This is one area where I'd be skeptical of "data" because I think there are too many variables to separate any signal out of the noise. First, there are not all that many startups to begin with, and they all do different things. Whether it's one founder or many seems way down on the list of factors influencing the success. If you had thousands of startups entering a single market within a short time span then maybe you could get something, but not in the current world.

Furthermore, I think as a founder if you concern yourself much over this point you are likely to fail. Building a startup is a game of applying extremely limited resources, if you are focused on structural details like number of founders then you are necessarily not focused on the unique metrics of success for your business. Of course there are pros and cons to being solo or multiple founders, but in the end a motivated individual should never let conventional wisdom get in the way of pushing forward (especially from investors who are grasping for any kind of signal to hang their hat on!). Conversely, if a person with the right talents and skills comes along you should absolutely allow them a co-founder seat even if you planned on going it alone as this will magnify your sphere of opportunity in terms of execution potential.

My understanding is that there is independent research and some of it has been done by YC.

See PG's comment here: https://news.ycombinator.com/item?id=77319

Keep in mind that, in general, any meaningful research showing differential rates of return will probably be proprietary because of the potential business advantage it confers.