When a political argument about Israel turns into an accusation of antisemitism, or the reverse, the question at the center is usually the same: are opposition to Zionism and prejudice against Jews the same thing, or different things that sometimes travel together? That question runs through campus hearings, newsroom debates, and legislative definitions of antisemitism alike. Researchers have now measured the relationship directly, and the answer is that the two attitudes are linked, not as one thing but through a specific channel: in three preregistered US studies, a model in which conspiracy beliefs implicating Israel and Zionists statistically accounted for the association between antisemitism and anti-Israel attitudes explained more than 55% of the variance in those attitudes (Harber and colleagues, 2026, peer-reviewed, abstract-sourced). Zionism is a political movement, support for a Jewish state in Israel; Judaism is a religion, a culture, and a peoplehood that long predates it. This explainer defines each term and reports what the evidence measures about how the two attitudes connect when they are treated as one.
Key Findings
- A psychological model with support across six national samples holds that perceived Jewish power, perceived control over valued resources, is associated with antisemitism through two opposing routes, a hierarchy-threat route and an egalitarian-threat route, which the model offers as an account of why the prejudice appears at both ends of the ideological spectrum (Hadar and colleagues, 2026, peer-reviewed, abstract-sourced).
- A model in which conspiracy beliefs implicating Israel and Zionists mediated the association between antisemitism and anti-Israel attitudes explained more than 55% of the variance in those attitudes; conspiracy beliefs about Jews unrelated to Israel did not carry the association (Harber and colleagues, 2026, peer-reviewed, abstract-sourced).
- The dual-loyalty claim is a minority view but skews young: 10% of US voters ages 18 to 34 agreed with all three antisemitic statements tested, including that Jews are more loyal to Israel than to America, against 2% of voters 65 and older, while two-thirds of all voters rejected every statement (Yale Youth Poll, 2025, academic poll).
- Under IHRA-based coding of an expert-annotated corpus, 62% of antisemitic conversations about Jews invoked the trope of Jewish collective power and 15% the denial of Jewish self-determination, separating the older power trope from the Israel-specific frame (Jikeli and colleagues, 2022, peer-reviewed).
- A large language model rated characters with Jewish names as more dominant, privileged, and hierarchical and as less likable than those with non-Jewish names, trait profiles aligned with historic antisemitic archetypes that persisted despite explicit bias-mitigation measures (Gutman and Gilead, 2026, peer-reviewed, abstract-sourced).
What each term actually names
Zionism and Judaism describe different kinds of things. Zionism is a political movement: support for the establishment and maintenance of a Jewish state in Israel, and in the research it is measured as anti-Israel attitudes and as denial of Jewish self-determination when opposed (Jikeli and colleagues, 2022, peer-reviewed). Judaism is a religion, a culture, and a peoplehood, an identity a person holds regardless of any view on statehood. One is a position on a political question; the other is who someone is.
The two are empirically distinguishable, not merely distinct by definition. A preregistered UK survey using a validated antisemitism scale measured anti-Jewish attitudes (prejudice toward Jews as Jews) and anti-Zionist attitudes as separate constructs, and found conspiracy suspicion correlated with both but more strongly with the Jew-directed form, while the anti-Zionist link was statistically accounted for by its overlap with it (Allington and colleagues, 2023, peer-reviewed). The forms share common ground and remain separable, which is exactly what a definitional distinction predicts.
How conflating them is associated with antisemitism
In the data, the two attitudes are linked through conspiracy beliefs implicating Israel and Zionists. Across three preregistered US studies, a model the authors call the Conspiracies Mediated Model of New Antisemitism accounted for more than 55% of the variance in anti-Israel attitudes (Harber and colleagues, 2026, peer-reviewed, abstract-sourced). The mediator is the detail that does the work. In the model, conspiracy beliefs implicating Israel and Zionists statistically accounted for (mediated) the association between antisemitism and anti-Israel attitudes, whereas conspiracy beliefs about Jews unrelated to Israel did not, and the result was not explained by a general conspiratorial mindset.
