Election polling

I am more aware in 2008 than I have ever been before of how easy it is to find the results of several polls compared, with political analysts talking about why they are giving different results. This is a good opportunity to read about how statistics is used.

• One analyst's description of the eight different national polls.
  http://www.fivethirtyeight.com/2008/10/tracking-poll-primer.html
  
  
• What's the difference between results from the population of registered voters and the population of "likely voters"?
  http://www.fivethirtyeight.com/2008/10/some-likely-voter-models-are-suspect.html
  
  
• If one "poll house" has a higher percentage for the candidate than other poll houses in one week, are they more likely to have a higher percentage for that candidate the next week?
  http://www.pollster.com/blogs/tracking_poll_house_effects.php

• Most surveys are taken by phone. Do they include cell-phones? What difference can that make?
  http://www.fivethirtyeight.com/2008/09/estimating-cellphone-effect-22-points.html
  
  
• What should the pollsters and reporters be doing to help us better understand why the polls give such different results?
  http://www.pollster.com/blogs/miller_what_pollsters_can_lear.php