The Effects of Identity and Beliefs on Perceptions and Judgments

September 2021

Cognitive psychologist Daniel Kahneman described how three heuristics (availability, representativeness, and anchoring) distort human judgments. Decades of research have confirmed the human information processing system has multiple channels (Zajonc, 1980; Broadbent, Fitzgerald, & Broadbent, 1986; Haidt, 2006).

One mode is fast, but another (the one that employs reason) is comparatively slow and of limited capacity (Porter, 1991). Haidt (2006) compared the slow, conscious, and rational system’s control of the quick, and often emotional, limbic system to a rider atop a wild elephant. Kahneman’s (2013) Thinking Fast and Slow uses these alternative processing modes to explain human error. Noise, A Flaw in Human Judgment (Kahneman, Sibony, & Sunstein, 2021) integrates the effects of random influences (viz., noise) to suggest improvements in human decision making.

Human information processing is integral to how humans live, work, and play. Industrial/Organizational Psychology is a “scientific study… that addresses psychological concepts and principles in the work world” (Muchinsky, 2009, 470). Understanding human cognition is necessary for developing and implementing effective organizational polices. This is particularly important at colleges and universities. Personnel policies must be considered from both a social and a legal perspective. From a socio-cultural perspective, institutions involved in higher learning have acquired several distinctive characteristics.

One of these is academic freedom. This concept emerged slightly over 100 years ago. The American Association of University Professors asserted the relevant principles in 1915. These principles have been reviewed, revised, reaffirmed, and interpreted often over the last century (AAUP, 2020). Although there are several versions of academic freedom (Fish, 2014), at its core, it asserts faculty members’ right to research and teach without intrusion. Unlike freedom of speech, however, academic freedom is not absolute. It protects bona fide professional efforts to seek and share the truth but both research and teaching are subject to peer review of quality.

Socio-culture norms are not the only constraint on personnel policies. In the United States, since passage of the 1964 Civil Rights Act, new restrictions have constrained personnel actions in nearly all organizations (Muchinsky, 2009; Caldwell, 2021). Title VII defines two types of sexual harassment: quid pro quo and hostile environment harassment. However, it was not until Title IX, the 1972 education amendment to the Civil Rights Act, that these rules applied directly to colleges and universities. Quid pro quo harassment (i.e., “a lay for an A”) was an obvious and universally condemned predatory behavior. However, the other form of discrimination involving protection from hostile environments was more controversial (Shibley, 2016; Chemerinsky & Gillman, 2017; MacDonald, 2018; Strossen, 2018; Whittington, 2018; Murray, 2019; Porter, 2021).

Here’s the rub: harassment by hostile environment is defined by Title VII as the discomfort experienced by the alleged victim and the subsequent self-reported interference with the victim’s participation in the educational activities. Intent is irrelevant. However, from the time of Socrates, asking disquieting questions and challenging social stereotypes have been integral to higher learning. As Mr. Dooley asserted, good education, like good religion, should “comfort the afflicted and afflict the comfortable” (Shields, 2004).

It is difficult to bring cases of academic freedom before federal courts, so American colleges and universities have been left alone to create and implement their own resolutions to these inherent tensions. Nonetheless, presented with the opportunity, federal courts have expressed strong support for academic freedom and freedom of speech as the Rodriquez v. Maricopa County Community College System decision (Creeley, 2010) illustrates:

“Intellectual advancement has traditionally progressed through discord and dissent, as a diversity of views ensures that ideas survive because they are correct, not because they are popular. Colleges and universities… have historically fostered that exchange. But that role in our society will not survive if certain points of view may be declared beyond the pale…. Teachers and students must always remain free to inquire, to study and to evaluate, to gain new maturity and understanding; otherwise our civilization will stagnate and die (Sweezy v. New Hampshire, 1957). We have therefore said that ‘[t]he desire to maintain a sedate academic environment … [does not] justify limitations on a teacher’s freedom to express… vigorous, argumentative, unmeasured, and even distinctly unpleasant terms’ (Adamian v. Jacobsen, 1975).”

Recent Studies:

Norris, Larsen, & Stastny’s (2010) study of torture demonstrated how scenarios could be used to collect information about respondents’ perceptions and judgments about hostile environments. They found respondents disagreed about which behaviors constituted “torture.” Also, respondents’ perceptions of torture showed biases caused by “extraneous factors” such as the political identity of the respondent and the national identity of torturers and their victims.

A survey of 1500 American college students by Villasenior (2017) concluded that “freedom of expression was deeply imperiled on US campuses.” About half of the national sample (62% Democrats and 39% Republicans) believed disrupting a speech by an invited but controversial speaker would be appropriate; fewer than half were aware the First Amendment protected “hate speech.” A subsequent survey of 4,400 college students by College Pulse (2019) confirmed that half the student respondents believed “shouting down speakers” was “sometimes acceptable.” Asking respondents to choose between these statements: “Some students are too easily offended” and “People need to be more careful about their language” revealed dramatic differences based on respondents’ gender and sexual orientation. Women and non-binary students were more supportive of the latter (inclusive) statement but 74% of heterosexual males chose the first statement.