The same study measured a partisan asymmetry that resists any single story. Democrats, compared with Republicans, expressed less antisemitism but stronger anti-Israel attitudes and greater endorsement of anti-Israel and anti-Zionist conspiracy beliefs (Harber and colleagues, 2026). The relationship is a measured, mediated association, not a finding that criticism of Israel is in itself antisemitism.
The perceived-power trope that ties them together
The trope that fuses Judaism and Zionism is one of hidden power, and it is associated with antisemitism through two opposite routes. A dual-threat model with converging support across six national samples, the United States, Mexico, Australia, the United Kingdom, Germany, and Poland, holds that perceived Jewish power, defined as perceived control over valued resources, registers among people who favor social hierarchy as a threat to ingroup dominance, and among people who oppose hierarchy as a threat to egalitarian ideals (Hadar and colleagues, 2026, peer-reviewed). That two-route structure is the model’s account of why the prejudice appears at opposite ends of the ideological spectrum.
This matters for the Zionism-Judaism conflation because, in the model, perceived Jewish power is associated with antisemitism through two routes, the same perceived-power belief that the conspiracy-mediation finding above measures. The model is theoretical and cross-cultural, drawn from the abstract of a 2026 paper, and reports converging support rather than a single prevalence figure. As a theoretical structure, it posits that a belief about concealed Jewish or Zionist control is the shared element across both routes, pairing the perceived-power model with the conspiracy-mediation result.
How the dual-loyalty claim shows up in surveys
The dual-loyalty claim, that Jews are more loyal to Israel than to their own country, is the clearest place the conflation appears in attitude data, and it is a minority view that skews young. In a 2025 academic poll of US registered voters, two-thirds rejected all three antisemitic statements tested, including the loyalty item, yet 10% of voters ages 18 to 34 agreed with all three, against 2% of voters 65 and older (Yale Youth Poll, 2025, academic poll). About one in five young respondents also supported boycotting Jewish American-owned businesses to protest the war in Gaza; the boycott item asked about boycotting Jewish American-owned businesses to protest the war in Gaza.
A 2025 advocacy survey measured the experience side of the same year. The American Jewish Committee, an established Jewish advocacy and community-relations organization, reported that 31% of American Jews said they were the personal target of antisemitism in 2025, with younger Jewish adults more likely than older ones to report being targeted (47% of those 18 to 29 versus 28% of those 30 and older) (American Jewish Committee, 2025, advocacy-mission survey). These are self-reported shares within a survey sample, a different kind of evidence from a peer-reviewed measured prevalence, and are labeled as such.
How the dual-loyalty trope appears in coded content
In coded online text, dual loyalty is a real but secondary frame, well behind the older power trope. Under IHRA-based coding of an expert-annotated corpus of English tweets, 62% of antisemitic conversations about Jews fit the IHRA paragraph on mendacious or stereotypical allegations about Jewish collective power, while 15% fit the denial of Jewish self-determination (Jikeli and colleagues, 2022, peer-reviewed). The power trope, not the Israel-specific frame, accounts for the largest share of the coded antisemitism even within conversations about Jews. Here the IHRA framework refers to the International Holocaust Remembrance Alliance working definition, a widely used and itself contested standard that several of these studies adopt as their coding rule.
A separate corpus measured the dual-loyalty dimension directly. Across coded implicit antisemitic terms in two QAnon subreddits, the share of terms associated with each antisemitism dimension was 57% distasteful traits, 56% hidden power, 20.7% dual loyalty, and 13.4% Holocaust minimization, with the categories non-exclusive (Weinberg and colleagues, 2025, peer-reviewed). These are four of several non-exclusive dimensions, not parts of one whole; a single term could invoke more than one. Both corpora rank hidden power above dual loyalty among the power-and-loyalty tropes, a consistent ordering across two platforms and methods.
What latent stereotypes reveal about the encoded trope
The trope is not only in human discourse; it is latent in systems trained on it. Using an indirect method that generated character biographies from Jewish and non-Jewish names and then stripped overt identity markers, a study prompted a large language model to rate psychological and sociocultural traits, and found characters tied to Jewish names were consistently rated as more competent, privileged, dominant, hierarchical, and obsessive, and as less likable and less collectivistic, than those tied to non-Jewish names (Gutman and Gilead, 2026, peer-reviewed). The resulting trait profiles aligned with archetypes historically associated with antisemitic tropes.
