Implicit Association Test (IAT)

Abstract

The Implicit Association Test (IAT) is a computerized measure developed by Greenwald, McGhee, and Schwartz (1998) in the United States. Its primary function is to assess implicit cognitions, attitudes, and associations that individuals may be either unwilling or unable to consciously report via self-report measures. The IAT overcomes limitations such as social desirability bias and lack of self-awareness by quantifying the relative strength of mental associations based on response latencies (reaction times) during rapid categorization tasks.

The test involves participants sorting words or images representing two contrasted concepts (e.g., young vs. old) and two contrasted attributes (e.g., good vs. bad). The speed with which a participant can categorize items when certain concepts and attributes are paired indicates the strength of the implicit association. This methodology has proven highly adaptable, allowing researchers to measure a wide range of constructs, including implicit bias, self-esteem, and personality traits.

Keywords

Implicit Association Test, IAT, implicit bias, reaction time, response latencies, implicit cognitions, attitude measurement, social psychology, self-concept, computerized testing.

Authors

Anthony G. Greenwald, Debbie E. McGhee, Jordan L. K. Schwartz

[quads id=5]

Purpose

The central purpose of the IAT is to provide a standardized, objective measure of implicit attitudes and beliefs. Traditional self-report methods often fail to capture true underlying attitudes because participants may be influenced by intentional misrepresentation or simply lack conscious access to their own implicit associations.

By relying on the speed of categorization rather than deliberate responses, the IAT aims to quantify automatic mental connections. It was initially designed primarily for research purposes, offering a tool to study constructs such as the development of racial bias and implicit bias among specific professional groups, such as healthcare practitioners. Furthermore, there is potential for the IAT to be used in clinical settings alongside explicit measures to gain a more comprehensive understanding of a client’s self-awareness and functioning.

Construct

The IAT measures the relative strength of associations between pairs of contrasted concepts (e.g., Male/Female, Black/White) and pairs of contrasted attributes (e.g., Good/Bad, Strong/Weak). The core construct is the implicit association, defined as automatic mental linkages between a target concept and an evaluative attribute.

The test operates on the principle that if two concepts are strongly associated in memory, categorization will be faster when they share a response key than when they require opposing responses. Examples of constructs successfully assessed using adapted IAT formats include self-esteem, self-concept, personality, and various forms of implicit bias.

Validity

The psychometric soundness of the IAT has been the subject of extensive academic debate, particularly concerning its validity. Research has shown good evidence for its convergent and discriminant validity (Gawronski, 2002; Grumm & von Collani, 2007) when compared to other measures.

However, concerns have been raised regarding its construct validity (Nosek et al., 2005) and predictive validity (Greenwald et al., 2009). Critics argue that the IAT might be measuring factors other than implicit bias, such as cultural familiarity or shared extra-personal associations, rather than genuine personal implicit attitudes. Nevertheless, studies have demonstrated that IAT effects persist even when controlling for familiarity (Dasgupta et al., 2000). A comprehensive meta-analysis indicated that the IAT does possess meaningful predictive validity, although the correlation between IAT scores and explicit self-report measures is often modest (Hofmann et al., 2005; Greenwald et al., 2009).

Reliability

The IAT generally demonstrates good test-retest reliability (Greenwald & Farnham, 2000). It is important to note that both the reliability and validity metrics can vary depending on the specific stimuli and conceptual categories employed in a given version of the IAT. Overall, the instrument is often regarded as a relatively psychometrically sound tool when compared to other available measures of implicit attitudes (Nosek et al., 2007).

Factor Analysis

As the IAT is a behavioral measure based on measuring the strength of association through response latencies, it does not rely on traditional factor analysis in the same way as multi-item self-report questionnaires. Instead, psychometric evaluation focuses on the reliability of the D-score algorithm (Greenwald, Nosek, & Banaji, 2003), which calculates the effect size of the difference in reaction times between compatible and incompatible tasks.

