Getting the Moral Foundations Charts Right
The Righteous Mind Figures are visual representations of Jonathan Haidt's Moral Foundations Theory. They show survey results mapped onto six psychological foundations: Care/Harm, Fairness/Cheating, Loyalty/Betrayal, Authority/Subversion, Sanctity/Degradation, and Liberty/Oppression. Most people encounter them as grouped bar charts comparing average scores across Liberals, Conservatives, and Collectivists (or similar ideological categories). I spent two years working with these figures in an organizational consulting context, trying to map team conflict onto the foundation framework. The charts look deceptively simple. They are not.
Where to Find The Righteous Mind Figures
The original figures come from Haidt's research at the University of Virginia and are published in his 2012 book The Righteous Mind. You can find the primary charts in the appendix of the book and on the Social Intuitions Project website (socialintuitions.org). There are also third-party recreations on Medium, academic slides, and various polarization dashboards. If you want the exact source material, go to the original. The republished versions often get the axis labels wrong or mix up the Likert scale direction. The key download to look for is the Implicit Association Test (IAT) and the Moral Foundations Questionnaire (MFQ-30) result scatter plots. Those are the high-resolution versions researchers actually use.
How to Read the Actual Data
Each foundation is scored on a scale. TheMFQ uses a 1-to-6 Likert scale for importance ratings and a separate set of questions measuring intuitive endorsement. The figures typically display mean scores with error bars representing standard error. When you see a conservative bar taller than a liberal bar on Loyalty, that does not mean conservatives are "more loyal." It means the average respondent in that group endorsed loyalty-related statements more strongly on the survey instrument. Here is a practical detail most people skip: the Liberal profile tends to show a heavy skew toward the two individualizing foundations (Care and Fairness) while scoring noticeably lower on the three binding foundations (Loyalty, Authority, Sanctity). The Conservative profile tends to distribute more evenly across all six. The Collectivist profile, which is less discussed, emphasizes Loyalty and Authority above the rest. These are population-level averages from American samples. Applying them to individuals produces garbage results every single time.
A Common Problem and What Actually Worked
I ran into this issue during a cross-departmental conflict analysis at a mid-size nonprofit. I pulled up the standard Moral Foundations Figures and tried to map two factions onto them. One group was clearly Care and Fairness dominant. The other group scored higher on Authority and Sanctity. I presented this as a diagnosis and it was received poorly. Not because the data was wrong, but because the chart format strips away the situational context that produces those scores. The workaround was to go back to the raw MFQ responses rather than the summarized figures. I pulled each respondent's individual scores and ran a clustering analysis instead of relying on the published aggregate bars. This revealed a subgroup within the "Authority/ Sanctity" faction that actually scored high on Care when the question was framed around institutional duty rather than personal empathy. The published figures would have completely missed that nuance because aggregation flattens it. If you are using these charts for anything beyond casual conversation, extract the underlying data and analyze at the respondent level.
What Beginners Miss
The first thing people get wrong is treating Liberty/Oppression as a standalone foundation. In the MFQ structure, Liberty is actually measured as a reaction against authority imposition, which means it intersects with the Authority foundation in the scoring. You will see some recreations of the figures show Liberty as a separate bar. That is a simplification that distorts the relationship between the two. The original figures from the research treat Liberty as inversely correlated with Authority endorsement in certain contexts, not as an independent dimension. The second thing people miss is the order of the foundations in the charts. Haidt arranged them as Care, Fairness, Loyalty, Authority, Sanctity, Liberty because that reflects the developmental and evolutionary weighting, not alphabetical convenience. When someone rearranges them, the visual argument changes subtly. A chart that puts Loyalty and Authority first tells a different story than one that leads with Care and Fairness. Pay attention to the ordering.
The Limitations
The figures are based on self-reported survey data from primarily American, Western, educated participants. That sample bias is baked into every bar. If you apply these figures to non-American contexts without validating the instrument locally, the results will mislead you. The MFQ has been translated and tested in other languages but the normative profiles shift significantly in collectivist cultures where the binding foundations dominate regardless of political identification. The charts also cannot capture moral foundations that fall outside the six-category model. Subsequent research has proposed additional foundations like Nature/Dignity and the treatment of non-human animals. The original figures do not include these. If your analysis involves environmental ethics or animal welfare framing, the standard Moral Foundations Figures will give you an incomplete picture. For organizational conflict resolution specifically, I found that the figures work best as a starting vocabulary rather than an analytical endpoint. They help people name what they are already feeling. They do not predict behavior well enough to be used as a diagnostic tool on their own. If you need something more predictive, combine the framework with actual behavioral data and situational analysis rather than relying on the chart profiles alone.
The raw MFQ survey instrument and scored data are available through the Social Intuitions Project. The published figures appear in Haidt's book and in the open-access papers that accompany the theory. Everything else is a derivative recreation of varying quality.
Get the Full Details
