Don't Joke About Me: Student Identities and Perceptions of Instructor Humor in College Science Courses.

Humor is a popular tool used by instructors to engage students. However, some instructor jokes may be perceived as less funny and more offensive by particular groups of students. Previous studies have shown that student gender impacts student perception of instructor humor; however, to our knowledge no studies have explored whether there are differences in how other identity groups interpret instructor humor. In this study, we surveyed 1,637 students across 25 different college science courses at a research-intensive institution in the Southwest United States. Students evaluated a set of topics that science instructors might joke about in class as to whether they were funny and offensive. Using binary logistic regression, we analyzed whether students of different identities, including race/ethnicity, political affiliation, LGBTQ+ status, religious affiliation, and native language, differentially perceived joke topics to be funny and offensive if told by an instructor in class. We identified that topics which tended to be perceived by students as funny rather than offensive were generally less likely to be perceived as funny to non-native English language speakers compared with native English speakers. We also found that students were more likely to be offended by jokes about their own identity group. This work identifies potentially humorous topics that instructors should avoid because they could be offensive to groups of students. This study also highlights topics that tend to be perceived as funny to most students, which indicates that instructors who joke about such topics may be universally benefitting college science students. ©2020 Author(s). Published by the American Society for Microbiology.

“I enjoy when professors make puns. My biology professor uses puns quite often and it helps me focus and enjoy the class more. An example of [one of] his pun[s] included a picture of a goat [with] the pun heterozygoat.”

“There was once when the instructor cracked a joke about race during a lecture. The students gave no reaction, but instead gave an awkward silence.”

INTRODUCTION

Humor can be a powerful tool that science instructors can use to connect with students, build classroom community, and create a more engaging classroom (1–5). Studies have shown that humor can help increase instructor immediacy, narrowing the psychological distance between students and instructors, and help students perceive instructors as more relatable (6–8). Science instructors have been previously described as uninteresting and unfriendly (9–13), so humor is one method by which they can seem more like a “real person” (6). Further, science content can be perceived as difficult, dry, or boring, so humor can give students a mental break from the high cognitive load of the classroom and help them engage with the material (1, 6, 14). Finally, the potential psychological benefits of humor are numerous, including reduced anxiety and stress levels, as well as enhanced self-efficacy and self-motivation (3, 5, 14, 15).

While humor can lead to an array of positive outcomes for students, they likely have to perceive the humor to be funny in order for it to be effective. In fact, a recent study we conducted across 25 college science courses demonstrated that while funny humor enhanced students’ sense of belonging, attention to course content, and relationships with the instructor, offensive humor decreased students’ sense of belonging, attention to course content, and relationships with the instructor (6). Importantly, humor is subjective and jokes that one person may find funny, someone else may find offensive. While there are certainly individual differences in what someone perceives as funny, there is evidence that people’s social identities and backgrounds are likely to influence whether they perceive certain topics as offensive, especially when the topics are related to social identities (16, 17). This can present a particular challenge in the context of 21st-century science classrooms, where the demographics of instructors typically do not mirror the demographics of students (18–21). What instructors may deem funny based on their own backgrounds and social identities may be quite different than what students perceive to be funny.

There is relatively little research that has explored the link between college student social identities and students’ interpretation of humor. Our research group explored the impact of gender on student perception of funny and offensive topics that college science instructors might joke about in class and identified striking differences between men and women (6). Specifically, men were more likely than women to report that they might find jokes about identity groups (e.g., gender, race/ethnicity) funny if a college science instructor were to joke about them. Conversely, women were more likely than men to report that they might find jokes about identities offensive. To our knowledge, we know of no other studies that have explored the relationship between other student identities and the interpretation of science instructor humor.

In addition to our original study on gender, there are studies outside of the college classroom linking social identities and humor that can provide some insight into how different identity groups (e.g., people of different religions, political affiliations, LGBTQ+ status, races/ethnicities) may react to instructor humor (22–26). These studies suggest that individuals in the same identity group may share a sense of humor or find similar topics funny. Social identity theory considers a social identity to be a person’s knowledge that they belong to a social group, which is defined as a set of individuals who view themselves as members of the same social category (27). Individuals who are similar to oneself are considered the in-group, while people who are different from oneself are considered the out-group. This self-categorization is an important step in identify formation, and as a consequence, one tends to accentuate the perceived similarities between oneself and other in-group members, particularly in attitudes, beliefs, values, affective reactions, behavioral norms, and styles of speech (28). Identity theory provides an important foundation for why students of different demographic groups may interpret humor differently. First, there may be similarities among students of the same identity, or students who perceive themselves to be in the same in-group, with regard to what their group finds funny. Additionally, intergroup discrimination, or discrimination against individuals who are not in one’s in-group, is well documented and motivated by multiple factors including the improvement of one’s self-esteem (27). Therefore, we hypothesize that students may find jokes about their own in-group offensive but perceive jokes about individuals who are not part of their in-group as funny. Additionally, one’s native language can affect how one perceives and interprets humor (29–32). If someone is exposed to humor in another language it may diminish their ability to interpret the humor as intended, particularly ironic humor or humor that uses word play such as puns (31, 33, 34).

