Is One Dimension Enough for Understanding the Political Spectrum?

Suppose you have a friend who is all about environmentalism and social justice, but also thinks businesses should be free from most government regulation. Or maybe you’ve heard about someone deeply patriotic, proud of religious traditions, but believes in wealth distribution and universal healthcare. These are the kinds of “mixed” positions that can get lost when we assume politics only runs from Left to Right.

For decades, scholars have treated liberal-to-conservative as the “big lane” where policy preferences fall. This tradition, rooted in works like Anthony Downs’s An Economic Theory of Democracy (1957) and in later studies like Carmines and Stimson’s Issue Evolution (1989), emphasizes how core cleavages, like the role of government in economics or social policy, can line up neatly on a single axis. Even modern surveys (e.g., the American National Election Studies) typically anchor participants’ responses along a Left–Right spectrum. But real people’s beliefs aren’t always so neat and orderly. While the Left–Right line remains a good quick measure, it can smooth over the complexities of human values and group identities. That’s where adding other relevant dimensions can reveal patterns we’d otherwise miss, and allow us to see, for instance, how strongly an individual’s outlook leans toward collective solutions (group-focused, structural changes) or individual autonomy (personal freedom, minimal regulation).

Beyond Left and Right: Mapping What People Actually Believe

At CARE, we recognized these complexities and developed a more detailed approach that measures how individuals endorse or reject a range of issue-specific positions, rather than relying on their self-reported ideological placement, so we can capture a more accurate picture of people’s beliefs. That’s the starting point for our more detailed approach to measuring political ideology: rather than having people declare “I’m liberal” or “I’m conservative,” we collect their views on issues that are commonly linked to different ends of the spectrum, asking them to endorse their top 3 choices. Examples of these include:

• Reducing foreign entanglements, focusing on national borders and economic interests.
• Promoting self-determination and cultural reclamation for formerly colonized peoples.
• Cleaning up covert and corrupt influences on the federal government.
• Wealth redistribution, universal healthcare, guaranteed basic income.
• Free-market capitalism, reduced taxes, minimal government interference.
• Ensuring transparent, secure, and fair elections.
• Aggressive action on climate change, renewable energy investment.
• Minimizing government intervention in personal and economic affairs.
• Preserving cultural standards, prioritizing citizens over immigrants.
• Advocating for traditional moral frameworks in American culture.
• Protecting gun ownership with minimal restrictions.
• Supporting anti-racism, LGBTQ+ Rights, and reproductive Rights.
• Upholding conventional family structures and moral standards.

We adapted these issues from an existing typology (Ramaciotti et al., 2024), and took an additional step in restricting respondents to their top three choices, so that we force prioritization, highlighting tradeoffs (making hard choices) and capturing a clearer hierarchy of their strongest convictions.

Step One: Reimagining the Left–Right Axis

Instead of saying, “Where do you see yourself on a Left-to-Right political scale” we look at the degree to which people endorse or reject each of these actual issue statements. This makes a big difference even for the singular Left to Right dimension, getting out of self-perception territory and into what people’s most important beliefs actually are. You might support certain “America Left” or “Right.” We sum up the “Left-leaning” issue scores for each person (like supporting environmentalism or social justice), sum up the “Right-leaning” issue scores (like advocating for minimal government interference or emphasizing religious/traditional values), and subtract one from the other – creating a single “composite” measure, wherea large positive number suggests a more Right-leaning orientation, and a more negative number is tilting more to the Left. In Figure 1, you can see how the composite score creates some natural clusters along the Left-to-Right spectrum. 

Figure 1: Histogram of Composite Scores with Clustering Boundaries

Of course, just having a continuous score from Left to Right might still feel like a single-lane highway—what if you want to highlight distinct groups, like “Far Left,” “Moderate Left,” “Moderate Right,” and “Far Right”? This is where k-means clustering comes in. It’s a method that automatically searches for natural groupings in the data (by minimizing the variation within each cluster and maximizing the variation between clusters). Instead of arbitrarily deciding, “Well, anyone over 0.75 is ‘Far Right,’” we let the data tell us where those boundaries should fall.

We take this a step further by defining those boundaries more clearly using a clustering algorithm (k-means). This approach for defining thresholds and groups helps to avoid guesswork about where “moderate” ends and “extreme” begins, and finds meaningful categories that reflect genuine patterns in the data. In our case, it happens to align well with common-sense notions of political leanings—if someone’s strongly pro-environment, pro-social justice, and skeptical of large corporations, they land in a more Left-leaning cluster. Essentially, we are letting the data show us where people naturally group instead of assuming they fit neatly in Left or Right boxes, or having them self-sort into those. 

By combining actual endorsements of policy stances with clustering, we get a richer narrative than a simple “Which side do you identify with?” We uncover folks who are overall to the Left but also hold a few socially conservative views, or those who might be overall to the Right but still believe in greater economic equity. But…. This still leaves us with a singular Left to Right dimension, even if we arrive there in a more beliefs-centered and data-driven, rather than self-perception, method. 

Step Two: Adding a Second Dimension—Collectivist vs. Individualistic

While our Left-to-Right composite does capture differences related to respondents’ beliefs about economic and social issues, it overlooks how people believe solutions should be pursued. That’s why we introduced a second dimension: collective vs. individualistic. This dimension gauges whether someone endorses more collective solutions (group-level or collective benefits) or individual autonomy (emphasis on personal freedoms and minimal regulation). For example, two people might both be “Left-leaning,” but one believes in top-down government programs, while the other prefers letting communities or individuals handle problems. A simple Left–Right scale doesn’t reveal that difference, leading us to lump them together even though their strategies for real-world solutions are distinct. Without this additional dimension, we risk painting an overly simplistic picture of ideology. 

