Boston Area Colleges Election Project

Exit polls had more problems than pre-election polls in 2004.

The NES is victim to a substantial amount of pro-winner bias in subpresidential elections.

Concludes that the 1989 Virginia gubernatorial race exit poll was wrong because it used face-to-face interviews.

1988 NES Senate Election Study has systematic respondent overstatement of support for winners.

Rebuts Gronke's criticisms in a 1990 article in Legislative Studies Quarterly, except he does not rebut Gronke's criticisms about exit polling

The race of an interviewer has an effect on the response, depending on the race of the respondent and the type of question asked.

Issue labeling and enhancement can have a substantial effect on public support for some issues. In other words, question wording and word choice have a directional effect.

Voter turnout has actually not been in decline. This misperception exists because ineligible voter population has increased more quickly than voter population.

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

Current predominant ecological inference methods used in voting rights litigation, especially the two predominant methods, have shortcomings.