Beyond our previous study focusing exclusively on whether males and females differentially perceive certain joke topics as funny and offensive in the context of college science courses, we know of no studies that explore the extent to which undergraduates of different identities (e.g., race/ethnicity, religious affiliation, etc.) differentially perceive specific joke topics as funny and differentially perceive specific joke topics as offensive when delivered by instructors of college science courses. In this study, we set out to explore how student social identities, including race/ethnicity, political affiliation, LGBTQ+ status, religious affiliation, and native language, affect students’ perceptions of instructor humor in the context of college science courses. Specifically, we were interested in whether students of different identities perceived certain potentially humorous topics to be funny and whether they perceived certain topics to be offensive. While we recognize that an instructor’s rapport with students, the delivery of the joke, and the classroom norms and expectations certainly affect students’ interpretation of humor, in this study we were most interested in exploring students’ perceptions of joke topics (17, 35, 36).

Our specific research questions were:

If a college science instructor were to tell a joke about a potentially humorous topic, are there significant differences in the percent of students in different demographic groups who might find it funny?

If a college science instructor were to tell a joke about a potentially offensive topic, are there significant differences in the percent of students in different demographic groups who might find it offensive?

METHODS

This study was done with an approved Arizona State University Institutional Review Board protocol #00005725.

We were interested in investigating the extent to which students perceived particular topics as funny or offensive if they were joked about by a college science instructor. We collected a list of potentially humorous topics by interviewing 95 college students about the most recent joke they heard that was funny and the most recent joke they heard that was offensive. We compiled a list of topics mentioned by at least three students. We also noted that some of the topics were phrased as broader categories (e.g., religious individuals), so we added more specific subsets of these topics (e.g., Mormons, Christians, Catholics, Jewish people, Muslims).

We created a survey with the list of potentially humorous joke topics and a question asking, “If a college science instructor were to tell a joke in class, which of the following joke topics might you find funny?” Students were instructed to check all topics that they might find funny. Then, students were presented with the same list of topics and asked, “If a college science instructor were to tell a joke in class, which of the following joke topics might you find offensive?” Students were then instructed to check all topics that they might find offensive. At the end of the survey, student demographics were collected, including gender, race/ethnicity, political affiliation, LGBTQ+ status, religious affiliation, and native language. A more detailed description of the development of this survey can be found in Cooper et al. (6). A copy of the analyzed survey questions can be found in Appendix 1.

We sent an email to all science instructors at a large research-intensive institution in the Southwestern United States asking them to send our survey to students in their science classes. The survey was conducted at the same institution where we had interviewed the 95 students for the pilot data collection. The instructors offered students a small amount of extra credit in exchange for completing the survey; in cases where instructors were not able to offer extra-credit, students were entered into a drawing to win a $200 gift card. In total, 1,637 students from 25 college science courses completed the survey.

In a previous study using these data (6), we identified a set of 11 topics that students overall were more likely to find funny rather than offensive; we used these topics to test whether there were demographic differences in whether students might find particular topics funny if a college science instructor were to joke about them. Similarly, we identified a set of 16 topics that students were more likely to find offensive than funny; we used these topics to test whether there were demographic differences in whether students might find particular topics offensive if a college science instructor were to joke about them. In this study, we used binary logistic regression to test whether there were differences among students in different identity groups who reported that they might find a particular topic funny and to test whether there were differences among students in different identity groups who reported that they might find a particular topic offensive if a college science instructor were to joke about it. Specifically, we used the identical methodology that we used in our initial paper to detect whether there were gender differences between the topics students perceived to be funny and offensive (6); for each potentially humorous topic, we regressed whether students reported that they might find that particular topic funny (e.g., whether a student perceived jokes about relationships funny) on student identity (e.g., race/ethnicity). An example model is: whether a student perceived jokes about relationships to be funny (Y/N) ~ race/ethnicity. Similarly, for each potentially offensive topic, we regressed whether students reported that they might find a particular topic offensive (e.g., whether a student perceived jokes about women as offensive) on student identity (e.g., political affiliation). All models are reported in Appendix 1 along with the respective regression results. Because we tested whether students in an identity group were particularly likely to find 11 topics funny, we applied the Bonferroni correction for significance at the p < 0.05 level for each comparison. The Bonferroni-adjusted p value needed for significance for all funny comparisons is p < 0.005 (0.05/11). Because we tested whether students in an identity group were particularly likely to find 16 topics funny, the Bonferroni-adjusted p value needed for significance for all offensive comparisons is p < 0.003 (0.05/16). In all regressions, we used the majority group, or the group with the highest number of students in it, as the reference group.