The detail that matters for the conflation question is persistence. The pattern held despite explicit bias-mitigation measures applied to the model, which the authors read as evidence the power-and-dominance trope is encoded at a latent level rather than surfaced only by overt prompts. The finding reports the direction and pattern of a measured experimental effect, drawn from the paper’s abstract, not a single prevalence rate.
Methodology and limitations
The findings draw on three kinds of measurement. Survey and experimental studies measure attitudes and the associations between them (Harber and colleagues, 2026; Hadar and colleagues, 2026; Allington and colleagues, 2023; Yale Youth Poll, 2025). Content-analysis studies code samples of online text against the IHRA framework and report shares by trope (Jikeli and colleagues, 2022; Weinberg and colleagues, 2025). One audit study uses an indirect experimental method on a generative model (Gutman and Gilead, 2026). One figure comes from an advocacy-mission survey and is labeled as such, distinct from a peer-reviewed measured prevalence (American Jewish Committee, 2025). Several peer-reviewed entries are sourced from the paper abstract and are flagged in the source list.
Three limits bound every figure. The relationships are correlational: the studies report that attitudes are associated, not that one causes the other, and several authors say so explicitly. The content and survey shares describe specific samples, keyword corpora, two subreddits, and survey panels, not a population, and the QAnon user shares are lower-bound proxies by the authors’ own description. And the coding rests on the IHRA framework, one widely used but contested standard for identifying antisemitism, so a different framework could classify some borderline cases differently. The Institute measures how the Zionism-Judaism conflation operates in attitudes and discourse; it does not adjudicate the contested question of where criticism of Israel ends and antisemitism begins.
Conclusion
So are opposition to Zionism and prejudice against Jews the same thing, or different things that sometimes travel together? The evidence does not rule on the contested boundary between criticizing Israel and antisemitism, but on the narrower question it can measure the answer is clear: the two attitudes are distinguishable yet linked, and they link through one specific channel the data brings into focus.
That channel is conspiracy belief, and the portrait holds across every axis the report measured. The two attitudes connect mainly through conspiracy beliefs implicating Israel and Zionists, a model that accounts for more than 55% of the variance in anti-Israel attitudes, while conspiracy beliefs about Jews unrelated to Israel do not carry the link. The shared element across the routes is a perceived-power trope that registers as a threat to dominance among people who favor hierarchy and as a threat to equality among those who oppose it, the model’s account of why the prejudice surfaces at opposite ideological poles. In attitude data the dual-loyalty claim is a minority view that skews young, held by 10% of voters ages 18 to 34 against 2% of those 65 and older. In coded online text it runs second to the older frame: 62% of antisemitic conversations about Jews invoked Jewish collective power against 15% the denial of Jewish self-determination. And the same power-and-dominance trope sits latent in a generative model, which rated characters with Jewish names as more dominant, privileged, and hierarchical, and held to that pattern even after explicit bias-mitigation measures.
That perceived-power trope does not stay in the survey panel. It lives in everyday talk about who controls the banks, the media, and foreign policy, in the charge that a Jewish neighbor’s real loyalty lies elsewhere, in conspiracy theories about hidden hands that long predate any argument about Israel. Cultural narratives are one of the inputs that shape how a group is seen, which is why the Institute measures them. The measurement has done its work: the conflation of Zionism with Judaism runs through a belief in concealed Jewish or Zionist power, and that belief tracks the same trope the data finds in survey panels, online text, and a generative model alike. The question the figures cannot close is the one a society has to: when this fusion of Zionism and Judaism circulates in ordinary talk, is it one of the quiet currents that feeds antisemitism? That is the question to sit with.
Frequently Asked Questions
Are Zionism and Judaism the same thing?
No. Zionism is a political movement, support for a Jewish state in Israel, and Judaism is a religion, culture, and peoplehood independent of any view on statehood. Measured as separate constructs, anti-Jewish and anti-Zionist attitudes correlate but remain distinguishable, with conspiracy suspicion correlating more strongly with the Jew-directed form (Allington and colleagues, 2023).
