Good afternoon Alfonso,
I hope this message finds you well. I wanted to follow-up with updates regarding my queries from last week.
1) Centering: We have attempted to analyse our data (total n=87) with and without the "centering" option, and while we find some general similarities, we have noted that the "centered" data produces vastly more significant clusters (albeit smaller), and slightly different peaks, when compared to the non-centered analysis. Considering this result, would you be so kind as to provide any insight as to which option (centered or not) would be best to proceed with?
2) Number of factors to retain: We are assuming that MVPA principles resemble those of PCA, i.e. that certain factors will have a stronger capacity of explaining variability. Is there a direct way to evaluate this for each factor? If no metric is available, is there any "eye test" we could perform ?
Thank you in advance for your expertise on the matter, it is greatly appreciated.
Warmest Regards,
Maria
I hope this message finds you well. I wanted to follow-up with updates regarding my queries from last week.
1) Centering: We have attempted to analyse our data (total n=87) with and without the "centering" option, and while we find some general similarities, we have noted that the "centered" data produces vastly more significant clusters (albeit smaller), and slightly different peaks, when compared to the non-centered analysis. Considering this result, would you be so kind as to provide any insight as to which option (centered or not) would be best to proceed with?
2) Number of factors to retain: We are assuming that MVPA principles resemble those of PCA, i.e. that certain factors will have a stronger capacity of explaining variability. Is there a direct way to evaluate this for each factor? If no metric is available, is there any "eye test" we could perform ?
Thank you in advance for your expertise on the matter, it is greatly appreciated.
Warmest Regards,
Maria
