Conducted by NORC at the University of Chicago
In partnership with the One8 Foundation
Antisemitic Attitudes in America: Topline Findings
ADL has measured antisemitic attitudes among Americans since the early 1960s. Building on this historic work and furthering it to ensure greater accuracy, ADL, the National Opinion Research Center at the University of Chicago (NORC) and the One8 Foundation embarked on a year-long process to study the research literature on bias and antisemitism, convene academic and communal leaders and conduct qualitative interviews.
Based on the learnings from this process, ADL updated its ways of measuring antisemitism to develop a more nuanced suite of questions that provide greater understanding. ADL then used these upgraded measurement tools to survey over 4,000 individuals, a representative sample of the American population from September through October of 2022. The topline results, presented in this report, show several trends that are cause for concern:
Widespread belief in anti-Jewish tropes, at rates unseen for decades
Over three-quarters of Americans (85 percent) believe at least one anti-Jewish trope, as opposed to 61 percent found in 2019. Twenty percent of Americans believe six or more tropes, which is significantly more than the 11 percent that ADL found in 2019 and is the highest level measured in decades.
Substantial rates of Israel-focused antisemitism
Many Americans believe in Israel-oriented antisemitic positions – from 40 percent who at least slightly believe that Israel treats Palestinians like Nazis treated the Jews, to 18 percent who are uncomfortable spending time with a person who supports Israel.
Trope-focused and Israel-focused antisemitism appear to overlap significantly
There is a nearly 40 percent correlation between belief in anti-Jewish tropes and anti-Israel belief, meaning that a substantial number of people who believe anti-Jewish tropes also have negative attitudes toward Israel.
Young adults have more anti-Israel sentiment than older generations, and only marginally less belief in anti-Jewish tropes
While young adults (between the ages of 18 and 30) show less belief in anti-Jewish tropes (18 percent believe six or more tropes) than older adults (20 percent believe six or more tropes), the difference is substantially less than measured in previous studies. Additionally, young adults hold significantly more anti-Israel sentiment than older adults, with 21 percent and 11 percent agreeing with five or more anti-Israel statements, respectively.
This report is the first in a series on anti-Semitic attitudes in the United States from this survey, with future reports covering topics including differences in anti-Semitic attitudes across the political and ideological spectrum; antisemitism among different racial and ethnic groups; views of Jews as privileged; similarities and differences between biases against Jews and other groups; and key drivers of belief in anti-Jewish tropes and anti-Israel sentiment.
In 1964, ADL began measuring to what extent Americans believe certain antisemitic tropes through a detailed quantitative survey conducted in partnership with researchers at the University of California Berkeley and the National Opinion Research Center at the University of Chicago (NORC). The index used by ADL focused on gauging belief in particular historic anti-Jewish tropes or stereotypes, such as whether “Jews stick together more than other Americans,” “Jews have too much power in the business world” and “Jews have a lot of irritating faults.”
As a result of the rise in antisemitic incidents over the past several years, as measured by the ADL Audit, ADL created a new research center focusing on antisemitism. Further, as antisemitism in the United States evolves to include not just traditional, authoritarian versions of antisemitism but also newer forms of antisemitism, including anti-Zionism, ADL identified the need for more comprehensive research methods.
Over the past year, ADL, in partnership with the One8 Foundation and NORC, undertook an effort to evolve its survey to include different forms of antisemitism that were not captured in prior research. The new survey augmented existing questions by probing respondents’ sentiments toward Israel and its supporters, investigating societal perceptions of Jewish stereotypes, ascertaining respondents’ previous encounters and experiences with Jews, and conducting novel experiments that attempted to expose anti-Jewish biases respondents may not have been willing to share. As a result, ADL now has several batteries of questions that together create a more comprehensive picture of antisemitic attitudes in the United States.
Researchers used this new instrument to survey 4,007 respondents in September through October of 2022. This report is the first in a series on antisemitic attitudes in the United States from this survey. It details topline findings on belief in anti-Jewish tropes and anti-Israel sentiment. The report also includes an analysis of the ways in which anti-Jewish attitudes and anti-Israel sentiments converge and diverge, illuminating several new avenues for understanding antisemitism. Future reports over the coming months will cover topics including differences in antisemitic attitudes across the political and ideological spectrum; antisemitism among different racial and ethnic groups; views of Jews as privileged; similarities and differences between biases against Jews and other groups; and key drivers of belief in anti-Jewish tropes and anti-Israel sentiment.
