Digital Library
In the Shadow of War Hotspots of Antisemitism on US College Campuses (Report 1)
Topic:
Antisemitism & Antizionism, Israel & Regional Politics
Principal Investigators:
Graham W Wright, Sasha Volodarsky, Shahar Hecht and Leonard Saxe
Study Date:
2023
Source:
Brandeis University,Cohen Center for Modern Jewish Studies
Key Findings:
This report is the first in a series aimed at identifying evidence-based strategies for effectively responding to antisemitism on campus.
-Antisemitic hostility varies dramatically from one campus to the next. At some schools, the vast majority of Jewish students we surveyed reported that there is a hostile climate toward Jews and Israel on their campus, while at other schools, substantially fewer Jewish students feel this way.
-At the schools with the highest perceived levels of antisemitic hostility, Jewish students were more likely to report experiencing insult or harassment in person and on social media, seeing antisemitic images, slogans, or graffiti, and being blamed for the actions of the Israeli government because they were Jewish. However, the variation between schools with respect to these experiences was significantly smaller compared to the variation in overall perceptions of hostility.
-At all schools, more students were concerned about antisemitism related to criticism of Israel than they were about antisemitism related to traditional anti-Jewish stereotypes. Concern about antisemitism related to criticism of Israel was not limited to those who, in the context of the war, had favorable views of the Israel government: Even among those who had unfavorable views of the Israeli government (44% of all respondents), nearly half (45%) were very concerned about antisemitism related to criticism of Israel.
-In the context of the war, at all schools, Jewish students were substantially more concerned about antisemitism coming from the political left than they were about antisemitism from the political right. Concern about antisemitism from the political left was not limited to politically moderate or conservative Jewish students: 41% of liberal Jewish respondents were very concerned about antisemitism from the political left (54% of all respondents were liberal).
-Jewish students are most likely to encounter anti-Israel hostility from other students on campus, as opposed to faculty or administrators. At the most hostile schools, about 80% of Jewish students reported encountering hostility toward Israel from other students “sometimes” or “often.” At these schools, 30% also reported encountering hostility toward Israel from faculty.
-Jewish students at campuses with higher levels of antisemitic hostility were much less likely to feel fully safe or comfortable on their campus and much less likely to feel that they “very much” belonged. At the same time, and regardless of campus, we found no significant association between the levels of antisemitic hostility on campus and Jewish students reporting that concerns about antisemitism affected their daily lives.
Methodology:
Schools are ranked according to an index of perceived antisemitic hostility, as reported by Jewish students. Campus-level variation is also discussed, with respect to an index of antisemitic incidents reported by survey respondents.
To identify the schools with the most or least hostile antisemitic climate, the researchers could simply compute the average of the antisemitic hostility scale for students at each school and compare those means. However, this approach risks producing biased results at schools where they have only a small number of respondents. To address this problem, they turn to multi-level modeling and use empirical Bayes estimation to generate campus-specific estimates that account for the sample size at each school.
Empirical Bayes estimation “shrinks” or “pulls” the estimate of the mean score (for antisemitic hostility in this case) at each school toward the overall mean (across all schools) in proportion to the number of respondents at that schools. At schools where there is data from many students, the estimate is only shrunk a small amount, reflecting stronger confidence in the data available on that school. At schools where there is data from a smaller number of respondents, the estimate is shrunk substantially, reflecting weaker confidence in the available data at that school.