Hello, 

I have two questions re the t_thr column: 

1. I'm filling out the sdm table and am struggling filling out the t_thr column. It seems that very few papers (at least the ones that I'm looking at) report the t value they used to threshold their statistical maps. I have noted the various p value the studies used to threshold, and entered these values into the converter on the sdm website (https://www.sdmproject.com/utilities/?sh...). However, the results I get seem odd: e.g. One-sample p to t converter: p = 0.001, n=20, t= .-3.5794

Is this correct? All the examples I see on the website are around 3.1. Is the p value input supposed to be thresholded or unthresholded? 

2. Is it worth the hassle of finding the individual t value for each study, or is it better to allow the automatic conservative sdm approach, p=.0001? If this is better, or necessary in some cases where I cannot get the t value, what do I input into the sdm table? 

Best wishes, 
Molly

Dear Molly Rowlands,

I have encountered the same question as you about how to filling out the t_thr column in the case that most studies only report the threshold of p value. Have you found any solution? If so, would you like to share your solution? thank you very much.

Best wishes,
Liangfang

Hi Molly and all,

At our paper in JoVE you'll find the guidelines of which t_thr to use for studies not reporting the value used.

Best,

Anton

https://www.jove.com/t/59841

Hi everyone,

I convert the p values with the converter, but same as Molly I got negative t_thresholds is this ok? Should we ignore the symbol and add just the value? Aditionally, in some papers (in the text) I found a p value for threshold but then in the table the authors report different threshold for some of the coordinates, what should i do then? Sorry but is not so clear for me in the Jove paper, can you please help me with this.

Thanks

Sol

Dear Sol, 

If you covert the pvalue to obtain the t_thr in the sdm table, you should ignore the symbol as that variable needs the absolute value and SDM will perform the upper and lower limits. If you are not sure about the threshold you should use, then, you should go to the results coordinate files for each study and use a slightly smaller value than the t-value of the smallest peak.

Best, 
Lydia