# Thread: Calculating standard deviation from mean and P value

1. ## Calculating standard deviation from mean and P value

I am doing a systematic review and one of the papers I am including has reported the data only as mean, sample size and P value. I need the mean and standard deviation in order to do a meta-analysis. Is it possible to calculate the SD?

2. ## Re: Calculating standard deviation from mean and P value

Originally Posted by CMP
I am doing a systematic review and one of the papers I am including has reported the data only as mean, sample size and P value. I need the mean and standard deviation in order to do a meta-analysis. Is it possible to calculate the SD?
As long as you have the data points..of course you can figure out standard deviation. 1. calculate how much each data points deviates from the mean, then square each of those numbers, then add all the squared numbers, then divide SS by n-1, and then take the square root of that.

3. ## Re: Calculating standard deviation from mean and P value

What test was the p-value for?

4. ## The Following User Says Thank You to Dason For This Useful Post:

hlsmith (06-22-2016)

5. ## Re: Calculating standard deviation from mean and P value

I don't have all the data points. The authors reported the mean and p value only. The p value was from a Mann-Whitney test.

6. ## Re: Calculating standard deviation from mean and P value

I re-read the paper and realized I was mistaken about the p-value being based in the Mann-Whitney. It was a Student t-test.

7. ## Re: Calculating standard deviation from mean and P value

What was the t-test testing?

8. ## Re: Calculating standard deviation from mean and P value

It was comparing the difference in the mean pain scores of one group of patients receiving one treatment with another group of patients receiving another treatment.

9. ## Re: Calculating standard deviation from mean and P value

Was the t-test assuming equal variances?

10. ## Re: Calculating standard deviation from mean and P value

Originally Posted by Dason
Was the t-test assuming equal variances?
I can't tell for sure but I would say yes, equal variances.

11. ## Re: Calculating standard deviation from mean and P value

Well, my suggestion to this problem is straight-forward. Assuming this case is a two-independent samples t-test, with equal variances, and the total sample size (N) is known, then just determine the associated t-statistic associated with the computed p-value (df = N-2). As such, it would become straight-forward to the find the standard deviation of the difference between the two means (i.e., the standard error (Se) of the statistic) by dividing:

Se = (Mean-diff between groups) / t

This is my interpretation of what the OP is asking.

12. ## The Following User Says Thank You to Dragan For This Useful Post:

CMP (06-24-2016)

13. ## Re: Calculating standard deviation from mean and P value

That's awesome! Thanks!

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts