What are the response rates in national polls?

[Nova Andromeda]

--Can anyone provide me with data on what the response rates for national polls are? I can't seem to find any hard data on this subject. I'm guessing it varies quite a bit depending on the poll and the organization running the poll. However, I must be blind since I can seem to find it in major polling websites like Gallop on any useful subject.
-This bothers me because if the response rate is 50% (for instance) then one could argue that a poll showing 100% support for X doesn't show the "real" 50% public support because only the supporters or X answered the pollers.

[Ariphaos]

You'd be better looking up polling methods, though the real question is what kind of poll you are talking about.

I'm not sure if the specific data gets published for all polls. I do know that this type of thing occurs and it's an extreme hassle, especially when looking for white males to answer polls and a woman or minority answers the phone. Polling agencies will try to get an even demographic in order to ensure that they have selected an appropriately random sample.

It's been a long time since I had to study this, but if I recall correctly the response rate was usually well under 50%, due to busy signals, no answer, and cases where the answering party gets offended, decides to troll the poll and so on.

I would certainly question the polling methods of anyone who got 100% of a poll result for anything even remotely controversial. I'd expect 'yes' answers to "Would you kill yourself if we asked you to?" polls.

In fact I think that'd be a great way to measure the trolling rate :-p

[Nova Andromeda]

--I'm not asking about any one poll specifically, but national polls in general. I have looked for polling methods and most polls are normalized for obvious things like gender and race, but the only thing that counts in the end is that the response rate doesn't vary with respect to one's position regarding the poll questions. If it does the poll will be off. While normalizing for race, gender, etc. may help with this problem it may or may not help very much.
-It seems like the response rate should be reported along with the results of any poll, but I can't seem to find them....
-One should note, however, that polls before major elections are quite acurate which lends some credebility to the polling methods being used.

[Ariphaos]

I think that's frequently identifiable by known false responses. Neopagan religions being the main example, I think - where they receive a very small result in the polls, but relevant data suggests that they are in fact more numerous. In such cases, effects on response rate would be similar to such false responses, and it will almost certainly be in favor of the perceived minority.