Since 1982 when long time Los Angeles Mayor, Tom Bradley, An African-American lost the California Governorship race to Republican challenger, George Deukmejian, a White, the term Bradley Effect has remained in the political lexicon of America. It is believed to be a phenomenon whereby white voters lie to pollsters about their voting preference for fear of accusations of racial bias. In that election, Tom Bradley was way ahead in polls until the election day. He was even projected as the winner after the close of voting on election day according to exit polls but ultimately lost the election after all votes were tallied.
In 2008 when President Barack Obama challenged Senator John McCain, the term was all over the place in the media. There were analyses that for President Obama to win the elections according to the polls, he needed to be ahead with considerable margin to make up for any Bradley Effect. Well, that election is now history and President Obama defeated McCain by an Electoral College landslide and by approximately 6.5 percent in the popular vote.
Even while the term Bradley Effect got a lot of mention in the 2008 Presidential election as a possible factor, there were divergent views as to whether Tom Bradley really lost his own election due to racial bias. After the 2008 elections, talk of Bradley Effect has since died down considerably. The consensus seems to be that the effect was not a factor in the 2008 election after-all, and polling data was used an evidence.
Just like every other topic under the sun, there are not just two sides but several sides to this Bradley Effect argument. There are hard questions to be answered if these analyses are to be taken seriously. For example, Rasmussen Reports and Pew Research polls which had Obama leading McCain by 6 points before Election Day where hailed as the most accurate polls of the 2008 Presidential election because they were off by mere 0.5 points. These polls were used as part of the arguments that there was no Bradley Effect.
What if there was really a Bradley Effect in 2008 and it was captured by Reuters/C-SPAN/Zogby poll which had Obama leading McCain by 11 points? The pollster was adjudged as the most inaccurate because its final poll was off the mark by 4.5 points, but what if it was really the most accurate and it was of the mark because it captured the so-called Bradley Effect? Perhaps, each election should be judged on it own merit.
If pundits are really looking for “effects” to pin election results to, have they thought about a possible Obama Effect in 2012? It could be Bradley Effect in another form. It is also quite possible that a lot of voters are lying to pollsters about their preference in this election to President Obama’s disadvantage. Voters who overwhelmingly chose Obama in 2008, may be coy to tell pollsters they are not voting him this year, lest they are called wimps or gullible. They may be secretly planning to vote for Mitt Romney in large numbers but have been telling pollsters they will either vote Obama or they are undecided. So, if Obama continues to lead Romney until Election Day, watch out for talk about possible Obama Effect.
Better still, different elections should be judged by their own merit instead of looking for effects to hold responsible. No two elections are the factors entirely parallel and there can never be a consensus about the reasons for the results of a particular one.