My main problem with the data is that, well, it doesn't say what it purports to say.
This is a survey taken among users of a specific app-development package that doesn't support WP7, as you say, that asks how interested they are in developing for WP7. It doesn't say anything about WP7 popularity or lack thereof. It's useful, if at all, for AppCelerator to determine how much of a market they have to adapt their product for WP7, not about developers in the wild.
What MJFoley is doing here is taking an existing but irrelevant data point and twisting it to create a headline: "Microsoft a distant third horse in the mobile race", even though the race itself is an irrelevant internal AppCelerator race.
I've spent the last two years as a tech journalist, so I know exactly what went on in their heads. In fact, I reported on that self-same survey a few months back, and took care to stress that these results are NOT indicative of WP7's success. Or lack thereof.