Phillip Atiba Goff

Introduction

Phillip Atiba Goff is an American social psychologist who has done more than anyone else to make policing a subject of psychological research. He is known for a body of laboratory and field work arguing that racial discrimination does not require racial prejudice: situational cues, institutional routines, and automatic associations can produce unequal outcomes even among people who hold no hostile beliefs. He co-founded the Center for Policing Equity and is the principal investigator of the National Justice Database, the largest collection of American police behavioural data assembled by an outside research group. Goff taught at Pennsylvania State University and the University of California, Los Angeles, held the inaugural Franklin A. Thomas Professorship in Policing Equity at John Jay College of Criminal Justice, and moved to Yale University in 2020, where he is the Carl I. Hovland Professor of Black Studies and Professor of Psychology. In recent years he has used the name Phillip Atiba Solomon.

Early Life and Education

Goff was born in Philadelphia in 1977, the son of Edwin L. Goff, an associate dean at Villanova University, and Florence Withers Goff. He earned an A.B. from Harvard University in 1999 in Afro-American studies — an unusual major for a future quantitative social psychologist, and one that located his later work in the study of race rather than merely in the study of bias. He then went to Stanford University, taking an M.A. in social psychology in 2001 and a Ph.D. in 2005. His dissertation, "The space between us: stereotype threat for whites in interracial domains," was supervised by Claude Steele and inverted the standard stereotype threat question: rather than asking how negative stereotypes depress the performance of stigmatised groups, it asked what happens to white participants when they anticipate an interracial interaction. In 1999 he also co-founded Deep Dickollective, an Oakland-based queer hip hop collective in which he performed as "Lightskindid Philosopher."

Academic and Professional Career

Goff taught at Pennsylvania State University in 2004–2005 before joining the University of California, Los Angeles, where he became an associate professor of social psychology and, in 2007 or 2008, co-founded the Center for Policing Equity as a research and action centre. He has also been a visiting scholar at Harvard's Kennedy School of Government. In 2016 he was appointed the inaugural Franklin A. Thomas Professor in Policing Equity at John Jay College of Criminal Justice — the college's first endowed professorship — and the Center moved with him. In 2020 both he and the Center moved to Yale; his named chair took effect in January 2021. He was a key figure in the founding of the National Initiative for Building Community Trust and Justice in 2014 and served as a panelist for President Obama's Task Force on 21st Century Policing. His awards include two American Psychological Association early career awards, the Association for Psychological Science Rising Star award, the National Organization of Black Law Enforcement Executives' Lloyd G. Sealy Award, the Society of Experimental Social Psychology's Career Trajectory Award, and the APA Award for Distinguished Contributions to Research in Public Policy.

Major Contributions to Psychology

Discrimination without prejudice

The organising claim of Goff's research is that the standard model — prejudice in the head produces discrimination in the world — fails to explain most of the racial inequality that actually shows up in police data. He argues that situational factors do the work: ambiguous cues, time pressure, departmental incentives, and the statistical structure of where police are sent. That framing has practical consequences, because it implies that screening officers for biased attitudes will not by itself change outcomes, while changing the situations will.

Dehumanization and the Black–ape association

In a 2008 Journal of Personality and Social Psychology article with Jennifer Eberhardt, Melissa Williams, and Matthew Christian Jackson, Goff reported that the implicit association between Black people and apes — a dehumanising stereotype with a long history in American imagery — was measurable in undergraduates and predicted real outcomes: the stronger the association in a jurisdiction, the more likely Black defendants there were to have been sentenced to death in cases involving white victims. The paper was among the first to link an implicit measure to a legally consequential outcome and became one of the most cited studies of its kind.

The essence of innocence

Goff's best-known study, "The essence of innocence: consequences of dehumanizing Black children" (JPSP, 2014), reported across four studies that Black boys are seen as older and less innocent than white boys of the same age. Undergraduates judged children up to age nine as equally innocent regardless of race, but from age ten onward rated Black children as significantly less innocent; they overestimated the age of Black boys by an average of 4.5 years and rated them more culpable, especially when the alleged offence was serious. In a sample of 176 mostly white male police officers, those higher in implicit Black–ape dehumanization were more likely to have used force against a Black child in custody. Notably, explicit prejudice did not predict use of force, while dehumanization did — the dissociation that gives the research programme its theoretical point. The term adultification that Goff and others use for this phenomenon has since appeared in state appellate opinions.

