Implicit Relationship Evaluations

A post by Tony Greenwald at Scientific American:

http://www.scientificamerican.com/article.cfm?id=i-love-him-i-love-him-not

Implicit race attitudes predicted vote in the 2008 U.S. presidential election

An article by Project Implicit researchers published this month in Analyses of Social Issues and Public Policy reports evidence that both implicit and explicit race attitudes were related to intended vote in the 2008 U.S. presidential election. 1,057 registered voters completed a study conducted at Project Implicit’s research website during the week before the presidential election. The participants completed multiple measures of racial attitudes including self-reported feelings of warmth toward Blacks and Whites, a measure of “symbolic” racism, two implicit measures of racial attitudes – a Brief version of the Implicit Association Test (IAT) and the Affective Misattribution Procedure (AMP), and reported their intended vote. Analyses suggested that participants who showed strong implicit and self-reported favoring of Whites compared to Blacks were also more likely to intend to vote for John McCain instead of Barack Obama. Collectively, the four race attitude measures accounted for 21% of the variation in intended vote. Further, after including liberalism-conservatism that is a (strong) predictor of vote and related to race attitudes, the race attitude measures still predicted 2% (p-value = 10e-24) of voting intention variance. Also, implicit and self-reported racial attitude measures each contributed unique predictive validity of intended vote. Of course, like any study of these relations, the data are correlational leaving open the possibility of unseen third-variables that are determinants of both racial attitudes and intended vote. However, in the absence of plausible alternative accounts, these results strongly suggest that race attitudes played a role in determining the 2008 Presidential vote.

The Surprisingly Limited Malleability of Implicit Racial Evaluations

When individuals are asked to report how much they prefer Black people to White people they might report egalitarian feelings, reporting no preference for either social group. However, psychological research has shown that sometimes individuals’ automatic evaluations do not reflect these explicit endorsements. In fact, sometimes individuals have preferences they might not publicly endorse, or even be consciously aware they hold. For example, an analysis of more than 900,000 completed Implicit Association Tests (IAT) at the Project Implicit website suggested that more than 70% of test takers associated White people with good and Black people with bad more strongly than the reverse. A popular question is whether these associations can be changed, and how. Brian Nosek and I contributed new research to this question in a new article that is scheduled to appear in a special issue of Social Psychology. 

Our research followed up on important demonstration by Buju Dasgupta and Tony Greenwald in a 2001 article. They found that implicit preferences for Whites compared to Blacks can be reduced by first exposing participants to admired Black and disliked White individuals such as Jackie Robinson and Jeffrey Dahmer. Their straightforward approach had a substantial effect on implicit preferences, and has potential for practical use because of its simplicity.

We initiated our research to examine the mechanisms that were contributing to the change in IAT effects in Dasgupta and Greenwald’s paradigm. Like many social psychologists, we believed that participants’ showed a reduced pro-White bias due to the exposure to admired Black individuals. As a result, in Experiment 1, participants only viewed positive exemplars prior to measuring their implicit bias. Despite having a very large sample (1,303 participants), we were unable to obtain a malleability effect. We conducted four more studies to replicate the original effect and figure out what factors are important for it to occur. In a follow-up study, we increased our procedure to more closely mirror Dasgupta and Greenwald’s experiment by exposing participants to admired Black and disliked White individuals. In addition, we gave participants repeated exposure to the individuals. With these procedural changes, we were able to successfully replicate the malleability finding, though our effect was much smaller than the one Dasgupta and Greenwald reported. To ensure our effect was real, we ran the same study and again found that participants viewing disliked White individuals and admired Black individuals showed less pro-White bias.

Having replicated the malleability effect, we turned our attention to why our effect was smaller than the one first reported by Dasgupta and Greenwald. Our first thought was that because our studies had been run on the Internet and not in the laboratory, our studies did not have an experimenter present. We hypothesized that an experimenter’s presence might increase the participants’ feelings of being judged. As a result, laboratory participants may be more motivated to show less bias. To test, we ran the same experiment both on the Internet as well as in the laboratory. Our results indicated that there were no differences between the samples, suggesting that participants in the laboratory did not show less implicit pro-White bias than Internet participants despite having an experimenter provide the materials in person.

Our research suggests that simply exposure to admired Blacks may not be enough for malleability to occur. Rather, repeated exposure and including negative White exemplars may be necessary to reduce bias. Together, these studies suggest that the original malleability effect is real, but weaker than previously reported.