Belief in Anti-Jewish Tropes
Respondents were asked to rate the truthfulness of 14 statements describing different traditional anti-Jewish tropes. These tropes reference common anti-Jewish conspiracies theories that previous research has shown lead to hostility and violence. As seen in the chart, researchers found significant levels of support for these tropes. Some of these statements view Jews as “clannish,” with 70 percent and 53 percent of Americans saying that Jews stick together more than others and go out of their way to hire other Jews, respectively. Other tropes relate to the concept of “dual loyalty,” with 39 percent of Americans saying that Jews are more loyal to Israel than the United States. Finally, we see over 20 percent support for several statements relating to Jews being too powerful in business and Wall Street.
Researchers also looked at the overall number of different anti-Jewish tropes that people believe. The chart below shows the percentage of Americans that selected either “Somewhat True” or “Mostly True” for a certain number of anti-Jewish tropes. As this graph demonstrates, 85 percent of Americans think at least one trope is somewhat true. Approximately 20 percent of Americans (corresponding to around 52 million people) believe in six or more of the 11 anti-Jewish tropes that ADL has measured since 1964, which in this index are counted among those who hold significant antisemitic attitudes. All 11 tropes are believed by three percent of the population, which corresponds to approximately 8 million people - more than the number of Jews in the United States.
Researchers compared the share of Americans who believe in six or more anti-Jewish tropes over time, with the caveats discussed in the methodological appendix. The share of Americans who believe six or more tropes appears to be significantly greater than measured in the previous study – from 11 percent in 2019 to 20 percent in 2022.
Researchers observed greater belief in anti-Jewish tropes among young adults (ages 18-30) in the 2022 survey than in prior research. While younger adults have modestly lower rates of believing in tropes than older Americans, this difference is far less than previously observed. For example, in 1992, ADL found a 19-percentage point gap between those under 40 and those over 40. Indeed, one of that study’s major findings was that “the steady influx of younger, more tolerant Americans into the adult population” had led to an overall decrease in antisemitism.
Israel-Focused Antisemitism and Sentiment
ADL has seen the ways in which criticisms of Israel can exceed policy critiques and instead morph into traditional anti-Jewish conspiracy theories and antisemitic tropes as well as be weaponized to malign or increase hostility toward Jews generally. As a result, researchers adapted measures of an Israel Sentiment Index, presenting eight statements about Israel and its supporters and asking respondents whether they agree with them on a 6-point scale. Just like the anti-Jewish tropes, the statements vary in kind and intensity, creating a broad measure of Israel sentiment, not solely measuring only extreme forms such as anti-Zionism.
The below chart shows support for each statement in the Israel Sentiment Index. Notably, 90 percent of Americans believe Israel has a right to defend itself against those who want to destroy it. Further, 79 percent of Americans see Israel as a strong ally of the United States. However, negative, antisemitic sentiments toward Israel are held by a broad swath of the American public – from 40 percent who believe, at least slightly, that Israel treats Palestinians like Nazis treated the Jews, to 18 percent who are uncomfortable spending time with a person who supports Israel.
Just as researchers examined the number of different anti-Jewish tropes that people believe in, researchers also looked at the different number of the eight statements about Israel that people agree with. The chart below shows the percentage of Americans that selected either “Slightly Agree,” “Agree,” or “Strongly Agree” for a certain number of negative statements about Israel.
Unlike with traditional anti-Jewish tropes, researchers found that young adults believed in significantly more anti-Israel statements than older adults. Twenty-one percent of young adults agreed with five or more of the eight anti-Israel statements – almost double the 11 percent of older adults.
The Relationship Between Antisemitism Revealed by Classical Tropes and Israel Sentiment
Researchers also wanted to explore the relationship, or correlation, between antisemitism focused on Jewish tropes and antisemitism focused on Israel. High correlation would mean that those who believe in classical anti-Jewish tropes are likely to also express anti-Israel forms of antisemitism, and vice versa. In contrast, low correlation means that the group of people who believe anti-Jewish tropes are largely separate from the group of people who express Israel-focused forms of antisemitism. Such analysis is important for predicting whether interventions that target trope-focused antisemitism are likely to also reduce Israel-focused antisemitism, and vice versa. Similarly, correlations help inform our prediction of whether increased belief of classical anti-Jewish antisemitism will lead to greater anti-Israel antisemitism.
To explore this relationship, the charts below show the percentage of anti-Jewish and anti-Israel statements believed by the average American, along with the degree to which they are correlated. Researchers found a substantive correlation (0.38) between belief in anti-Jewish tropes and anti-Israel sentiment across all respondents. Older adults have a much higher correlation (0.43) than we find in younger adults (0.28).