RESULTS

A total of 1,637 students from 25 different college science courses consented to participating in this study. The relevant demographics are displayed in Table 1. Differences between male and female interpretation of humor are reported in Cooper et al. (6). Therefore, in this manuscript, we focused on additional identities including race/ethnicity, political affiliation, LGBTQ+ status, religious affiliation, and native language.

TABLE 1

Demographics of students who completed the humor survey (n=1,637).

Gender
Female	61.3%
Male	37.0%
Other	0.6%
Declined to state	1.0%
Race/ethnicity
Asian	14.6%
Black or African American	4.2%
Latinx	12.5%
Multiple races	11.7%
Other	4.5%
White	49.8%
Declined to state	2.7%
Political affiliation
Democrat	33.4%
Republican	16.4%
Independent	16.2%
Libertarian	5.9%
Not political	18.4%
Other	2.6%
Declined to state	7.0%
LGBTQ+ status
LGBTQ+	8.5%
Non-LGBTQ+	88.9%
Declined to state	2.6%
Religious affiliation
Atheist or agnostic	17.7%
Buddhist	2.1%
Catholic	24.1%
Hindu	1.9%
Jewish	2.1%
Mormon	3.1%
Multiple religions	5.6%
Muslim	6.3%
Nothing in particular	15.5%
Other	6.0%
Protestant	10.2%
Declined to state	5.3%
Native language
Native English speaker	82.8%
Non-native English speaker	17.0%
Declined to state	0.2%

Race/ethnicity

For all analyses regarding student race and ethnicity, we included students who identified as Asian, Black or African American, Latinx, or white. There were too few students who identified as other races/ethnicities (e.g., Native Hawaiian or other Pacific Islander) to include them in the analyses, and grouping students who identify as different races was not appropriate because there is no reason to assume that they would be affected by humor in similar ways. We chose white students as our reference group, since this was the majority of our student population.

Identifying disproportionately funny topics by race/ethnicity

Compared with white students, Asian students were significantly less likely to report that they would find jokes about sports funny if a college science instructor were to joke about them. Black students were significantly less likely than white students to report that they would find instructor jokes about cute animals funny (Table 2).

TABLE 2

The percent of students by race/ethnicity who reported that they might find a particular subject funny if a science instructor were to tell a joke about it.

Potentially Humorous Subject	% of White Students (n=816)	% of Asian Students (n=239)	% of Black students (n=68)	% of Latinx students (n=205)
Science	90.2%	87.9%	86.8%	88.3%
College	87.3%	80.3%	79.4%	83.4%
Television	77.9%	71.1%	76.5%	77.1%
Food puns	70.0%	64.4%	55.9%	71.7%
Relationships	63.8%	63.6%	63.2%	59.5%
Cute animals	60.2%	51.9%	35.3%¶	53.2%
Sports	55.3%	40.6%¶	60.3%	50.7%
Students	52.0%	52.7%	51.5%	48.3%
Politics	49.6%	47.3%	51.5%	48.3%
Sex	45.5%	44.4%	41.2%	45.9%
Farts or poop	32.1%	34.3%	25.0%	38.5%

Significant differences are bolded.

indicates that students of a particular race/ethnicity were significantly less likely than white students to find a particular subject funny if a college science instructor were to joke about it (¶ p<0.005).

We used the Bonferroni-adjusted p value to indicate significance (0.05/11) (p<0.005).

The binary logistic regression model tested for each potentially humorous subject was: (whether student perceived the subject as funny [Y/N]) ~ race/ethnicity.

Identifying disproportionately offensive topics by race/ethnicity

We found that Black students were significantly more likely than white students to be offended by jokes about African Americans. We also found that Latinx students were significantly more likely than white students to be offended by jokes about immigration. Further, we identified that Asian students were significantly less likely than white students to be offended by jokes about Democrats and Republicans (Table 3).

TABLE 3

The percent of students by race/ethnicity who reported that they might find a particular subject offensive if a science instructor were to tell a joke about it.