Morning Meeting on MSNBC features the IAT and Tony Greenwald

A 10-minute segment features a lively discussion about implicit race bias and Tony Greenwald providing color commentary:

http://www.msnbc.msn.com/id/21134540/vp/32186549#32186549

Demonstrations of Various Implicit Measures

There are quite a few different measures that researchers use to measure automatic evaluations and stereotypes. Evaluative Priming was the first measure and it is still very popular. The IAT is probably the most popular implicit measure at the present, and it has many variants, such as the GNAT, the Brief-IAT, the Single-Target IAT and the Single-Block IAT. The SPF combines features of Evaluative Priming and the IAT, with one or two unique features of its own. The AMP is a promising measure that has some unique qualities compared with other implicit measures.

One way to learn about all these measures is through this page that I recently put together. The page lists many of the implicit measures that were developed over the years, and has links to sources of information about each measure. And, it also has a bonus: online demonstrations of many of these implicit measures. These demonstrations can give you some idea about what each implicit measure is about.

I did not create this page as the definitive source for implicit measures knowledge. At best, it could serve as a gate to the world of implicit measures. As such, I would like to add links to people’s more detailed sources of information about implicit measures, especially sources that summarize recent developments in the field. If you have a page or a website with useful materials, papers and other information related to one implicit measures or another, please contact me so I could add the link to my implicit measures page.

The full web address of my implicit measures page: http://www.bgu.ac.il/~baranany/imp.html

BBC podcast on gender-science stereotypes

A BBC podcast of an interview with Brian Nosek about Project Implicit‘s recent gender-science stereotypes article is available at the BBC World Service’s Science in Action series.

Implicit stereotypes predict sex differences in science and math achievement

[Press release from EurekaAlert]

Implicit stereotypes – thoughts that people may be unwilling to say or not even know that they have – may have a powerful effect on gender equity in science and mathematics engagement and performance, according to a new study published this week in the Proceedings of the National Academy of Sciences.

The international study involving more than half a million participants in 34 countries revealed that 70 percent harbor implicit stereotypes associating science with male more than with female.  Moreover, countries whose citizens stereotyped more strongly had larger sex gaps favoring boys in eighth-grade science and math achievement.

Implicit stereotypes may contribute to continuing underachievement and under participation among girls and women in science compared to their male peers.

“We found a general tendency, across every country that we investigated, that people on average have an easier time associating science concepts with male, rather than with female,” said lead investigator Brian Nosek, an associate professor of psychology at the University of Virginia.

“We correlated our data with a measure of actual science achievement among eighth graders in those 34 countries and found that in the countries with the largest sex gap – where the boys were performing much better than girls in math and science – there also was strongest implicit stereotyping of science as a male endeavor.”

The science and math achievement scores across nations came from the TIMSS (Trends in International Mathematics and Science Study) and were compared with the implicit stereotype data collected through Project Implicit led by these researchers publishing the study.

Surprisingly, there was no sex gap in the tendency to implicitly stereotype science as male. Male and female study participants showed equally strong associations of science with male.

Among nations represented in the study, the United States falls roughly in the middle of the pack in stereotyping science as male, and in the actual achievement of boys compared to girls at the eighth grade level.

The study is part of Project Implicit, a publicly accessible research and education Web site at which visitors can complete the Implicit Association Test to measure their own implicit associations. The test is available for a variety of topics involving gender, race, religion and politics.

In Project Implicit’s more than 10 years of existence, more than 10 million tests at the Web site have been completed by visitors around the world. A dozen years of research and hundreds of published studies suggest that people have implicit belief systems that may differ from their declared beliefs. These implicit beliefs are related to behavior, such as, interracial behavior, voting, and even drug use. A recent meta-analysis led by Anthony Greenwald, one of the researchers on the current study, provides evidence of the relationship between the IAT and a variety of behaviors from more than 100 studies.

“Participants are often surprised to learn that they may have unconscious biases involving gender or race or religion that are quite different from their stated beliefs,” said Fred Smyth, a co-investigator on the study and research assistant professor at the University of Virginia. This divergence between implicit and explicit beliefs, and the relation of both to behavior, suggests that behavior is influenced both by deliberate, explicit beliefs and by automatic, implicit reactions.

