I would be interested to know if they controlled for the gender of the investors

Controlling this might not necessarily be desirable. One of the problem with randomized studies is that it's difficult to show they generalize well. Given investing does not take place in a randomized environment, that investors were representative of real investors is far more important than they be controlled for.