Potentially Offensive Subject	% of White Students (n=816)	% of Asian Students (n=239)	% of Black Students (n=68)	% of Latinx Students (n=205)
Gender
Women	62.6%	55.6%	67.6%	68.3%
Race/ethnicity/immigration
African Americans	60.9%	56.1%	80.9%*	64.4%
Mexicans	60.0%	53.1%	69.1%	68.8%
Immigration	47.4%	43.9%	52.9%	63.4%*
Religious affiliation
Christians	51.7%	43.9%	64.7%	51.7%
Jewish people	58.5%	53.1%	55.9%	61.0%
Muslims	61.5%	59.4%	75.0%	62.4%
LGBTQ+ status
Gay or lesbian people	59.3%	56.1%	58.8%	65.4%
Transgender people	60.8%	56.9%	58.8%	64.4%
Disability status
People with disabilities	65.4%	58.6%	72.1%	66.3%
Age
Old people	28.1%	28.5%	35.3%	35.1%
Political affiliation
Democrats	41.9%	31.0%¶	39.7%	42.9%
Republicans	39.2%	28.0%¶	25.0%	33.7%
Other
Genitalia	34.4%	27.6%	38.2%	37.1%
Weight	48.2%	41.8%	55.9%	55.6%
Divorce	28.6%	24.7%	27.9%	29.3%

Significant differences are bolded.

indicates that students of a particular race/ethnicity were significantly more likely than white students to find a particular subject offensive if a college science instructor were to joke about it (*p<0.003).

indicates that students of a particular race/ethnicity were significantly less likely than white students to find a particular subject offensive if a college science instructor were to joke about it. (¶p<0.003).

We used the Bonferroni-adjusted p value to indicate significance (0.05/16) (p<0.003).

The binary logistic regression model tested for each potentially offensive subject was: (whether student perceived the subject as offensive [Y/N]) ~ race/ethnicity.

Political affiliation

For all analyses regarding student political affiliation, we included students who identified as Democrat, Republican, Independent, Libertarian, and not political. There were too few students who identified as a member of another political group to include them in the analyses and grouping students who identify as different political groups was not appropriate because there is no reason to assume that they would perceive humor in similar ways. We chose Democrats as our reference group, since this was the majority of our student population.

Identifying disproportionately funny topics by political affiliation

Overall, there were few differences among students of different political affiliations with regard to which topics they might perceive to be funny if an instructor were to joke about them. We observed only two differences: Libertarian students were significantly more likely than Democrat students to report that they might find jokes about sex funny, and students who considered themselves not political were significantly less likely than Democrats to find jokes about politics funny (Table 4).

TABLE 4

The percent of students by political affiliation who reported that they might find a particular subject funny if a science instructor were to tell a joke about it.

Potentially Humorous Subject	% of Democrat Students (n=546)	% of Independent Students (n=266)	% of Libertarian Students (n=97)	% of Not Political Students (n=301)	% of Republican Students (n=269)
Science	89.2%	93.6%	90.7%	90.0%	87.4%
College	85.0%	91.0%	90.7%	80.7%	85.9%
Television	78.4%	78.9%	77.3%	73.8%	77.7%
Food puns	69.2%	71.4%	70.1%	64.1%	68.4%
Relationships	62.1%	63.9%	72.1%	64.5%	64.7%
Cute animals	57.5%	59.0%	58.8%	55.1%	55.8%
Sports	48.4%	56.8%	61.9%	51.2%	58.7%
Students	52.9%	51.5%	66.0%	51.2%	50.6%
Politics	52.4%	58.6%	57.7%	39.9%¶	44.6%
Sex	46.0%	46.6%	61.9%*	40.2%	40.9%
Farts or poop	32.8%	36.8%	33.0%	34.6%	31.2%

Significant differences are bolded.

indicates that students of a particular political affiliation were significantly more likely than Democrat students to find a particular subject funny if a college science instructor were to joke about it (*p<0.005).

indicates that students of a particular political affiliation were significantly less likely than Democrat students to find a particular subject offensive if a college science instructor were to joke about it. (¶p<0.005).

We used the Bonferroni-adjusted p value to indicate significance (0.05/11) (p<0.005).

The binary logistic regression model tested for each potentially humorous subject was: (whether student perceived the subject as funny [Y/N]) ~ political affiliation.

Identifying disproportionately offensive topics by political affiliation

Compared with Democrats, students who identified as Independents, Libertarians, Republicans, and students who consider themselves not political were significantly less likely to be offended by jokes about women, African Americans, Mexicans, immigration, Muslims, gay or lesbian people, transgender people, and Democrats (Table 5).

TABLE 5

The percent of students by political affiliation who reported that they might find a particular subject offensive if a science instructor were to tell a joke about it.

