Wisconsin Diversity Opinion Maps (WISDOM)
By Dr. Markus Brauer and Angela Byars-Winston
The WISconsin Diversity Opinion Maps (WISDOM) is an evidence-based tool to counter misinformation, increase data-informed decision-making in policy and practice, and contribute to a better, fairer society where people from all backgrounds can flourish and be healthy by increasing policy and practice informed by data.
At a moment when diversity and inclusion initiatives face growing hostility, having accurate data on the public’s true attitudes toward these issues has never been more critical to building an equitable society. Currently, no reliable data exists on diversity-related opinions, and research shows that most Americans significantly underestimate how much support actually exists for diversity and inclusion efforts. As a result, decision-makers are making consequential choices with far-reaching societal impact based on incomplete or misleading information.
Our goal is to change that by gathering trustworthy data that captures genuine attitudes and beliefs about diversity and inclusion across every region of the United States. We then present these findings in a way that is simple, accessible, insightful, and interactive for the public to explore.
Diversity Maps
Methodology
This site offers detailed estimates of American attitudes toward diversity-related issues, covering general attitudes, self-reported behaviors, and policy beliefs. It serves as a high-resolution source of public opinion data capable of guiding decision-making, policy development, and educational efforts at the national, state, and local levels. These estimates are generated through a statistical technique called multilevel regression with post-stratification (MRP), which draws on a large-scale national survey combined with demographic and geographic population data.
The modeling employs data from a nationally representative survey of American adults (18+) conducted in July 2026 with a sample size of N = 4,775. The survey was administered by Ipsos and drawn from the Ipsos KnowledgePanel®, an online panel of members drawn using probability sampling methods. All questionnaires were self-administered by respondents in a web-based environment, and computers were loaned to individuals who were chosen to participate but did not have access. Respondents hailed from all 50 states and the District of Columbia, and (insert # here) counties.
Survey Question Wording: Diversity-Related Opinions
Model estimates in the maps were derived from public responses to the following survey questions. For each question, the response categories were analyzed on its own original scale, not collapsed to agree/disagree first. Six questions really are Yes/No, and those are reported as such. One is a 0–100 slider, and its full distribution — spike at “Both equally” included — is shown, not thrown away. The other 18 are a 6-point agree/disagree scale, and their full 6-category distributions are charted
General Attitudes and Beliefs
Broad measures capturing how respondents generally think about diversity-related issues, across a range of groups and topics.
Is discrimination against each of the following groups a serious problem in the US today? Racial/ethnic minority groups, Women, Religious minorities (for example Jews, Muslims), People with disabilities, LGBTQ people (Note: LGBTQ = lesbian, gay, bisexual, trans, and queer), Immigrants
- Yes
- No
Diversity of employee backgrounds – including race, gender, age, and life experiences – helps companies be more successful.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
White people face as much discrimination as people of color in the U.S. today.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
What are racial disparities in education and wealth due to? [Please use the slider to indicate your response] ***need to figure out what to do for this question***
──────────────────●─────────────────
| Individual factors (e.g., lack of effort, lack of self-control, low motivation) | Both equally | Societal factors (e.g., unequal access to resources, residential segregation, ongoing discrimination) |
Policies and Programming
Support for specific policies and programs addressing diversity-related issues, across a range of domains.
I am in favor of diversity and inclusion programs that have been shown to be effective in rigorous evaluation studies.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
We should fund more research on the best methods to reduce disparities between groups defined by race, gender, and other demographic characteristics.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
Many diversity training workshops blame White people or try to make them feel guilty.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
I am in favor of policies that protect members of minority groups from discrimination.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
It is important to develop affordable housing options in middle-class neighborhoods to promote racial integration.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
Public schools should not be required to teach the history of racism and discrimination in the U.S.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
I support policies aimed at reducing racial disparities in policing.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
I support policies aimed at reducing racial disparities in health outcomes.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
I support policies that promote equality of access to employment regardless of race, gender, religion, disabilities, and sexual orientation.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
Behaviors
Measures of respondents’ own diversity-related behaviors and personal reactions in situations involving race or diversity.
