Need a hand? All the help you want just a few clicks away
Dear allI have some questions about MFA analysis.1. Is MFA possible only when the number of tables is more than ‘3’? When I tried to process with ‘2’ tables, it didn’t work, for the default value is always set as ‘3’. How do you analyze with ‘2’ tables?2. Is the analysis automatically corrected even though scale range or types of data are different between tables? For example, one table is the result of 9-point scale and another table is the result of 6-point scale, or the result of frequency data, then can I still use the analysis? What I know is that the ‘Mixed option’ is for the case where qualitative data and quantitative data are mixed. In case of the example that I mentioned above, should I analyze it as quantitative data without any selecting other options?3. Is there any method of presenting p-values for RV coefficients? How you do validate the significance of the RV coefficient values?Thanks in advanced :-)
1. It is definitely possible to run MFA’s on two tables. Try it out :)
2. Yep, all the variables are rescaled to the same scale, and every table has the same weight in the analysis. Our “mixed” option means that you have tables of quantitative data and tables of qualitative data. XLSTAT does not support tables with different data types within the same table.
3. You may obtain RV coefficients p-values by applying our RV coefficient feature on the PCA and / or MCA observation coordinates: https://help.xlstat.com/customer/en/portal/articles/2062458
Hi, I'm trying to perform a pCA analysis of a multivariate database. I received an error as follows: The selection is not valid. A multiple selection can be by rows only if the Transposition option is activated.Please advise.Annie
If you wish to launch a PCA using rows as variables, just activate the transposed mode by clicking on the blue square arrow at the bottom left of the dialog box.
XLSTAT first steps and statistical learning resources
The place where you will master all of XLSTAT features
Any feature or option you’d like us to develop?
For those of you who prefer video help