Many other models, whether purely Left vs. Right or purely economic vs. social, don’t capture important aspects of how people believe society should organize itself. This leads to gaps in our understanding of voter behavior and policy preferences about who should solve what, and what roles should the state, community, and individual play in it? Following Haidt’s work mapping moral foundations to ideology (Haidt et al. 2009), we reviewed issue statements and survey items to see which ones signaled group-focused thinking versus individual responsibility. Items emphasizing broad social safety nets or communal decision-making add to respondents’ collective-leaning score and items emphasizing personal choice, autonomy, and opposition to what’s seen as overreach are added to their individualistic score. We once again, calculated the difference to arrive at an Individualistic-collectivist composite score, and applied a clustering algorithm to see what groups emerge. 

Figure 2: Two-Dimensional Political Map

The result is a more information-rich, two-dimensional map of ideology, as shown in Figure 2. Instead of lining everyone up from Far Left to Far Right on a single row, we now see six distinct clusters (color-coded) that show not only whether someone leans Left or Right, but also how collective or individualistic their preferred solutions tend to be. This affords us greater predictive power (increasing explained variance, adding to predictive accuracy by adding more details and distinctions), better identification of overlooked groups (like Right-leaning supporters of social safety nets or Left-leaning skeptics of federal overreach) and insights into policy debates (where unexpected alliances can form around specific approaches, regardless of “Left” or “Right” labels).

Conversations around healthcare, education, and climate often hinge on whether the solution should be federally administered or individually led. By tracking collective vs. individualistic leanings, we can see which groups are likely to support or oppose legislation, even if they share a nominally similar stance on the Left–Right scale.

Overall, adding a second dimension helps us avoid one-dimensional thinking. We end up with a two-axis map (Left–Right vs. collective–individual) that shows far more varied clusters, capturing a richer portrait of political ideology than a simple “Which side do you identify with?” could ever achieve.

Once we place people in this multi-dimensional space using their endorsement and prioritized beliefs, we can begin investigating questions like: Why do certain Left-leaning individuals still champion individual liberty over group-based solutions? How do pro-environment conservatives reconcile their stance on market freedom with federal regulations? Our approach provides rich data to explore how people arrive at their unique political philosophies, beyond overly simplistic labels.

Rather than relying on how people think of themselves, our project starts by building on what they actually endorse, making our measure more grounded in real attitudes and more flexible for uncovering new ideological patterns that may not be visible on a standard 1-to-7, Left or Right scale. It helps to reveal that American political life is more varied than just “Democrat” vs “Republican,” and that our beliefs, sometimes surprising or even seemingly contradictory, can be understood in greater depth.

What This Reveals—and Why It Matters

Most of us don’t fit neatly into ideological boxes. Our political views often reflect a blend of instincts, relationships, and personal experiences that don’t line up perfectly with a single party or label. Some of us value tradition and social cohesion, but also care deeply about fairness and justice. Others champion personal freedom while also believing we have a responsibility to protect the vulnerable. These aren’t contradictions—we each hold a mix of values that guide us in different contexts. This internal diversity is part of what makes us human. It’s also why political identity often feels more complicated than the labels we’re given. Instead of seeing Left and Right as opposing teams, we can begin to see political differences as expressions of different priorities and problem-solving strategies. The traditional Left–Right line helped organize large-scale politics, but it doesn’t fully capture how real people think, feel, and act.

Why an Emergence Approach Helps Understand Reality

At CARE, we take an emergence-based view. That means we don’t assume people fit into predefined categories—we let patterns emerge from the ground up. Starting with people’s strongest convictions on real issues, we look at how these beliefs naturally cluster. What we find are new groupings that reflect not only where people stand on issues, but also how they think solutions should be approached—whether through shared responsibility or individual freedom.

This two-dimensional lens—Left vs Right and Collective vs Individual—gives us a more accurate picture of what drives political behavior. It helps us understand why unlikely alliances form, why some policies get broad support, and why others spark intense resistance. It even helps us spot warning signs for polarization or authoritarian drift.

Most importantly, this approach honors the complexity of human nature. We’re not just collecting data on ideology, we’re developing a new grammar for understanding how people form beliefs, build coalitions, and shape our future.

Want to Go Deeper?

This post is only the beginning! If you’re curious about how understanding the split brain of U.S. politics can help us build more durable coalitions, navigate differences with compassion, and move toward a healthier republic—check out our extended essay:

Healing the Split Brain of U.S. Politics
A reflection on political cognition, moral pluralism, and the path forward.

In it, we explore how Left and Right are not just opposing teams, but different cognitive styles—each with something vital to contribute to the whole. Healing the divide begins with understanding ourselves.

Reference: 

Bergson, A. (1958). [Review of An Economic Theory of Democracy, by A. Downs]. The American Economic Review, 48(3), 437–440. http://www.jstor.org/stable/1809781

Carmines, E. G., & Stimson, J. A. (1989). Issue Evolution: Race and the Transformation of American Politics. Princeton University Press. https://doi.org/10.2307/j.ctv141636r

Haidt, J., Graham, J., & Joseph, C. (2009). Above and Below Left–Right: Ideological Narratives and Moral Foundations. Psychological Inquiry, 20(2–3), 110–119. https://doi.org/10.1080/10478400903028573

Further Reading

Why the self-reported Left-Right spectrum is problematic:

https://www.cambridge.org/core/journals/political-science-research-and-methods/article/selfreported-political-ideology/C2BED995008303104F4D43819B5FCC1E

https://www.annualreviews.org/content/journals/10.1146/annurev-psych-020124-115253