PULLMAN, Wash. – During your favorite TV show, an ad comes on for an athletic drink featuring a white athlete. During the next commercial break, an almost identical ad is shown, but it features a black athlete. Does this difference have an effect on the audience? This is a question one WSU professor is trying to answer.
Ioannis Kareklas, an assistant professor in the Department of Marketing, began research into the effects of racial matching in advertising as part of his dissertation at the University of Connecticut. At WSU he is delving deeper into how time, region, age and cultural norms can change a person’s preference for the white or black athlete.
41 years of analysis
Kareklas began with a meta-analysis. A meta-analysis, he explained, is when a researcher collects all of the previous data collected on the subject, organizes it according to certain factors and then looks for patterns over time.
The first pattern Kareklas noted was the increase in acceptance of blacks in advertising. Research into racial matching effects began in 1969, a year after the Civil Rights Act of 1968 was signed, when there was a sudden pressure to include black people in advertisements. Since 1969, black people have begun to show more and more preference for seeing their own race in advertisements, he said.
This parallels the increase he found in white preference as well. Since 1970, white subjects have shown an increasing preference for white models in ads.
Gender, age and region
After simply looking at results over time, Kareklas began coding for other factors. When he began looking at gender and age differences, he also found patterns.
“I’ve found that black males prefer black models in ads more often than black females do,” Kareklas said. “I’ve also found that young black people have a larger preference for black models in ads than older black individuals.”
He also found that, in regions like the Midwest and the West, the black community’s preference for ads featuring black models was greater.
“Perhaps in areas with a large population of blacks, there is less of a preference,” Kareklas said. “And in areas where there is a smaller population, there is a greater preference.”
Cultural events
He organized his data according to significant cultural events for the black community, like the trial of Bobby Seale and the election of President Barack Obama, and found more variations.
“When there is an event that is positive for the black community, black subjects want to see more ads featuring black models,” Kareklas explained. “When there is a negative event, much less so.”
Throughout his analysis, Kareklas found that subjects generally preferred to see their own race in advertisements. This seems like the obvious conclusion. But then he investigated color preference and found the answer wasn’t so simple.
Is racial preference simply black and white?
During his research, Kareklas became interested in how an individual’s color preference affected racial preference. Kareklas theorized that a subject’s preference for the color black or white would coincide with which race was preferred.
He began conducting implicit association tests, a technique developed by Dr. Anthony Greenwald at the University of Washington, which allow a researcher to measure subjects’ attitudes without them knowing. This helps to eliminate bias. No person is going to admit they prefer white or black people in ads for fear of being labeled a racist. IATs help eliminate this.
During IATs, subjects are asked to classify images or words presented to them on a computer as pleasant or unpleasant by pressing keys on the keyboard. Subjects are supposed to finish the test as quickly as they can, which helps to eliminate bias.
This test is based on the idea that individuals will associate positive things with other positive things quicker in their minds than they associate two negative things. If subjects perform the test quickly, they won’t have time to consider what is socially right or wrong and they will answer truthfully.
Kareklas performed multiple IATs, testing for color preference, race preference and advertisement preference. His results showed that color preference predicted racial preference, even in subjects who preferred the color opposite of their race.
He theorizes this link between color preference and racial preference has to do with the cultural meaning behind the two colors. In Western culture white is considered a good, pure color while black is associated with evil and darkness. This connotation leads subjects to prefer white things, including people. 
Effect on advertising
Kareklas continues to explore the link between color preference and racial preference and has found that controlling for color preference yields results similar to his original studies. Black subjects will prefer black models and white subjects will prefer white when color preference is eliminated.
As for other aspects of his research, Kareklas’ feels they could help businesses market to certain demographics.
“Now geographic regions, different ages and genders can be more effectively marketed to because we know their (racial) preference,” he said.