I frequently try to create a welcoming environment for individuals from other racial groups in my community.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
I would confront a friend if they discriminated against someone else.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
I would be hesitant to apply for a job at a company that is known to be biased against a particular social group defined by race, gender, or another demographic characteristic.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
I feel uncomfortable interacting with people whose race or cultural background is different from my own.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
Who Should Act
Measures of who respondents believe bears responsibility for addressing diversity-related issues.
Elected officials should put forth legislation that ensures equal access to personal and professional opportunity.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
Leaders in organizations should be held accountable for fostering an inclusive workplace culture and achieving measurable diversity goals.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
I am bothered by the projection that White people will eventually be a numerical minority in the U.S.
- Strongly agree
- Agree
- Somewhat agree
- Somewhat disagree
- Disagree
- Strongly disagree
Frequently Asked Questions
What do these maps depict?
These maps show estimated percentages of American adults (18 and older) who hold specific beliefs, attitudes, and policy views related to diversity issues. Rather than relying on raw survey data alone, the estimates come from a statistical model that blends survey responses with U.S. Census demographic data. This approach allows the model to project opinions across different population groups based on factors like gender, race and ethnicity, and educational background, while also accounting for shifts in public sentiment over time.
Note: This methodology draws inspiration from the Yale Climate Opinion Maps 2025, which pioneered a similar approach for estimating localized public opinion by combining survey data with demographic modeling.
Where does the survey data come from?
The underlying dataset contains a sample size of 4,775 participants collected in July 2026. Participants were recruited via KnowledgePanel. KnowledgePanel, operated by Ipsos, is a premier, probability-based online research panel designed to be nationally representative of the U.S. population. Members were randomly recruited via address-based sampling, ensuring valid, scientific insights. Eligible panelists were invited to complete a voluntary online survey after providing informed consent. ***(check if it’s this or the combined sample size)***
How large was your survey sample in my state or county? If no one from my area was surveyed, how can you draw conclusions about opinion there?
The number of survey respondents differs by geographic unit, but totals a very large number nationwide. Because our surveys are designed to be nationally representative, respondent counts in each state and county reflect that area’s share of the total U.S. population; meaning more populous states and counties include more respondents than smaller ones. For some smaller counties, only a handful of respondents (or none at all) may have been surveyed directly. In these instances, our model draws on county population data, patterns from other counties within the same state, and relevant county characteristics (such as past voting behavior) to generate estimates.
By understanding a county’s (or other unit’s) political leanings alongside its demographic makeup, we can predict residents’ views with strong accuracy. Incorporating additional variables, such as rural versus urban setting or other local characteristics, further sharpens these predictions. Furthermore, for each geographic unit, our model pools information from neighboring areas and broader regional data (for instance, state-level patterns inform estimates for every county within that state).
This layered approach allows us to generate reliable estimates for all counties, even those where direct survey sampling was minimal.
How accurate are the estimates?
No model is perfect and there are uncertainties in the model estimates. ***fill in this section***
Do the maps account for differences in population density across the country?
***verify if this is true***
No, the maps illustrate the estimated share of people within each geographic area who would respond a given way to each question, and they are not adjusted to account for differences in population density. It’s worth keeping in mind that some areas may cover a large land area while having relatively few residents (e.g., Montana), whereas others may be geographically small yet densely populated (e.g., Rhode Island).
Can I use this data?
Yes. We welcome and encourage the use of these maps and their findings in independent research and analysis. The underlying data may be accessed via the Data Download tab at the top of this page, enabling users to conduct their own analyses and develop custom visualizations. Researchers who use this data in an academic publication are requested to acknowledge the source using the following citation:
*citation here*
For any news article, visualization, blog post, or other publication that makes use of this data or these maps, please include a link to the WISconsin Diversity Opinion Maps (WISDOM) website, along with attribution to the Institute for Diversity Science.