Subjects: ten provisional areas of public life

The Subjects section of My Vote Forward covers ten broad areas of public life, from education to defense, that people can rank by how much attention they think each deserves. The ten subjects are a provisional working list chosen by the project for its demonstration, not a scientific or official taxonomy, and every ranking of them depends on how the list was drawn. Each subject has three pages: what the public prioritizes and why it is hard to measure, who actually decides, and what the evidence shows.

In plain words

A subject is a big area, like “Housing and Communities.” It is not a plan. Saying housing matters most to you does not say what you want done about housing.

My Vote Forward lets people say which subjects deserve attention first. To ask that question, somebody has to write the list. We wrote this one. It has ten items. The number ten is a design choice, not a fact about the world, and nobody has yet tested whether it is the best size.

So we show you the list, tell you who made it, and explain what it leaves out. For each subject you can read three things: what people say about it, which government body actually has the power to act, and what the research shows.

Defined terms

Subject. A broad area of public concern used to ask about relative importance, such as education or health and care. A subject has no cost, no legal authority and no author. See Priorities.

Proposal. A specific action that someone has written down, with a responsible authority, a cost, a funding source, dependencies and a status. A proposal can be approved, amended or rejected. A subject cannot. See Proposals.

Subject list (taxonomy). The set of subjects a person is asked to compare. Its boundaries decide what counts as one thing and what counts as two.

Coverage experiment. A list adopted to test whether people can use it, what they feel is missing, and how much it shapes answers, rather than a claim that it is correct.

Open-ended question. A question that lets people name a problem in their own words, which researchers then code into categories.

Closed-ended question. A question that offers a fixed list of options.

The ten starting subjects

The ten subjects in the My Vote Forward demo are stored as versioned configuration, labeled subjects-v1, with the status “provisional.” The configuration describes the list as “a coverage experiment, not a locked taxonomy,” and adds that “subjects overlap; they are not budget accounts.” Each has a one-line scope:

Each subject links to its own page, which in turn links to its “Who decides” and “Evidence” pages.

Evidence: who chose the ten subjects

The ten subjects were chosen by the people building My Vote Forward, as a working list for the demonstration. No public process, citizens’ panel, legislature or outside expert body selected them, and no study has yet measured whether this list is easier to use or fairer than another. That is the honest answer to the question “who chooses the ten?” and it is the reason the list is labeled provisional.

The list was not invented from nothing. Researchers and statisticians have long sorted government activity into categories, and the ten subjects overlap heavily with two established systems.

The first is the Classification of the Functions of Government (COFOG), maintained through the United Nations Statistics Division. COFOG was first published in 1980, and a revised version was approved by the UN Statistical Commission in March 1999.⁠¹ It has ten top-level divisions: general public services; defense; public order and safety; economic affairs; environmental protection; housing and community amenities; health; recreation, culture and religion; education; and social protection.⁠¹ The UN Statistical Commission recommended a revision in 2022, and the UN expects that update to be completed by 2028.⁠¹ COFOG classifies spending by purpose, which is a different job from asking people what matters to them, but the overlap shows that “ten” is a common size for a top-level list.

The second is the Comparative Agendas Project, a research network that codes laws, hearings, budgets and media coverage across countries. Its master codebook sorts each entry “into one of 21 major topics and 220 subtopics”,⁠² including macroeconomics, civil rights, health, labor, education, environment, energy, immigration, transportation, housing, defense and government operations.⁠² A list of 21 is more precise than a list of 10. It is also harder to rank on a small screen.

Comparing the three lists shows what any short list must merge or drop. My Vote Forward folds labor into “Economy and Work,” puts energy with environment, and combines courts and policing in “Safety and Justice.” It has no separate subject for agriculture, technology, public lands, culture, or social welfare, all of which appear in the Comparative Agendas codebook.⁠² COFOG gives “social protection” its own division,⁠¹ while the demo spreads income support and caregiving across “Economy and Work” and “Health and Care.” These are judgment calls. A person whose main concern is farm policy or retirement security has to decide which box it belongs in, and some will decide none does.

Why the list shapes the answers

The subject list shapes the answers because people choose from what they are shown. Pew Research Center reports a direct test: when a 2008 survey asked which one issue mattered most in deciding a presidential vote, 58% chose the economy when it was offered as a response, but only 35% volunteered the economy when the question was open-ended.⁠³ Among those given the closed list, fewer than one in ten (8%) gave an answer other than the five offered; among those asked the open version, 43% gave an answer that was not on the closed list.⁠³ Pew also notes that the order of options matters, that randomizing them “does not eliminate order effects” but spreads the bias randomly.⁠³ See Question design.

Two well-known U.S. measures of public concern show how large the difference can be.