The effects of gender and sexual orientation on respondents’ perceptions is problematic (Clark, 2021b). Equally concerning is the apparent influence of political identity on judgments concerning administrative policies. Suggesting that both social scientists and organizational administrators share liberal biases, Clark (2021c) opines that the premature application of findings has created “various false positives” and many “failed interventions.”

The Course:

The first author found himself teaching an Industrial/Organizational Psychology course in the spring of 2018 on a campus sharply divided on views about inclusion and academic freedom. Students were acutely aware of these tensions; many were distressed by the administration’s lack of transparency and penchant for imposing politically correct solutions rather than encouraging open academic debate suggested by federal courts and expected as part of the college’s commitment to “liberal arts education.” The course’s research component would involve a comprehensive survey of the campus community about these issues.

The Survey:

The first author had been an advisor for a recent Title IX prosecution and observed apparent procedural inconsistencies contrary to published policies, due process, and academic freedom. Class discussions highlighted many other relevant issues. The survey contained variables suggested by the literature or students’ own “lived experiences.” Suggestions from six faculty reviewers and enthusiastic class discussions produced a survey of just over 70 items. These included demographic questions about race, gender, and sexual orientation. There were also questions concerning respondents’ explicit beliefs about freedom of speech, academic freedom, activism, hostile environments, and political orientation. The largest portion of the survey was devoted to realistic scenarios which described hypothetical but ecologically valid events. Each scenario asked respondents to rate the extent to which they perceived the incident described as violating college policy prohibiting hostile environments. The subsequent survey item asked if the behaviors that created the situation would be protected by academic freedom. Responses were on a 6-point Likert scale. The survey was offered through the Qualtrics anonymous response option (i.e., collecting no individually identifying information).

The Sample:

Drawings for several $50 gift cards were offered and enrolled students individually endorsed the survey invitation to the campus. We had hoped for 500 respondents. Unfortunately, after an impassioned social media posting appeared that falsely claimed the survey was retaliation intended to punish past Title IX grievants and prevent future victims from filing complaints, the campus erupted. Shortly afterwards, the survey itself was sabotaged; over a thousand “dummy” submissions made it impossible for other students to participate. Nonetheless, 120 valid and complete responses had already been received. Better yet, an examination of the demographic data indicated that respondents approximated the campus characteristics in terms of race, gender, and sexual orientation. Our sample was 73% white, 58% female, and 67% heterosexual. These data also indicated the campus had a significant list to the political left (70% liberal; 20% uncertain; & 10% conservative).

The Statistics: Multiple Regressions and Path Analysis:

A complete account of the statistical analysis is beyond the scope of this essay. (Much of this information, including the survey itself with item-by-item results, is available here. Our course text suggested multiple regression was an appropriate way to examine multiple causation. Cohen & Cohen (1983) recommended using path analysis to organize and present multiple regression analyses. For the non-social scientists, the chart below may appear daunting; hopefully, a brief explanation will enhance understanding.

The chart contains nine variables and sixteen predictive paths. Each variable is represented as a circle. Variables are arranged in a predictive flow from left to right. Demographic variables are assumed to predict, but not be predicted by, other variables; they are along the left margin. The next four variable represent explicit beliefs. The placement of the Politics variable to the left of the other belief variables (Activism, Hostile Environment Protection, and Academic Freedom) reflects a notion that human beliefs are deductive. Our broad beliefs inform (i.e., predict) specific beliefs. To maintain cognitive consistency, our beliefs also affect one another. When we encounter a novel situation (e.g., reading a survey scenario), our identity and beliefs may affect our perception. Theoretically, when then asked to decide, our perception as well as the demographic and belief variables may predict that judgment. Although the chart appears busy, it only contains those pathways found to be “statistically significant” (i.e., the observed relationships were unlikely to have occurred by chance).

The circles representing each non-demographic variable contain important information. The R2 is the portion of the variance explained by the predictive pathways leading to each variable. In general, values of 25% are considered adequate; 33%, good; and 50%, outstanding. Of the six dependent variables depicted, only the variable reflecting respondents’ explicit support for academic freedom has an R2 of less than 25%. This is likely the result of too little variance; everyone explicitly endorsed the importance of academic freedom.


Gender and Sexual Orientation combine to predict over a third of the variance in political identification. As in Clark’s (2021b) study, heterosexual males differed from the rest of the campus community. Hostile Environment Perception was the variable that was predicted best by the preceding variables; over half its variance was “explained” by their influence. (The number within each predictive pathway provides an estimate of its relative predictive value.) While such a high level of predictability is a positive outcome for the study, its implications for institutions are worrisome. Over half the total variance in how individuals perceive situations is predicted by who they are and what they believe. All four of these variables are extraneous (to use Norris, et al.’s terminology) and thus reflect bias.

