Priority versus salience: why “the most important problem” is not the same as what matters most

Salience is how prominent an issue is in public attention at a given moment; priority is how much weight people would give an issue when choosing among competing uses of public effort. The standard survey measure, Gallup’s open-ended “most important problem” question, captures salience well but, as political scientist Christopher Wlezien showed, it mixes an issue’s importance with how much of a problem it currently is. A priority measure needs a different design.

In plain words

Ask people “What is the most important problem facing the country?” and the answer often tracks the news. When prices jump, many say the economy. When there is a war, many say war. When those calm down, other answers rise, even if people care about them no more than before.

That does not mean the economy stopped mattering when prices settled. It means it stopped being the biggest problem right now. “Most urgent this month” and “most important to me” are different questions.

A tool that asks people to rank public priorities has to decide which question it is asking. If it does not, it will report headlines and call them values. This page explains the research behind that distinction. The general effect of wording on answers is in Question design, and the salience glossary entry gives the short definition.

Defined terms

• Salience. Prominence of an issue in attention, at the individual or public level. See the glossary.
• Importance (I). How much an issue matters to a person, in Wlezien’s framework.⁠¹
• Problem status (P). How much of a problem an issue currently is, in the same framework.⁠¹
• Most important problem (MIP). The standard open-ended survey question asking what the most important problem facing the nation is.
• Most important issue (MII). A variant wording asking for the most important issue rather than problem.⁠²
• Agenda-setting. The influence of news coverage on which issues the public considers important.⁠³ See the glossary.
• Issue-attention cycle. Anthony Downs’s concept that public attention to an issue rises and falls in a recurring pattern, examined empirically by Peters and Hogwood.⁠⁴
• Priority (framework sense). A participant’s stated relative importance among a fixed list of public subjects, as defined in Priorities.

Evidence

What the MIP question measures

Wlezien’s paper “On the Salience of Political Issues” begins with the history: scholars measuring salience “initially turned to voters’ responses to the standard survey question that asks about the ‘most important problem’ (MIP) facing the nation,” and “at about the same time, scholars turned to aggregate MIP responses to characterize the broader public salience of issues over time”.⁠¹ His argument is that “the research confuses at least two very different characteristics of issue salience, the ‘importance’ of issues and the degree to which they are a ‘problem,’” and that “the confusion is rooted in basic measurement”.⁠¹

He formalizes the point. For individual i and domain j, define importance and problem status, and let IP_ij be how much domain j is an important problem. Then:

MIP_i = max over j of { IP_ij }

In Wlezien’s words, MIP “not only combines ‘importance’ and ‘problem,’ it is singular,” and in the aggregate “the public’s most important problem represents the plurality IP winner”.⁠¹ Three consequences follow, which he states directly: MIP for a domain “will vary over time when the importance of the domain changes”; it “also will vary when the problem status of the domain changes, even if importance remains the same”; and even if both stay fixed, it “will vary simply because of shifts in the degree to which other issues are important or problems”.⁠¹ His example: “Give us peace and we need to look elsewhere for our MIP. Add in prosperity and we have to look further still”.⁠¹

Using Gallup data from 1945 to 2000, Wlezien concluded that “most of the variance reflects changes in problem status within the domains and in the other domains,” with some surplus variance that may reflect importance.⁠¹ The published version appeared in Electoral Studies in 2005 under the title “On the salience of political issues: The problem with ‘most important problem’”.⁠¹

Changing the wording does not fix it

If the word “problem” is the issue, perhaps asking about the most important “issue” would help. Jennings and Wlezien tested this with British data in Public Opinion Quarterly (2011). They found that “MII responses are strikingly similar to MIP responses,” and that while MII “may be slightly closer to a valid indicator of issue importance, MII responses mostly reflect variation in assessments of problem status.” Their conclusion: “An effective measure of issue importance remains elusive”.⁠²

What current data look like

Gallup still asks the open-ended question monthly. In its September 2026 reading, 31% named an economic problem (net), including 13% for the high cost of living or inflation and 11% for the economy in general; 30% named the government or poor leadership; 10% named immigration; and environment or climate change drew 1%.⁠⁵ The column of percentages totals 133%, because some respondents name more than one problem.⁠⁵

Two methodological points follow from the format. First, a 1% mention rate for the environment is not evidence that people think environmental policy is unimportant; it is evidence that few named it as the single most pressing problem that month. Second, the multiple-mention total means the published figures are shares of respondents naming a category, not shares of a single ranked choice, which complicates any comparison with ranked priority data.

