Is it methodologically correct to, in some specific situations, merge related binary variables into one continuous? I have 10 yes/no binary variables regarding different types of economic risks associated with running a startup. I have a database of around 100 different startup companies evaluated on each of those variables (whether a particular risk occurred). Is it methodologically correct to merge all those binary variables into one continuous variable, as to assess the overall economic risk experienced by each of the startups? If yes, do you know any literature to which I could refer? 
 A: Fuzzy sets are useful for this sort of compositing, especially if you need to weight the variables in a non-uniform way. The techniques are applicable to Boolean variables too, although I've only used them to composite continuous columns myself. There is a learning curve involved, but it's a useful addition to one's toolbelt. My favorite go-to resource is Klir, George J. and Yuan, Bo, 1995, Fuzzy Sets and Fuzzy Logic: Theory and Applications. Prentice Hall: Upper Saddle River, N.J. You can get a used copy for a few bucks if you hunt around. If you use SQL, parts of my series of self-tutorials on fuzzy sets might be of help (although I don't directly discuss compositing). 
One "gotcha" I ran across when combining continuous columns like this was forgetting to make sure that all of the indicators point in the same direction; for example those that measure some undesirable property should all have the same lower and upper bounds, while those that have desirable ones should be reversed. One simple method for reorienting variables on a 0-1 scale is obviously to subtract them from 1, like a distance measure. I hope that helps.
