So the answer is, "I feel like..." I get it, but when talking about macro policies like this, it's an awfully flimsy way to build an argument. I get that there are problems with data sets, but the answer has to revolve around trying to solve those problems, not just disregard all data.
The Heritage data I posted, above, about voter fraud might be my favorite one. Yes, there is voter fraud, but looking at that very consrvative organization's attempt to document it, and you'll see that it is an a near infinitessimal level (and it gets prosecuted).
What if the reason you think there is a bigger immigration problem now than in 2008 is because it's talked about so much more now than it was then? It's like the meme about Gen-X-ers being a lot more worried about quick sand than it turns out we should have been. If it is talked about a lot, it appears to be a bigger problem. But appearances can be deceiving. We all know that. So we turn to data. And yes, it's fair to point out the inadequacies in the data, but there has to be something more than just, "I feel like..."