Media coverage moves salience

McCombs and Shaw’s 1972 study, “The Agenda-Setting Function of Mass Media,” argued that “readers learn not only about a given issue, but also how much importance to attach to that issue from the amount of information in a news story and its position,” so that “the mass media may well determine the important issues” of a campaign.⁠³ Agenda-setting research since then has examined how coverage volume affects what people name as important. For a priority instrument, the implication is that answers collected during a news spike will partly measure the spike.

Attention rises and falls

Peters and Hogwood’s 1985 article in the Journal of Politics tested whether Downs’s issue-attention cycle, which Downs “introduced ... based on his perception of the ecology issue rather than on a quantitative analysis of issue salience,” holds across U.S. policy areas, using changes in federal government organizations as an indicator.⁠⁴ They found organizational activity “is indeed found to take a cyclical form, with the timing differing for different policy areas,” and that “peak periods of organizational activity occur either during the period of peak public concern with an issue or in the period immediately after that peak”.⁠⁴ Salience matters for what government does, which is why measuring it accurately, and separately from importance, matters.

A simple formal contrast

The following is a design analysis, not a finding. Suppose that for each person i and subject j there are three quantities: urgency U_ij (how pressing now), importance I_ij (how much it matters over time), and tradeoff weight W_ij (share of a fixed budget the person would assign). Then:

• An open MIP question measures, roughly, argmax over j of f(U_ij, I_ij), one subject per person.
• A ranking of importance with an explicit time horizon measures the order of I_ij.
• An allocation task measures W_ij, which also reflects beliefs about cost and effectiveness.

A subject can be low in U and high in I (maintenance of infrastructure between failures is a plausible example, offered as a hypothesis, not a finding). Another can be high in U and moderate in I (a short-term price spike). A single question cannot separate these.

Competing interpretations

“Salience is what matters politically.” On this view, the MIP question measures what moves votes and officials’ attention, and Peters and Hogwood’s evidence that government activity tracks peak concern.⁠⁴ supports measuring salience directly. Priorities detached from urgency may describe values no one acts on.

“Importance is what a deliberate public would want measured.” On the opposite view, salience is too driven by coverage⁠³ and by the current state of problems⁠¹ to serve as guidance on long-term choices. See Short-term thinking and Agenda capture.

“Importance cannot be measured cleanly at all.” Jennings and Wlezien’s conclusion that “an effective measure of issue importance remains elusive”⁠² supports skepticism that any wording isolates importance. Every instrument measures a blend; the task is to know which blend.

“Allocation reveals priorities better than ranking.” Forcing a budget split exposes tradeoffs that rankings hide, as discussed in Collective priorities. Critics respond that allocation also measures beliefs about cost and effectiveness, not only importance.

Application to this framework

These are design proposals for My Vote Forward, not results.

1. Ask separate questions, labeled. “Which of these feels most pressing right now?” (salience) and “Rank these by how much they should matter to public decisions over the next ten years” (importance), shown as separate results and never merged.
2. Record the news context. Each release states its collection dates and notes major events during the window, so readers can judge whether a spike in a subject reflects a news cycle. This builds on the snapshot dating in Measurement.
3. Show the urgency-importance gap. Publishing both results side by side lets readers see subjects that are important but not currently pressing, which is the group Wlezien’s analysis predicts single questions will miss⁠¹
4. Do not compare directly with MIP polls. Gallup’s format allows multiple open-ended mentions totaling more than 100%,⁠⁵ so a ranked priority share is a different quantity. Any comparison must say so.
5. Use repeated measurement to separate stable priorities from passing salience. Because participants can update over time (see Participation over time), the framework can show which subjects hold their rank across news cycles. Whether this works is a hypothesis to test in Evaluation.

Hypothetical example. In a demonstration, 40% of participants mark “economy and work” as most pressing during a month of high inflation, but only 22% rank it first for the next ten years, where “infrastructure and transport” rises from 5% pressing to 14% first. A release that showed only the first question would report a mandate for economic action. A release that shows both reports something more useful: what feels urgent and what people think should guide longer-term decisions differ.

Unresolved questions

• Does adding an explicit time horizon (“over the next ten years”) actually shift answers toward importance, or do people still answer with current urgency?
• How long must a priority be tracked before a stable rank can be distinguished from a salience spike?
• Should a priority instrument include an open-ended “anything missing?” item to catch problems the fixed list omits, and how would that interact with privacy rules for small categories (see Small-group suppression)?
• Can allocation tasks separate importance from beliefs about effectiveness, or is that separation impossible in practice?
• If media coverage shapes importance as well as salience,⁠³ is there any baseline of “what people would think without coverage,” and is that baseline even meaningful?