The National Justice Database

The second half of Goff's career has been infrastructural. American police agencies are under no general obligation to publish use-of-force data, and the federal systems that do collect it have well-documented coverage gaps. The National Justice Database, funded by the National Science Foundation and hosted at the Center for Policing Equity, collects incident-level data on stops and force from agencies that volunteer to participate and returns department-level diagnostics. Its analytic framework has been described as a potential national model for police data accountability and has been used in jurisdictions including Austin, Texas.

Major Publications

Goff, P. A., Eberhardt, J. L., Williams, M. J., & Jackson, M. C. (2008). Not yet human: Implicit knowledge, historical dehumanization, and contemporary consequences. Journal of Personality and Social Psychology, 94(2), 292–306.
Goff, P. A., Jackson, M. C., Di Leone, B. A. L., Culotta, C. M., & DiTomasso, N. A. (2014). The essence of innocence: consequences of dehumanizing Black children. Journal of Personality and Social Psychology, 106(4), 526–545.
Goff, P. A., Thomas, M. A., & Jackson, M. C. (2008). "Ain't I a woman?": Towards an intersectional approach to person perception and group-based harms. Sex Roles, 59(5–6), 392–403.
Goff, P. A., & Kahn, K. B. (2013). How psychological science impedes intersectional thinking. DuBois Review: Social Science Research on Race, 10(2), 365–384.
Goff, P. A. (2014). America's lack of a police behavior database is a disgrace. The New Republic.

Impact and Legacy

Goff moved the study of policing from the margins of criminology into mainstream social psychology, and he did it while holding two roles that are usually kept separate: university researcher and chief executive of an advocacy-adjacent organisation advising the very departments he studies. His framing — that bias lives in situations and systems, not only in individuals — has shaped police reform programmes, congressional testimony, and the design of use-of-force reporting across dozens of agencies.

That dual role is also the source of the sharpest criticism. The Center for Policing Equity both produces the research on participating departments and sells or provides reform services to them, which raises an obvious question about whether an organisation with a stated mission "to eliminate racism in public safety" can audit its own clients impartially. Defenders point out that the alternative — no data at all — is worse, and that publishing methods is itself a form of accountability.

The methodological criticism is more tractable and more damaging. Participation in the National Justice Database is voluntary, so the agencies in it are self-selected departments that were already willing to be examined; they are not a random sample of American policing, and generalising from them in either direction is unsafe. More fundamentally, disparity estimates depend on the choice of benchmark — the population against which stop rates are compared — and this choice is not a technical detail. Criminologists including Greg Ridgeway and, more recently, Ratcliffe and Hyland have shown that comparing stops to census population, to calls for service, to crime locations, or to arrestee demographics produces dramatically different disparity estimates from the same data; in one published illustration, a census benchmark produced a disparity nearly three times the size of the two activity-based benchmarks. Because police deployment follows crime and calls rather than population, census-based denominators systematically overstate disparity, and activity-based denominators rest on data that is itself produced by police. There is no benchmark without weaknesses, which means that headline disparity numbers are far less stable than their public presentation suggests. The underlying administrative records have their own problems: they are collected for administrative purposes, record the officer's perspective, and frequently omit the contextual detail needed to explain a given encounter.

The laboratory work has drawn parallel criticism. The 2014 officer sample consisted of 176 mostly white male officers from a single department, and the authors themselves acknowledged that dehumanization could be a consequence of negative encounters with Black children rather than a cause of force against them — a reverse-causation caveat that has largely disappeared from the popular retellings. The ape-priming experiment, like most social priming effects of its era, rests on a between-subjects design with a small number of participants and has not been the subject of a large registered replication; given what the replication crisis did to comparable findings, that absence is a liability rather than a technicality. The 2008 death-sentence finding has been criticised for the small number of capital cases involved and for the difficulty of ruling out confounding by case severity and jurisdiction. Goff's response has been consistent and partly effective: he argues that waiting for perfect data is itself a choice that favours the status quo, and he has invested in building the infrastructure that would allow the questions to be answered properly.

Learn More

Jennifer Eberhardt — his Stanford collaborator on the dehumanisation and racial-bias work.
Anthony Galt Greenwald — whose implicit-association research supplies the measurement tradition he works in.
Prejudice — the phenomenon this research programme targets.
Bias — the measurement of implicit and explicit bias.

References

Goff, P. A., Eberhardt, J. L., Williams, M. J., & Jackson, M. C. (2008). Not yet human: Implicit knowledge, historical dehumanization, and contemporary consequences. Journal of Personality and Social Psychology, 94(2), 292-306.