How is conflating Zionism and Judaism linked to antisemitism in the data?
The conflation is the measured channel between the two attitudes. A model in which conspiracy beliefs implicating Israel and Zionists mediated the association explained more than 55% of the variance in anti-Israel attitudes, while conspiracy beliefs about Jews unrelated to Israel did not (Harber and colleagues, 2026). The shared element is a belief in concealed Jewish or Zionist power.
How common is the belief that Jews are more loyal to Israel than to their own country?
It is a minority view that skews young. Two-thirds of US voters rejected all three antisemitic statements tested, including the loyalty item, but 10% of voters ages 18 to 34 agreed with all three, against 2% of those 65 and older (Yale Youth Poll, 2025). The survey reports the share agreeing, not a judgment about the claim.
Which antisemitic trope appears most often in coded online content?
Hidden or collective Jewish power, not the Israel-specific frame. Under IHRA-based coding, 62% of antisemitic conversations about Jews invoked Jewish collective power and 15% the denial of Jewish self-determination (Jikeli and colleagues, 2022), and a separate corpus found 56% of implicit terms invoked hidden power versus 20.7% dual loyalty (Weinberg and colleagues, 2025).
Why does the perceived-power trope show up on both the left and the right?
Because it is associated with antisemitism through two opposite routes. A dual-threat model with support across six national samples found perceived Jewish power registers as a threat to dominance among hierarchy supporters and as a threat to equality among hierarchy opponents (Hadar and colleagues, 2026). That structure is the model’s account of why the prejudice surfaces at opposing ends of the spectrum.
Do AI systems reproduce the Zionism-Judaism conflation?
They reproduce the underlying power trope. A large language model rated characters with Jewish names as more dominant, privileged, and hierarchical and as less likable than those with non-Jewish names, profiles aligned with historic antisemitic archetypes that persisted despite explicit bias-mitigation measures (Gutman and Gilead, 2026). The study measured latent stereotypes, not overt output.
Does the research say criticizing Israel is antisemitic?
No study in this evidence base measures that, and several authors caution against the inference. The research measures association, for example that antisemitism and anti-Israel attitudes link through conspiracy beliefs implicating Israel and Zionists (Harber and colleagues, 2026), not a judgment about any individual’s criticism. The boundary itself is a contested scholarly question the Institute does not adjudicate.
Sources
- Allington, Hirsh, Katz, 2023. Correlation Between Coronavirus Conspiracism and Antisemitism: A Cross-Sectional Study in the United Kingdom. Scientific Reports, DOI 10.1038/s41598-023-41794-y. Peer-reviewed.
- American Jewish Committee, 2025. The State of Antisemitism in America 2025. American Jewish Committee, ajc.org/AntisemitismReport2025. Established monitoring organization, advocacy mission, survey.
- Gutman, Gilead, 2026. From Myth to Model: Representation of “the Jew” in Generative AI. American Psychologist, DOI 10.1037/amp0001668. Peer-reviewed, abstract-sourced.
- Hadar, Halevy, Cohen, Apfelbaum, Chan, 2026. The Perils of Perceived Power: The Dual-Threat Model of Antisemitism. American Psychologist, DOI 10.1037/amp0001693. Peer-reviewed, abstract-sourced.
- Harber, Bulska, Malloy, Vila, 2026. Antisemitism, Conspiracy Beliefs, and Anti-Israel Attitudes. American Psychologist, DOI 10.1037/amp0001635. Peer-reviewed, abstract-sourced.
- Jikeli, Axelrod, Fischer, Forouzesh, Jeong, Miehling, Soemer, 2022. Differences Between Antisemitic and Non-Antisemitic English Language Tweets. Computational and Mathematical Organization Theory, DOI 10.1007/s10588-022-09363-2. Peer-reviewed.
- Weinberg, Levy, Edwards, Kopstein, Frey, and colleagues, 2025. Hidden in Plain Sight: Antisemitic Content in QAnon Subreddits. PLoS ONE, DOI 10.1371/journal.pone.0318988. Peer-reviewed.
- Yale Youth Poll, 2025. Fall 2025 Poll on Antisemitic Attitudes and Israel-Palestine. Yale University. Academic poll.