Potentially Offensive Subject	% of Democrat Students (n=546)	% of Independent Students (n=266)	% of Libertarian Students (n=97)	% of Not Political Students (n=301)	% of Republican Students (n=269)
Gender
Women	74.2%	62.8%¶	51.5%¶	53.8%¶	46.8%¶
Race/ethnicity
African Americans	73.3%	59.8%¶	53.6%¶	52.8%¶	49.1%¶
Mexicans	73.8%	57.1%¶	52.6%¶	53.2%¶	49.8%¶
Immigration	63.6%	49.6%¶	42.3%¶	41.9%¶	31.6%¶
Religion
Christians	54.2%	52.6%	43.3%	47.5%	55.0%
Jewish people	67.8%	54.5%¶	53.6%	49.5%¶	48.3%¶
Muslims	75.5%	61.7%¶	53.6%¶	55.1%¶	47.6%¶
LGBTQ+ status
Gay or lesbian people	71.1%	59.0%¶	52.6%¶	51.8%¶	45.0%¶
Transgender people	73.4%	60.9%¶	53.6%¶	51.2%¶	45.4%¶
Disability status
People with disabilities	73.3%	66.2%	54.6%¶	56.1%¶	56.5%¶
Age
Old people	35.0%	29.7%	28.9%	27.6%	22.3%¶
Political affiliation
Democrats	51.3%	36.1%¶	32.0%¶	32.6%¶	31.2%¶
Republicans	33.9%	32.0%	35.1%	33.6%	45.0%*
Other
Genitalia	39.0%	33.5%	32.0%	28.9%¶	32.0%
Weight	57.1%	48.1%	46.4%	42.9%¶	36.8%¶
Divorce	30.4%	28.2%	21.6%	27.9%	27.5%

Significant differences are bolded.

indicates that students of a particular political affiliation were significantly less likely than Democrat students to find a particular subject offensive if a college science instructor were to joke about it. (¶p<0.003).

We used the Bonferroni-adjusted p value to indicate significance (0.05/16) (p<0.003).

The binary logistic regression model tested for each potentially offensive subject was: (whether student perceived the subject as offensive [Y/N]) ~ political affiliation.

LGBTQ+ status

For all analyses regarding student LGBTQ+ status, we included students who identified as lesbian, gay, bisexual, asexual, queer, and transgender in one group, which we call LGBTQ+. Unfortunately, there were too few students in each unique group to include individual identity groups in the analyses, so we decided to group students together as LGBTQ+ because even though students in some groups (e.g., those who identify as transgender) may have experienced more overt discrimination than others, we believed that students who are in this community may share similar marginalized experiences and thus it is reasonable to group them for the analyses (37–39). Students who did not identify within the LGBTQ+ community were the reference group for these analyses.

Identifying disproportionately funny topics by LGBTQ+ status

Students who are part of the LGBTQ+ community were just as likely as non-LGBTQ+ students to find ten of the eleven potentially humorous topics funny. There was only one difference: compared with non-LGBTQ+ students, LGBTQ+ students were significantly more likely to report that they might find jokes about sex funny if a college science instructor were to tell a joke about it (Table 6).

TABLE 6

The percent of students by LGBTQ+ status who reported that they might find a particular subject funny if a science instructor were to tell a joke about it.

Potentially Humorous Subject	% of Non-LGBTQ+ Students (n=1,455)	% of LGBTQ+ Students (n=139)
Science	89.1%	93.5%
College	85.0%	88.5%
Television	76.2%	77.0%
Food puns	66.7%	78.4%
Relationships	62.5%	64.7%
Cute animals	55.1%	66.9%
Sports	52.9%	41.0%
Students	51.3%	58.3%
Politics	48.2%	54.0%
Sex	43.0%	56.8%*
Farts or poop	33.2%	33.8%

Significant differences are bolded.

indicates that LGBTQ+ students were significantly more likely than non-LGBTQ+ students to find a particular subject funny if a college science instructor were to joke about it (*p<0.005).

We used the Bonferroni-adjusted p value (0.05/11) (p<0.005).

The binary logistic regression model tested for each potentially humorous subject was: (whether student perceived the subject as funny [Y/N]) ~ LGBTQ+

Identifying disproportionately offensive topics by LGBTQ+ status

Compared with non-LGBTQ+ students, LGBTQ+ students were significantly more likely to be offended by jokes about women, African Americans, Mexicans, immigration, gay or lesbian people, transgender people, people with disabilities, and jokes about weight (Table 7).

TABLE 7

The percent of students by LGBTQ+ status who reported that they might find a particular subject offensive if a science instructor were to tell a joke about it.

