We’ve often noted here that when partisans (of both sides) see a poll they don’t like about their candidate, they sometimes will talk about how the methodology of the poll was wrong. And usually the partisans don’t see the same problem with polls that favor their candidates.
The issue of methodology and whether political preferences may influence how polling is used burst upon the Internet recently with some questions raised about the site Real Clear Politics (which we link to often here on TMV).
The issue of methodology, polling and perceived biases is an interesting one. And here is a MUST-READ discussion of it.