or higher in the new drug group than the old drug group.

**Q: **8. A trial comparing a new to an old weight loss program reports that the new program results in 1 ounce more weight loss than the old program with a p-value of .01. What is the best interpretation of this result?
A. Because the weight loss difference was 1 ounce, the result is likely due to chance.
B. Because the p-value is .01, the result is likely due to chance.
C. Because the weight loss difference was 1 ounce, the result is not likely to be clinically significant.
D. Because the p-value is .01, the result is not likely to be clinically significant.
show answer

**A: **C. Because the weight loss difference was 1 ounce, the result is not likely to be clinically significant.

## References

- 1. Wasserstein R & Lazar N. The ASA Statement on P-Values: Context, Process, and Purpose. The American Statistician. 2016;70(2):129-33. doi:10.1080/00031305.2016.1154108
- 2. Tigges S. To P or Not to P: A P-Value Comic. Journal of the American College of Radiology. 2021;18(8):1163-75. doi:10.1016/j.jacr.2021.03.014 - Pubmed
- Browner W. Are All Significant P Values Created Equal? JAMA. 1987;257(18):2459. doi:10.1001/jama.1987.03390180077027

## 1 public playlist includes this case

## Related Radiopaedia articles

## Promoted articles (advertising)

### How to use cases

You can use Radiopaedia cases in a variety of ways to help you learn and teach.

- Add cases to playlists
- Share cases with the diagnosis hidden
- Use images in presentations
- Use them in multiple choice question

Creating your own cases is easy.

ADVERTISEMENT: Supporters see fewer/no ads