About the Exit Poll

After the discrepancies in exit poll results in the 2000 and 2004 Presidential elections, controversies surrounded exit polling methodology – specifically that of poor sampling techniques, inaccurate results and skewed reporting of underrepresented subgroups, such as African American and Latino voters.

To date, exit poll researchers have used random sampling techniques that select too few racially homogenous precincts, such as those studied by Edison-Mitofsky (1989), or studies that interview a majority of mixed-race precincts, as was conducted by the Los Angeles Times (Barreto, Guerra, Marks, Nuno, & Woods, 2006).

To address these methodological issues, in 2005 a team of LMU researchers implemented the use of an innovative sampling technique known as the “racially stratified homogenous precinct approach” to accurately sample voters in the city of Los Angeles. As explained in their article, Controversies in Exit Polling: Implementing a Racially Stratified Homogenous Precinct Approach, this methodology theorizes that polling data is more accurate when it targets subgroup enclaves, as opposed to singling out individuals from these subgroups who live in more integrated communities. When compared to other exit polling methods, this methodology has been deemed to produce accurate results that are indicative of the desired sampling population, regardless of diverse populations.

This research has contributed to new evidence on behaviors, preferences and needs of ethnic minorities, and barriers to voting in Los Angeles (Barreto, Cohen-Marks, Woods, 2009). As a result of this evidence, the Center for the Study of Los Angeles established this signature study and now has a definitive record of how Los Angeles votes.

For more information about the previous research or questions about the upcoming exit poll, please contact Brianne Gilbert, the Center’s Associate Director, at brianne.gilbert@lmu.edu.

Research. Action. Justice.