Home › Forums › General Help › reviewer questions
This topic contains 1 reply, has 2 voices, and was last updated by Joana 2 months, 2 weeks ago.

AuthorPosts

veroniqueDear Mite et collegues,
I have submitted a manuscript in which I am using EEGBRAPH. Good news, it should be accepted but I have 2 questions of one of the reviewer: Question 1: “Please provide the calculated effect size, raw pvalues and corrected pvalues for all the group comparisons.”
What can I answer ?? Note that if I can I do not want to add all of these data….. !! 
Question 2 ” Although the authors claimed that they used the FDRcorrection to control the falsepositive significant differences, there are no details provided in the Material and Methods. For example, how the authors performed the FDRcorrection for the nodal level indexes? Did the authors considered the number of frequency band and the number of brain regions (ROIs) at the same time?”
In the BRAPH helpforum, I found this “In BRAPH, to calculate the false discovery rate (corrected by using the BenjaminiHochberg procedure, http://braph.org/manual/braingraphs/) the pvalues are ranked in ascending order and compared with their false corrected values. Once the largest pvalue that is smaller than the corresponding falseratecorrected value is identified, all the pvalues smaller than this value are considered significant. These multiple comparisons are performed at a particular density for a given measure across all of the regions in the brain atlas.” Can I used it like that exactly ??
Hope you will be able to help me,
Dear Veronique,
For the effect sizes, you can check any statistical website on how to calculate them.
What do you mean exactly with raw or corrected p values?
Braph will provide you the raw p values, if you want to correct them for multiple comparisons you can use FDR, Bonferroni or any other test. Note that the global graph measures (if calculated across different densities/thresholds) are not normally corrected for multiple comparisons.
Braph corrects the nodal measures by FDR across all nodes. You can use the explanation given in the manual.
Best wishes,
Joana
 Question 1: “Please provide the calculated effect size, raw pvalues and corrected pvalues for all the group comparisons.”

AuthorPosts