Okay I see what your getting at. But even from this article it doesn't seem like the people who did the studies were trying to claim that the gay population grew. I got on the Williams Institute and it is even more clear that that is not what they are claiming on it, and that these are different studies. Heck, the study that showed 3.5% of the population would be in the LGBT demographic was a Gallup poll and wasn't directly related to the Williams institute at all. A writer for the William institute were simply commenting on the study and never even tried to compare it to their own 2011 survey. But I can understand how the author of the article in the opening post had that misconception.The problem comes when you change what it is you are looking at if one group was not added the first time and then added the second time it makes it look like the numbers have changed when if fact they never really did. It changes the parameters of the survey and gives out false numbers. It makes it look like the gay population grew in proportion when in fact it didn't.
Here is a little bit from the Williams institute's analysis of the Gallup poll that I found amusing; "larger portion of lower educated Americans identify as LGBT compared to college graduates". [Link]