Analysis often focuses on the instrument’s ability to consistently measure the relative strengths of the four associations being tested, and how these scores correlate with criterion variables (predictive validity) or other measures of similar constructs (convergent validity).

Instrument

Test Type: Computerized Performance Task (Implicit Measure)

Format: Computer-administered sequence of rapid categorization/discrimination tasks using keyboard presses (left/right key assignments). Raw data consists of response latencies (reaction times).

Language Available: Highly adaptable; stimuli can be presented verbally or textually, making it suitable for translation across numerous languages and cultures depending on the specific concepts being measured.

Population Group: General population, including adolescents, adults, and specific clinical/research cohorts (e.g., healthcare practitioners).

Age Group: Children (as young as 5 years old via the ChildIAT version), adolescents, and adults.

Population Details: The Child-oriented version (ChildIAT) incorporates modifications such as larger keys and verbal recordings of words to accommodate varied reading abilities in younger participants (Baron & Banaji, 2006).

Test Methodology: The IAT measures the speed of association between two target concepts and two attribute categories across five sequential blocks (initial discrimination, attribute discrimination, combined compatible task, reversed concept discrimination, combined incompatible task). Faster reaction times in the combined blocks indicate a stronger automatic association.

Keywords

Implicit attitudes, social desirability bias, reaction time, D-score, psychometric soundness, Implicit Association Test, self-esteem, personality, construct validity, predictive validity.

[quads id=5]

Authors

Author ORCID Identifier: N/A (Not provided in source material)

Affiliation Email addresses: N/A (Not provided in source material)

Correspondence Address: N/A (Not provided in source material)

Permissions & Fee and Test Year

Permission to use the IAT instrument is generally not required, as the core software and stimuli are often made freely available for research purposes. Guidelines regarding the proper use of the IAT have been suggested by Nosek et al. (2005).

The software and stimuli for the IAT are freely available for download via http://www.millisecond.com/download/. A wide variety of stimuli and personalized packages are also available via http://projectimplicit.net/nosek/papers/pcias/. Further information and tests for personal use are available through Harvard’s Project Implicit website at http://implicit.harvard.edu.

The original PDF detailing the improved scoring algorithm for the IAT can be downloaded here: Greenwald, A. G., Nosek, B. A., & Banaji, M. R. (2003).

Test Year: 1998 (Original publication)