This topline report shows that antisemitic attitudes in the United States are widespread and likely increasing: 85 percent of Americans believe at least one anti-Jewish trope, as opposed to 61 percent in 2019. Furthermore, 20 percent of Americans believe six or more anti-Jewish tropes, substantially more than the 11 percent found in 2019. Unlike in years past, researchers found that young people seem to hold similar levels of belief in anti-Jewish tropes compared to older adults. This report also shows that anti-Israel sentiment, including anti-Israel sentiment rooted in antisemitic conspiracy theories, is held by broad swaths of the population. Analysis further revealed that while belief in traditional anti-Jewish tropes and anti-Israel sentiment are discrete, they overlap in substantial ways.
Over the coming months, ADL will release a series of reports outlining key findings from its 2022 national survey. These reports will probe deeply into antisemitic attitudes among different subpopulations; views of Jewish stereotypes; similarities and differences between prejudices against Jews and other groups; and key drivers of belief in anti-Jewish tropes and anti-Israel sentiment.
The sample contains 4,007 respondents from the National Opinion Research Center’s AmeriSpeak panel surveyed from September through October of 2022. Since its founding by NORC at the University of Chicago in 2015, AmeriSpeak has produced more than 900 surveys, been cited by dozens of media outlets and become the primary survey partner of the nation's preeminent news service, The Associated Press. AmeriSpeak was chosen for this study because of its scientifically rigorous panel and its strong representation of hard-to-reach populations, including low-income households, less educated persons, young adults, rural households, persons who are less interested in the news, and social and political conservatives.
Indeed, careful attention was paid to ensure the probability sample included a broad swath of respondents from a range of socio-economic, political and ethno-racial backgrounds. This is a weighted, representative sample of Americans generally, and of the sub-populations that researchers oversampled: those between the ages of 18 and 30 (1,292 respondents), those on the political Right (420) and political Left (663), Black Americans (578) and Hispanic Americans (626).
The study was registered with the University of Chicago’s Institutional Review Board for the Ethical Treatment of Human Subjects. The data are weighted to represent the general population of the United States in addition to the aforementioned subpopulations using the benchmarks of the demographic profile of the National Opinion Research Center’s AmeriSpeak panel and its 2021 General Social Survey.
ADL and its partners seek to ensure the data and methodology are transparent and shared with the public. To that end, the data will be released through ICPSR at the University of Michigan.
The survey questionnaire was designed as a large-scale collaboration between the staff at the ADL Center for Antisemitism Research, NORC, the One8 Foundation, Jewish communal and civil rights leaders, and an academic advisory board of scholars. Wherever possible, researchers incorporated questions that had been asked and validated elsewhere (such as the statements related to Israel) or had been asked previously by ADL (such as the ADL Index of anti-Jewish tropes).
Decades-long studies, such as the ADL Index, must balance continuity around measures while updating methods to ensure greater rigor and modern practices. Researchers opted to use the same 11 classic statements that probe, in particular, authoritarian anti-Jewish conspiratorial belief. In addition, researchers added back in a predictive statement that, while asked in the original 1964 study, had been dropped more recently. Further, researchers added two positively phrased statements in order to ensure greater rigor in the presentation of the attitudinal battery.
For the set of anti-Jewish tropes that have been asked in some form since 1964, comparisons in survey responses over such a long period of time are never perfect given that sampling techniques, modes of administration, societal understandings of terms, and norms in responding to surveys change over time. Additionally, slight changes were also made to the response options presented to respondents and to question wording in line with best practice of surveys. For this survey, the primary goal was to ensure that researchers accurately capture antisemitic attitudes in America.
In some cases, researchers made changes based on methodological best practice. For example, for the current survey, researchers opted to remove the “Unsure/Don’t Know” option for anti-Jewish tropes, considering it a crucial adjustment to the rigor of the study. While it is important to retain the “Unsure/Don’t Know” option for fact-based questions (some people just do not know a fact), for opinion-based questions, it is broadly better to compel some meaningful selection.
Additionally, over time, the mode of surveys has generally changed. For example, in ADL’s 2019 survey, researchers primarily opted for internet distribution rather than person-to-person interviews because research has shown that internet administration generally reduces what may be people’s reluctance to admit to prejudiced views. This is referred to as social desirability bias. Indeed, taken altogether, the changes to the index were designed to reduce social desirability bias, likely revealing some beliefs that otherwise may have gone undetected.
Below shows the share in belief of specific tropes at different points of survey administration, including the first and most recent.
Researchers also conducted Principal Component Factor (PCF) analysis of the anti-Jewish tropes and anti-Israel sentiments to examine, among other things, whether the statements included in each index measure the same underlying phenomenon. As seen in the below factor loadings, the statements for each of the indices score highly, showing that each index is composed of complementary statements. This shows that while some readers might think that Israel being a technology leader is not a measure of anti-Israel sentiment, the factor loadings indicate that those who hold anti-Israel beliefs do find it relevant.
