Bonferroni correction

#1
Hi! This is the first time for me approaching Bonferroni correction for multiple comparisons and I have a doubt about that. I am going to run bivariate correlation to see association between 5 scales of a questionnaire and 6 scales of an interview (each scale of the interview vs each scale of the questionnaire). I would like to apply Bonferroni correction for multiple comparison but I am not sure how to calculate the number of comparison to correct p. Hope I was clear and thank you in advance to those who will help me!
 
#3
Thank you for the answer. I would like to be sure I have calculated the number of comparisons well. I need to test correlations between 6 scales of a questionnaire and 5 scales of an interview. So, reducing p 0.05 to 0.05/30 (6x5) is right?
 

hlsmith

Less is more. Stay pure. Stay poor.
#4
I don't know what you mean by scale. Please describe the context in detail and/or provide a snippet of data.

Thank you.
 
#5
I need to correlate data of an interview made of 5 scores (communication, motor abilities, daily life, social abilities and total score) with scores from a questionnaire made of 6 domains (anxiety, depression and so on). So, correlations will be run between communication and anxiety, communication and depression [...], motor abilities and anxiety, motor abilities and depression and so on. Lot of comparisons and I need to correct for multiplicity. I am not sure my calculation of the number of comparison (6x5) is right. I know it is a stupid doubt, but I need to be sure the calculation is correct. Thank you in advance.
 

Karabiner

TS Contributor
#6
Personally, I perform Bonferroni corrections very rarely. If there are only few comparisons to be made,
then usually in my work there is one primary comparison, and some secondary comparisons, and I do
not mind making the secondary comparisons without Bonferroni (except a reviewer would demand it).
If there are many comparisons, Bonferroni destroys power and would therefore lead to a meaningless
analysis. But there certainly are circumstances, e.g. with a very large sample and/or the absolute priority
to prevent false-positive results, where Bonferroni (or preferably Bonferroni-Holm) can be justified.

I do not know whether you'd be better off with multiple/multivariate approaches such as multiple regression.
If you need to make 30 distinct comparisons, and all null hypotheses are true, then the probability of at least
one false-positive finding is about 80%. Instead of Bonferroni (0.05/30=0.0017) I would be inclined to choose
a conservative, but not so crazy significance level, such as 0.01. If all 30 null hypotheses were true, the chance
of at least one false-positve result would be 26% then. But if some null hypotheses were not true, you would
retain enough power with an alpha of 0.01.

Just my 2pence

Karabiner
 
Last edited:
#7
Thank you so much! I agree with you: such a correction should not be applied to my data, cause it is a merely exploratory analysis. Unfortunately, one co-author asked for that, so this attempt should be done. Thank you for your reply
 

Karabiner

TS Contributor
#8
Your co-author is not open to consider the loss of power with p=0,0017? That is sad.
As mentioned above, Bonferroni-Holm could be used, which is a little bit less conservative.

With kind regards

Karabiner