Reference’s

• Baron, A. S., & Banaji, M. R. (2006). The development of implicit attitudes: Evidence of race attitudes from children and adults. Developmental Science, 9(1), 75–83.
• Blair, I. V., Steiner, J. F., & Havranek, E. P. (2012). Unconscious bias and racial disparities in health care. Journal of the American Board of Family Medicine, 25(5), 721–728.
• Dasgupta, N., McGhee, D. E., Greenwald, A. G., & Banaji, M. R. (2000). Automatic preference for White Americans: Eliminating the familiarity explanation. Journal of Experimental Social Psychology, 36(3), 316–328.
• FitzGerald, C., & Hurst, S. (2017). Implicit bias in healthcare professionals: A systematic review. BMC Medical Ethics, 18, 19.
• Frantz, C. M., Cuddy, A. J. C., Burnett, M., Ray, H., & Hart, A. (2004). A threat in the computer: The race implicit association test as a stereotype threat experience. Personality and Social Psychology Bulletin, 30(12), 1611-1624.
• Gawronski, B. (2002). What does the Implicit Association Test measure? A test of the convergent and discriminant validity of the IAT. Experimental Psychology, 49(3), 171–180.
• Greenwald, A. G., & Farnham, S. D. (2000). Using the Implicit Association Test to measure self-esteem and self-concept. Journal of Personality and Social Psychology, 79(6), 1022–1038.
• Greenwald, A. G., McGhee, D. E., & Schwartz, J. L. K. (1998). Measuring individual differences in implicit cognition: The implicit association test. Journal of Personality and Social Psychology, 74(6), 1464-1480.
• Greenwald, A. G., Nosek, B. A., & Banaji, M. R. (2003). Understanding and using the Implicit Association Test: I. An improved scoring algorithm. Journal of Personality and Social Psychology, 85(3), 197-216.
• Greenwald, A. G., Poehlman, T. A., Uhlmann, E., & Banaji, M. R. (2009). Understanding and using the Implicit Association Test: III. Meta-analysis of predictive validity. Journal of Personality and Social Psychology, 97(1), 17–41.
• Grumm, M., & von Collani, G. (2007). Measuring implicit personality constructs with the Implicit Association Test: Convergent and discriminant validity of the IAT-personality. European Journal of Psychological Assessment, 23(1), 21–29.
• Hofmann, W., Gawronski, B., Gschwendner, T., Le, H., & Schmitt, M. (2005). A meta-analysis on the correlation between the implicit association test and explicit self-report measures. Personality and Social Psychology Bulletin, 31(10), 1369-1385.
• McConnell, A. R., & Leibold, J. M. (2001). Relations among the implicit association test, discriminatory behavior, and explicit measures of racial attitudes. Journal of Experimental Social Psychology, 37(5), 435–442.
• Nock, M. K., & Banaji, M. R. (2007). Prediction of suicide attempts by implicit associations. Journal of Personality and Social Psychology, 92(4), 629–636.
• Nosek, B. A., Greenwald, A. G., & Banaji, M. R. (2005). Understanding and using the Implicit Association Test: II. Method variables and construct validity. Personality and Social Psychology Bulletin, 31(2), 166-180.
• Nosek, B. A., Greenwald, A. G., & Banaji, M. R. (2007). The Implicit Association Test at age 7: A methodological and conceptual review. In J. A. Bargh (Ed.), Implicit measures of social cognition (pp. 265–292). Guilford Press.
• Olson, M. A., & Fazio, R. H. (2004). Reducing the influence of extrapersonal associations on the implicit association test: Personalizing the IAT. Journal of Personality and Social Psychology, 86(5), 653–667.
• Rachlinski, J. J., Johnson, S. L., Wistrich, A. J., & Guthrie, C. (2009). Does unconscious bias affect trial judges? Notre Dame Law Review, 84(3), 1195–1241.
• Rudman, L. A., & Ashmore, R. D. (2007). Discrimination and the Implicit Association Test. Group Processes & Intergroup Relations, 10(3), 359-372.
• Siegel, E. F., Dougherty, M. R., & Huber, D. E. (2012). Manipulating the role of cognitive control while taking the implicit association test. Journal of Experimental Social Psychology, 48(5), 1057-1068.
• Teachman, B. A., Smith-Janik, S. B., & Saporito, E. (2007). Implicit and explicit anxiety processes: Their relationship with worry and fear. Cognitive Therapy and Research, 31(1), 1–22.
• Thush, C., & Wiers, R. W. (2007). Implicit and explicit alcohol-related cognitions and alcohol use in adolescents. Addictive Behaviors, 32(5), 996–1004.

[quads id=5]

Items of the Implicit Association Test (IAT)

IMPORTANT: The following scale items must be preserved in their original language and must not be changed in any way.

Implicit Association Test (Race)

Questions Used to Obtain Optional Self-Report Measures Prior to Implicit Association Test (IAT) Measures

Likert Items

One 5-point Likert item was used in conjunction with each IAT, illustrated rated here for the Age IAT:

Which statement best describes you?

• I strongly prefer young people to old people.
• I moderately prefer young people to old people.
• I like young people and old people equally.
• I moderately prefer old people to young people.
• I strongly prefer old people to young people.

For the Race IAT, the italicized concept words were replaced with European Americans and African Americans. For the Election 2000 IAT the concepts were George W. Bush and Al Gore.

For the Gender–Science IAT, the Likert item was as follows:

Which statement best describes you?

• I strongly associate liberal arts with females and science with males.
• I moderately associate liberal arts with females and science with males.
• I associate males and females with science and liberal arts equally.
• I moderately associate science with females and liberal arts with males.
• I strongly associate science with females and liberal arts with males.