Potentially Offensive Subject	% of Non-LGBTQ+ Students (n=1,455)	% of LGBTQ+ Students (n=139)
Gender
Women	60.3%	78.4%*
Race/ethnicity
African Americans	59.7%	76.3%*
Mexicans	59.3%	76.3%*
Immigration	48.0%	66.9%*
Religion
Christians	52.0%	46.8%
Jewish people	56.5%	69.1%
Muslims	61.3%	75.5%
LGBTQ+ status
Gay or lesbian people	57.0%	79.9%*
Transgender people	57.9%	83.5%*
Disability status
People with disabilities	62.5%	81.3%*
Age
Old people	29.7%	29.5%
Political affiliation
Democrats	39.5%	43.9%
Republicans	36.4%	25.9%
Other
Genitalia	34.0%	36.7%
Weight	46.7%	66.2%*
Divorce	28.5%	28.8%

Significant differences are bolded.

indicates that LGBTQ+ students were significantly more likely than non-LGBTQ+ students to find a particular subject offensive if a college science instructor were to joke about it (*p<0.003).

We used the Bonferroni-adjusted p value (0.05/16) (p<0.003)

The binary logistic regression model tested for each potentially offensive subject was: (whether student perceived the subject as offensive [Y/N]) ~ LGBTQ+

Religion

For all analyses regarding student religion, we grouped students who identified as Catholic, Protestant, and Mormon into a single “Christian” group. We also included Muslims, students who identified as atheist or agnostic, and those who identified with having no particular religious identity. Unfortunately, there were too few students who identified as Buddhist, Hindu, or Jewish to include in the analyses. We also did not include students who identified with multiple religions. We chose Christian students as our reference group, since this was the majority of our student population.

Identifying disproportionately funny topics by religious affiliation

Compared with Christian students, students who identified as atheist/agnostic were significantly more likely to report that they might find jokes about cute animals, politics, and sex funny if a college science instructor were to tell a joke about them. Muslim students were less likely than Christian students to report that they might find jokes about food puns funny (Table 8).

TABLE 8

The percent of students by religious affiliation who reported that they might find a particular subject funny if a science instructor were to tell a joke about it.

Potentially Humorous Subject	% of Christian Students (n=611)	% of Atheist and Agnostic Students (n=290)	% of Muslim Students (n=103)	% of Students Who Identify with No Particular Religion (n=253)
Science	88.2%	92.4%	87.4%	94.1%
College	85.8%	89.0%	74.8%	87.0%
Television	78.2%	75.9%	68.0%	78.3%
Food puns	68.1%	71.7%	44.7%¶	70.8%
Relationships	62.0%	71.0%	59.2%	59.7%
Cute animals	55.5%	65.5%*	43.7%	56.5%
Sports	54.8%	53.8%	47.6%	52.2%
Students	53.5%	60.0%	42.7%	50.6%
Politics	43.9%	62.1%*	50.5%	51.4%
Sex	40.9%	56.6%*	33.0%	41.1%
Farts or poop	29.8%	33.4%	36.9%	35.6%

Significant differences are bolded.

indicates that students of a particular religious affiliation were significantly more likely than Christian students to find a particular subject funny if a college science instructor were to joke about it (*p<0.005).

indicates that students of a particular religious affiliation were significantly less likely than Christian students to find a particular subject funny if a college science instructor were to joke about it (¶p<0.005).

We used the Bonferroni-adjusted p value (0.05/11) (p<0.005)

The binary logistic regression model tested for each potentially humorous subject was: (whether student perceived the subject as funny [Y/N]) ~ religious affiliation.

Identifying disproportionately offensive topics by religious affiliation

Compared with Christian students, atheist or agnostic students were significantly less likely to report that they might be offended if a college science instructor were to tell a joke about Christians, Republications, genitalia, or divorce. Additionally, compared with Christian students, students with no particular religion were less likely to report that they might be offended if a college science instructor were to tell a joke about Christians or genitalia. We also found that Muslim students were significantly more likely than Christian students to be offended by jokes about Muslims (Table 9).

TABLE 9

The percent of students by religious affiliation who reported that they might find a particular subject offensive if a science instructor were to tell a joke about it.

