Errrrnttt. Nice try. You are correct that the article is committing a statistical error, but so are you. Technical support calls range the gamut from questions, software problems, hardware problems, user errors, help on setup and installation, etc. And you can't lump all the nonhardware issues into "problems strictly with the software" per se because a question could be as simple as "how do I install this" but could be more intricate like asking value added questions about how to best set up wifi or what settings are best to get maximize battery life. And some of those calls might fall outside the realm of the phone like "sorry the problem is not in your phone but your router/printer/computer/etc."
You also have to factor in user fatigue. If your first problem is a hardware problem where the phone doesn't even work, you're less likely to keep calling back about other problems because you just get tired of dealing with it.
Also the only scrap of information we have here from the article is that they separated out hardware calls from everything else. We have absolutely NO idea what those other issues are. They state "problems" but problems aren't the only reason to call technical support and by their own paramaters of the study they didn't appear to really look at those other calls to state they were actual problems.
Given the data in the article, the statement that 14% of Android phone calls are on hardware issues is meaningless. The statement that percentage wise, that 92% of iPhone calls are software problems is also meaningless. So you have an article and a comment, both meaningless. More numbers need to be revealed to make this meaningful.