This report benefited from the contributions of many people, including the team at the National Opinion Research Center, specifically, David Dutwin and Martha Stapleton. Significant contributions were also made by Emma Tsurkov, Jacob Dennen, Ayal Feinberg, Karyn Cohen Leviton, Alyssa Arens, Laura Royden, Michael Zanger-Tishler and Huseyin Emre Ceyhun. The researchers would like to thank the research advisory board and reviewers, including: Eitan Hersh, Alan Cooperman, Susan Fiske, Joshua Kalla, Brendan Nyhan, Donald P. Green, and Ilana Horwitz. The board offered insight into and criticism of the design, development, analysis, and reporting of this study; however, the content of the report is the authors’ alone. Additional thanks to the Crimson Lion/Lavine Family Foundation, the ADL Lewy Family Institute for Combating Antisemitism, the Diane and Guilford Glazer Foundation and the Rosalinde and Arthur Gilbert Foundation.
 Adorno, Theodor, Else Frenkel-Brenswik, Daniel J. Levinson, and R. Nevitt Sanford. The authoritarian personality. Verso Books, 2019.
 Respondents were administered this battery of statements in a randomized order with the rule that the first statement shown always had to be one of the three positively phrased statements, so as to ensure that both sequence and statement valence did not bias results.
 ADL Antisemitism Survey. May 1992. 37.
 Researchers adapted questions from Allington and Hirsh’s Antizionist Antisemitism Scale. Allington, Daniel and Hirsh, David. "The AzAs (Antizionist Antisemitism) Scale: Measuring Antisemitism as Expressed in Relation to Israel and Its Supporters" Journal of Contemporary Antisemitism, vol. 2, no. 2, 2019, pp. 43-52. In line with the anti-Jewish trope statements earlier, researchers reverse coded the positive statements for ease of interpreting analysis results. Statements presented to respondents varied in the strength of the negativity they conveyed, and in their focus on Israel versus its supporters. These eight statements are not meant to test knowledge or adjudicate criticism of Israel as legitimate, accurate or not. Rather, the Israel Sentiment Index is primarily meant to measure different dimensions of sentiment towards Israel. Accordingly, the statements are not meant to be as closely related to each other as possible, but, to the contrary, they are geared towards finding how different can these statements be while still measuring the same underlying social phenomenon.
 Inclusion of a statement in the index is not meant to imply that any one statement is strictly a measure of sentiment towards Israel, as opposed to anti-Jewish prejudice. Finally, given the low awareness and debates around the definition of Zionism and anti-Zionism, researchers avoided using these terms.
To measure whether these different statements do in fact measure the same underlying phenomenon, Principal Component Factor (PCF) analysis was conducted. For a more detailed explanation of PCF and its application to the anti-Jewish and anti-Israel sentiment, see the appendix.
 A Pearson Correlation coefficient can take any value between 1 and -1. The further away the coefficient is from zero, the stronger the relationship. 1 (or -1) means perfect correlation. The closer the correlation coefficient is to zero, the weaker the relationship. A correlations coefficient of zero means the two variables are independent of each other. Generally, any correlation above 0.5 suggests high correlation, and between 0.25 and 0.5 suggests moderate correlation.
 See, e.g., Croucher, Stephen M., Elvis Nshom Ngwayuh, Diyako Rahmani, and Cheng Zeng. "Social Desirability Bias among Prejudice Instruments: An Integrated Threat." Journal of Intercultural Communication 2019, no. 50 (2019). Also: Pew Research Center. Race in America 2019.
 To measure whether these different statements do in fact measure the same underlying phenomenon, Principal Component Factor (PCF) analysis was conducted. Sampling adequacy was measured as well as internal reliability of score, whether agreement with these eight anti-Israel statements reflects a single underlying latent dimension, and the degree to which each of the tropes influence the factor. The Kaiser-Meyer-Olkin measure of sampling adequacy is 0.92 for the entire scale and more than 0.9 for every variable (every anti-Israel statement) individually. This means that analytically, the data is suitable for factor analysis of all the statements together and that each variable individually is suited for inclusion in the factor.
The internal reliability of the scale, measured using Cronbach’s Alpha, is 0.88 on a scale of zero to one, where zero means no internal reliability. A Cronbach’s alpha score of 0.7 and above is considered good in the social sciences.
The first, and best, factor has an eigenvalue of 3.9, while the second has an eigenvalue of 0.18. This can be interpreted to mean that the eight statements used in the index share an underlying phenomenon, and only one.
 The factor loading is the correlation between the item and the factor. It can range in value from 1 to -1. The further away a value is from zero, the stronger is the justification to include the corresponding variable in the index. A factor loading of more than 0.30 indicates a moderate correlation between the item and the factor and is commonly used as the cutoff for inclusion.
 Unique variance represents the variation in the item that reflects unknown latent causes, and random error due to measurement error.