Gallup asks an open-ended question: “What do you think is the most important problem facing this country today?” In its September 1 to 17, 2026 telephone survey of about 1,000 adults (margin of error ±4 percentage points), 30% named the government or poor leadership, 13% the high cost of living or inflation, 11% the economy in general and 10% immigration.⁠⁴ Healthcare was named by 3%, education by 1%, unemployment or jobs by 1%, and infrastructure did not register in that month.⁠⁴ Because people can name more than one problem, the September total was 133%.⁠⁴

Pew asks a closed question about priorities for the president and Congress. In its January 16 to 21, 2024 survey of 5,140 adults on its American Trends Panel, 73% called strengthening the economy a top priority, 49% said the same of improving the job situation, and about six in ten across age groups gave that rating to reducing health care costs and to improving education.⁠⁵

Neither measure is wrong. They answer different questions. An open-ended “most important problem” captures what is on people’s minds, often the news of the month. A closed list of priorities captures how many things people consider important when asked about each one. Education looks marginal in the first and broadly important in the second. A ranking of ten subjects sits between the two: it forces a choice, like the open question, but within a fixed list, like the closed one. Its results inherit the strengths and the distortions of both. See Measurement.

How a subject differs from a proposal

A subject differs from a proposal the way a topic differs from a decision. “Education” is a subject. “Hypothetical example: the county school district asks voters to approve a bond to replace two elementary schools, repaid by property taxes over 25 years” is a proposal. The proposal has an author, an authority with legal power to adopt it, a price, a funding source and a date. The subject has none of those.

This distinction matters for three reasons.

First, people who agree on a subject often disagree on every proposal within it. Two people who both rank housing first may want opposite things: one more construction, the other stronger limits on development. A group ranking of subjects cannot be read as support for any proposal. See Priorities.

Second, a subject is usually spread across many governments. The Census Bureau counted 90,837 local governments in the United States in 2022 and 91,438 in 2025, an increase it attributes mainly to growth in special districts in Colorado, Florida, Nebraska and Texas.⁠⁶ Most subjects are divided among federal agencies, states, counties, cities, school districts and special districts. That is why each subject here has a “Who decides” page. A priority that does not reach the body with authority is a signal with no address. See Jurisdictions.

Third, subjects compete for the same money, but a ranking does not say how much. Ranking schools above roads is not the same as moving any amount from roads to schools. See Tradeoffs.

What each subject’s three pages cover

Each subject page in My Vote Forward follows the same structure, so readers can compare them.

• The subject page asks what the public says it prioritizes, why that is hard to measure in this area, the real tradeoffs, and the strongest arguments made from different perspectives. Contested claims are attributed, not asserted.
• Who decides maps which authority controls what: federal law and agencies, state constitutions and legislatures, counties, cities, school boards, special districts, courts and ballot measures. It names example states and notes where the rules vary.
• Evidence reports what public data and research show, with dates and denominators, where researchers disagree, and what is not known.

Competing interpretations

Critics and supporters of subject lists read the same facts differently, and each view has force.

The list is a useful map. One view holds that a short, stable list helps ordinary people think about government as a whole and lets results be compared over time. Statistical systems such as COFOG exist because consistent categories make comparison possible.⁠¹

The list is a frame that does the deciding. Another view, grounded in agenda-setting research and in survey experiments like Pew’s,⁠³ holds that whoever writes the list sets the terms of the answer. On this view, the omission of agriculture, retirement security or technology is not neutral; it tells participants those concerns are secondary.

The list hides conflict. A third view argues that broad subjects paper over disagreement. “Safety and Justice” contains people who want more policing and people who want less. Ranking the subject high reveals concern, not direction, and can be spun by either side.

Any list is better than an unstated one. A fourth view notes that every public conversation already runs on an implicit list: the topics covered by news outlets, debated in campaigns or funded in budgets. A written, versioned list at least makes the frame visible and open to challenge.

My Vote Forward does not resolve this argument. It adopts the safeguards each view implies.

Application to this framework

• The subject list is published with its version, scope lines and status. Any change creates a new version; earlier answers keep the version they were given against.
• Results report which list version they used, the order method, and the denominator. See Collective priorities.
• A subject ranking is never presented as support for a proposal. Opinions on proposals are a separate step with specific wording. See Opinions.
• Each subject points to the authorities that decide it, so a priority can be directed to the body that can act.
• Omissions are named on this page rather than left for readers to discover.

Unresolved questions

• Who should own the subject list in a real deployment: the operator, an independent review board, a randomly selected citizens’ panel, or participants through write-in suggestions? See Governance.
• How much does this particular list, and its order, move rankings compared with an open-ended question? That has not been measured.
• Should subject lists differ by level of government, since a city does not run defense and a nation does not run school districts? See Jurisdictions.
• Should overlapping subjects be split (for example, energy apart from environment), and at what cost to usability?
• How should results be compared across list versions without implying that a change in the list was a change in the public?