Boston Area Colleges Election Project

Discusses the networks' move to jointly hire the same exit polls for election projections.

Research of 1998 democratic primary in New York City shows that the having his/her name first on a list helps a candidate do better

This study examines 2000 polling data and discovers that for the 2000 polls, placing the horse race question later in the survey (after "warm-up" questions) makes it less likely for voters to respond

From abstract: Methodological details are presented of a survey research technique known as exit or election day polling.

Discusses the Utah Colleges Exit Poll. Is a good overview of a student run exit poll, its history, and uses. Also reviews exit polling in general, providing a list of key terms.

Investigates the reason for the difference between the exit poll estimates and the actual vote count in the 2004 Presidential election.

Preelection polls for the 1989 VA gov. election overestimated the vote share of the black candidate. This was due to social desirabilty and race-of-interviewer effects

Looks at the New Hamphsire polling gaffe and samples various pundits explanations, which include the Bradley effect, likelihood of voting estimates, greater possibilty of voters changing minds just be

Rebuts Wrights 1990 Legislative Quarterly Article. Wright's study does not explain why respondents overreport only their Senate vote and not presidential or House votes.

There is a proincumbent response bias in the 1978 and 1980 NES.