“Culture is a powerful force for shaping the beliefs and behavior of its members,” Nosek said.  “Even if one’s explicit beliefs change, the cultural residue may persist in memory and continue to influence behavior.”  This presents a challenge for addressing sex gaps in scientific engagement.

Over the past decade, Nosek, Mahzarin Banaji of Harvard University, and test creator Anthony Greenwald of the University of Washington have led the development of the Implicit Association Test to assess mental associations that may be different than what people know or say about themselves. For the gender and science study they worked with colleagues at universities and institutes across the globe.

For additional information about the study see: http://briannosek.com/papers/timss/

New Religion IAT launched

A new featured task measuring relative preferences for religions appeared today at the Demonstration site.  The task uses the new Brief IAT (Sriram and Greenwald, in press, Experimental Psychology).  The Brief IAT (patent pending) is an efficient version of the IAT allowing measurement of associations with perhaps 1/3 of the trials of a “normal” IAT.  The Religion IAT uses a novel 4-category format of the Brief IAT.  In this format, it is possible to measure relative associations strengths of four different religions (Christianity, Judaism, Islam and either Hinduism or Buddhism) with good.  Feedback is provided visually as a relative ranking of the four religions.

We employed this relative ranking format during the 2008 U.S. presidential primaries as well to obtain relative ranks of the Democratic and Republican candidates.  There is no formal manuscript reporting these results, or the methodological details of the 4-category Brief IAT format.  Such a manuscript may be forthcoming later this year.

Predictive validity meta-analysis of the Implicit Association Test is published

In the decade since the Implicit Association Test was introduced, its most surprising and controversial finding is its indication that about 70 percent of those who took a version of the test that measures racial attitudes have an unconscious, or implicit, preference for white people compared to blacks. This contrasts with figures generally under 20 percent for self report, or survey, measures of race bias.

A new study published this week validates those findings, showing that the Implicit Association Test, a psychological tool, has validity in predicting behavior and, in particular, that it has significantly greater validity than self-reports in the socially sensitive topics of race, gender, ethnicity, sexual orientation and age.

The research, published in the Journal of Personality and Social Psychology, is an overview and analysis of 122 published and unpublished reports of 184 different research studies. In this analysis, 85 percent of the studies also included self-reporting measures of the type generally used in surveys. This allowed the researchers, headed by University of Washington psychology Professor Anthony Greenwald, to compare the test’s success in predicting social behavior and judgment with the success of self-reports.

“In socially sensitive areas, especially black-white interracial behavior, the test had significantly greater predictive value than self-reports. This finding establishes the Implicit Association Test’s value in research to understand the roots of race and other discrimination,” said Greenwald. “What was especially surprising was how ineffective standard self-report measurers were in the areas in which the test measures have been of greatest interest – predicting interracial behavior.”

Greenwald created the Implicit Association Test in 1998 and he and Mahzarin Banaji, a Harvard psychology professor, and Brian Nosek, a University of Virginia associate professor of psychology, further developed it. Since then the test has been used in more than 1,000 research studies around the world. More than 10 million versions of the test have been completed at an Internet site where they are available as a self-administer demonstration.

The research looked at studies covering nine different areas – consumer preference, black-white interracial behavior, personality differences, clinical phenomena, alcohol and drug use, non-racial intergroup behavior, gender and sexual orientation, close relationships and political preferences.

Findings also showed that:

* Across all nine of these areas, measures of the test were useful in predicting social behavior.

* Both the test, which is implicit, and self-reports, which are explicit, had predictive validity independent of each other. This suggests the desirability of using both types of measure in surveys and applied research studies.

* In consumer and political preferences both measures effectively predicted behavior, but self-reports had significantly greater predictive validity.

Studies in the research came from a number of countries including Germany, the Netherlands, Italy, the United Kingdom, Australia, Canada, Poland and the United States. They looked at such topics as attitudes of undecided voters one-month prior to an Italian election; treatment recommendations by physicians for black and white heart attack victims; and reactions to spiders before and after treatment for arachnophobia, or spider phobia.

“The Implicit Association Test is controversial because many people believe that racial bias is largely a thing of the past. The test’s finding of a widespread, automatic form of race preference violates people’s image of tolerance and is hard for them to accept. When you are unaware of attitudes or stereotypes, they can unintentionally affect your behavior. Awareness can help to overcome this unwanted influence,” said Greenwald.