Potentially Offensive Subject	% of Christian Students (n=611)	% of Atheist and Agnostic Students (n=290)	% of Muslim Students (n=103)	% of Students Who Identify with No Particular Religion (n=253)
Gender
Women	62.7%	61.4%	64.1%	59.3%
Race/ethnicity
African Americans	60.7%	58.3%	68.9%	57.3%
Mexicans	61.4%	59.0%	65.0%	55.7%
Immigration	49.3%	50.3%	50.5%	49.4%
Religion
Christians	61.5%	32.8%¶	55.3%	43.9%¶
Jewish people	57.1%	55.5%	56.3%	54.5%
Muslims	60.7%	61.7%	84.5%*	57.3%
LGBTQ+ status
Gay or lesbian people	57.4%	61.4%	51.5%	58.9%
Transgender people	60.2%	62.1%	56.3%	57.7%
Disability status
People with disabilities	67.3%	63.4%	60.2%	60.1%
Age
Old people	32.2%	26.2%	32.0%	26.1%
Political affiliation
Democrats	41.6%	36.2%	36.9%	40.3%
Republicans	39.4%	27.9%¶	29.1%	34.8%
Other
Genitalia	40.6%	26.9%¶	32.0%	28.5%¶
Weight	49.1%	45.5%	41.7%	47.4%
Divorce	34.9%	19.3%¶	25.2%	26.5%

Significant differences are bolded.

indicates that students of a particular religious affiliation were significantly more likely than Christian students to find a particular subject offensive if a college science instructor were to joke about it (*p<0.003).

indicates that students of a particular religious affiliation were significantly less likely than Christian students to find a particular subject offensive if a college science instructor were to joke about it (¶p<0.003).

We used the Bonferroni-adjusted p value (0.05/16) (p<0.003)

The binary logistic regression model tested for each potentially humorous subject was: (whether student perceived the subject as offensive [Y/N]) ~ religious affiliation.

Native language

For all analyses regarding student native language, we grouped students into two groups: those whose native language is English and those who are not native English speakers. Unfortunately, we did not collect specific information about what students’ native language was, only whether it was English or not. However, since the students were taught in English and the jokes referenced were in English, we hypothesized that anyone who is a non-native English speaker may be likely to perceive humor differently since many jokes are based on specific wording or meaning in English (34). Native English speakers were the reference group for these analyses.

Identifying disproportionately funny topics by native language

Non-native English language speakers were significantly less likely than native English language speakers to report that they might find any of the joke topics funny if a college science instructor were to joke about them, with two exceptions: jokes about sex and jokes about farts or poop (Table 10).

TABLE 10

The percent of students by native language who reported that they might find a particular subject funny if a science instructor were to tell a joke about it.

Potentially Humorous Subject	% of Native English Speaking Students (n=1,356)	% of Non-Native English Speaking Students (n=279)
Science	90.8%	82.8%¶
College	86.9%	74.2%¶
Television	77.9%	66.7%¶
Food puns	69.7%	56.3%¶
Relationships	64.5%	52.3%¶
Cute animals	58.6%	43.0%¶
Sports	53.5%	43.4%¶
Students	53.4%	42.7%¶
Politics	49.3%	44.8%¶
Sex	45.4%	36.6%
Farts or poop	32.5%	37.3%

Significant differences are bolded.

indicates that non-native English speaking students were significantly less likely than native English speaking students to find a particular subject funny if a college science instructor were to joke about it (¶p<0.005)

We used the Bonferroni-adjusted p value (0.05/11) (p<0.005)

The binary logistic regression model tested for each potentially humorous subject was: (whether student perceived the subject as funny [Y/N]) ~ native language.

Identifying disproportionately offensive topics by native language

Compared with native English speaking students, non-native English speaking students were significantly less likely to report that they might be offended if a college science instructor were to tell a joke about gay or lesbian people, transgender people, people with disabilities, or Republicans (Table 11).

TABLE 11

The percent of students by native language who reported that they might find a particular subject offensive if a science instructor were to tell a joke about it.

Potentially Offensive Subject	% of Native English Speaking Students (n=1,356)	% of Non-Native English Speaking Students (n=279)
Gender
Women	63.1%	54.5%
Race/ethnicity
African Americans	62.3%	54.5%
Mexicans	61.7%	55.9%
Immigration	49.1%	50.9%
Religion
Christians	52.1%	46.6%
Jewish people	58.3%	52.0%
Muslims	63.4%	57.7%
LGBTQ+ status
Gay or lesbian people	60.8%	49.8%¶
Transgender people	62.2%	49.1%¶
Disability status
People with disabilities	65.4%	55.6%¶
Age
Old people	29.2%	31.5%
Political affiliation
Democrats	41.0%	33.7%
Republicans	37.0%	26.5%¶
Other
Genitalia	34.6%	30.1%
Weight	49.5%	41.6%
Divorce	29.1%	23.7%

Significant differences are bolded.

indicates that non-native English speaking students were significantly less likely than native English speaking students to find a particular subject offensive if a college science instructor were to joke about it (¶p<0.003)

We used the Bonferroni-adjusted p value (0.05/16) (p<0.003)

The binary logistic regression model tested for each potentially offensive subject was: (whether student perceived the subject as offensive [Y/N]) ~ native language.

