If it would be easy, then perhaps you should do it. Such a conclusion would have a nice amount of academic value.
I imagine there are a large number of small and/or untraded businesses with dishonest or illegal practices. Perhaps data would even show that larger market cap indicates corruption is less likely, depending on your measurement for corruption?