Even more dire is the prediction of decisions about academic freedom. In this case, two variables predict 42% of the variance in judgments. However, the influence of the prior perception of a hostile environment is much greater than the influence of the respondent’s explicit beliefs about academic freedom. Most members of the Berea College campus community are only willing to apply academic freedom to incidents they perceive not to be hostile or offensive.

The administration’s unwillingness to address these issues is understandable. However, the measures they took to suppress discussion of these issues violated their promises concerning academic freedom, due process, and other Constitutional rights. More information about these issues can be found here, here, and here. More information about the study is here.


  1. American Association of University Professors (AAUP). (2016). Executive summary of The History, Uses, and Abuses of Title IX.
  2. American Association of University Professors (AAUP). (2020). “1940 Statement of Principles on Academic Freedom and Tenure.”
  3. Broadbent, D. E., Fitzgerald, P. & Broadbent, M. H. P. (1986). Implicit and explicit knowledge in the control of complex systems. British Journal of Psychology, 77, 33—50.
  4. Caldwell, C. (2020). The Age of Entitlement; America since the sixties. New York: Simon & Schuster.
  5. Chemerinsky, E. & Gillman, H. (2017). Free Speech on Campus. New Haven, CT: Yale University Press.
  6. Clark, C. (2021a). The evolutionary advantage of playing victim. Quillette, (Feb 27).
  7. Clark, C. (2021b). The gender gap in censorship support; research suggests women favor inclusivity over academic freedom. Psychology Today Blog. (Apr 28).
  8. Clark, C. (2021c). Biased science makes bad policy; organizations frequently apply scientific findings to policy making prematurely. Psychology Today Blog (June 29).
  9. Cohen, J. & Cohen, P. (1983). Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences (2nd ed). Mahwah, NJ: Lawrence Erlbaum & Associates.
  10. College Pulse. (2019). Free Expression on College Campuses (May, 2019). Knight Foundation.
  11. Creeley, W. (2010). “Ninth Circuit delivers resounding victory for First Amendment on Campus.” Foundation for Individual Rights in Education (May 21, 2010).
  12. Fish, S. (2014). Versions of Academic Freedom; from professionalism to revolution. Chicago, IL: University of Chicago Press.
  13. Haidt, J. (2006). The Happiness Hypothesis: Finding modern truth in ancient wisdom. New York: Basic Books.
  14. Kahneman, D. (2011). Thinking Fast and Slow. New York: Farrar, Straus, & Giroux
  15. Kahneman, D., Sibony, O., & Sunstein, C.R. (2021). Noise, A Flaw in Human Judgment. New York: Little, Brown Spark.
  16. Kaufman, E. (2021). “Academic Freedom in crisis: punishment, political discrimination, and self-censorship.” Center for Science in the Public Interest Report #2. (Knight Foundation).
  17. Lukianoff, G. & Haidt, J. (2018). The Coddling of the American Mind; How good intentions and bad ideas are setting up a generation for failure. New York: Penguin Press.
  18. Muchinsky, P.M., (2009). Psychology Applied to Work (9th Edition). Summerfield, NC; Hypergraphic Press.
  19. Mac Donald, H. (2018). The Diversity Delusion: How race and gender pandering, corrupt the university and undermine our culture. New York: St Martin’s Press.
  20. Murray, D. (2019). How did the world lose its marbles? A new dogma has turned beliefs that once seemed common sense into hate crimes. Daily (Sept 1, 2019).
  21. Norris, J.I., Larsen, J.T. & Stastny, B.J. 2010. Social perceptions of torture: genuine disagreement, subtle malleability, and in-group bias. Peace and Conflict, 16, 275-294. DOI: 10:1080/10781919.2010.488164.
  22. Porter, D.B. (1991). Computer games and cognitive processes: two tasks, two modes, too much? British Journal of Psychology, 82. 343-357.
  23. Porter, D.B. (2021). “Hostile Environment Concerns May Cancel Academic Freedom.” Written testimony provided to Congressional Hearing Panel – Office of Civil Rights Consideration of Title IX Modifications. SAVE News and Commentary (June 6th).
  24. Shibley, R.L. (2016). Twisting Title IX: Encounter Broadside No. 49. New York: Encounter Books.
  25. Shields, M. (2004). The timeless wisdom of Mr. Dooley. Netscape- (March 1, 2004)
  26. Strossen, N. (2018). Hate; Why we should resist it with free speech, not censorship. New York: Oxford University Press.
  27. Tversky, A. & Kahneman, D. (1974). Judgment Under Uncertainty: Heuristics and Biases. Science, 27 Sep. Vol 185, Issue 4157, pp 1124-1131. DOI: 10.1126/science. 185.4157.1124
  28. Villasenor, J. (2017). Views among college students regarding the First Amendment: Results from a new survey. Brookings Blog (Sep 18, 2017).
  29. Whittington, K.E. (2018). Speak Freely; Why Universities Must Defend Free Speech. Princeton, NJ: Princeton University Press.
  30. Zajonc, R. B., (1980). Feeling and Thinking: Preferences Need No Inferences. American Psychologist, 35, 151-175.