DISCUSSION

This study highlights that the topics that instructors may joke about in class can be perceived differently by different groups of students.

Humor can be dependent on a shared culture and on language

Overall, it was difficult to discern any patterns with regard to whether particular identity groups were more or less likely to find instructor humor funny. However, non-native English speakers were less likely to find jokes about nearly every topic funny. It is well known that while the use of humor is present across cultures, the extent to which certain types of humor are appreciated can be culture-dependent (29, 31, 33). Unfortunately, we did not collect any information on the culture of these students, so we do not know to what extent cultural differences were responsible for the differences in humor interpretation between native and non-native English speakers. However, studies have also demonstrated that humor is dependent on language, and this can be particularly problematic in the college setting (29, 31, 34). For example, one study found Chinese students studying in the United Kingdom reported not understanding humor in class and struggling to pick up on jokes made by tutors (31). It is important for instructors to note that even though it may seem like non-native English speakers are enjoying or understanding humor, they may not be. Studies have found that non-native English speakers will often pretend to understand a joke to reduce the chances that others will perceive them as less smart (34). In sum, it is important for instructors to realize that while humor can be a powerful tool to build relationships with students and among students, this may not be the case for all students.

Students were significantly more offended by jokes about their own demographic groups

Disposition theory of humor suggests that the way in which someone responds to humor is dependent on how they feel about the target or the “butt” of the joke (17, 40). If someone dislikes or feels animosity toward the target of a disparaging joke, then they may perceive the joke to be funny, whereas if they like or feel empathetic toward the subject of the joke, they may perceive it to be unfunny or even offensive (40). Combined with social identity theory, or the idea that students perceive that they belong to a social group or a set of individuals who view themselves as members of the same social category (27), disposition theory of humor helps explain why students were almost always more likely to be offended by jokes about their own identity group. Specifically, compared with white students, Black students were more likely to be offended by jokes about African Americans, Republicans were more likely than Democrats to be offended by jokes about Republicans and vice versa, LGBTQ+ students were more likely than non-LGBTQ+ individuals to be offended by jokes about gay or lesbian people and transgender people, and Muslims were more likely than Christians to be offended by jokes about Muslims.

Additionally, we acknowledge that students hold multiple different identities and that students’ other identities could influence our findings about whether a particular identity group was more likely to be offended by a particular topic. In this study, we were most interested in how a topic was perceived by students with a particular identity regardless of other identities they hold (however, see Appendix 1 for tables of each identity and the percent of students that also identify as belonging to another identity group). We found that students who identified as LGBTQ+ and students who identified as Democrat were significantly more likely than their respective counterparts to be offended by jokes about most identity groups, particularly marginalized groups. Perhaps LGBTQ+ students in particular perceive that they belong to a broader group of marginalized identities.

In sum, we found that, in agreement with social identity theory and disposition theory of humor, students are more likely to be offended by jokes about identities that they hold. Instructors should recognize that even if they intend for a joke to be innocuous, if it is about an identity, they are most likely to offend students in that identity group.

Limitations

A limitation of this study design is that we did not actually record the jokes that are being told in college science classrooms but instead asked students how they would feel in a hypothetical situation. We acknowledge that the person telling the joke, the delivery of the joke, the specific content of the joke, the context of the class discussion surrounding the joke, and the relationship that the instructor has developed with the students all have potential to impact student perceptions of instructor jokes (17, 35, 36). Not having insight into the specific joke that the students were thinking of is a potential limitation of this study design, although we were hoping for students to think of these categories more broadly than one specific incident. Another limitation is that students may not have personally experienced instructors telling jokes in one or more of the categories, particularly some of the more offensive categories. However, another study has demonstrated that instructors do joke about the potentially offensive topics included in our study; the study recorded over 500 student-reported instances of college instructor humor that they perceived to be inappropriate, and examples included jokes that disparaged people based on gender, race, religion, and sexual orientation (17). Finally, this study was conducted at an institution in the United States, and because humor is dependent on culture and language (41–43), the findings may not be generalizable beyond the United States and more specifically, beyond college campuses in the United States.

CONCLUSION

In this study, we explored whether there were demographic differences with regard to whether college science students perceived potentially humorous topics as funny and offensive. We found that non-native English speakers were less likely to find topics funny if a college science instructor were to joke about them. Further, in alignment with social identity theory and disposition theory, we found that students were most likely to be offended by science instructor jokes about their own identities. This work highlights that while universally funny humor can positively impact students, instructors should be cautious about joking about identity groups in order to avoid offending students with particular demographic traits.

Click here for additional data file.