What is driving the midterms: living costs, corruption and the Iran war

As the November 3 midterm election approaches, the clearest message from recent surveys is about what Americans want government to do: make daily life more affordable, address concerns about corruption and confront the consequences of the war with Iran. Those concerns overlap, but they are not interchangeable. Neither are the questions pollsters use to measure them.

A newly released CNBC survey puts government corruption and grocery costs near the top of a limited-choice issue list. Separately, AP-NORC finds strong support for making affordability and ending the Iran war priorities for the next Congress. Read together, the research describes a demanding electorate. It does not provide a single national ranking that can be assembled by mixing percentages from different questionnaires.

Why the questions matter

The CNBC questionnaire, prepared by Hart Research Associates and Public Opinion Strategies, asks respondents to select one or two of the most important issues facing the country from a list. Government corruption appears first in the resulting table, followed by food and groceries. Protecting democracy and gasoline prices each receive 23%; immigration and border security receive 17%; Iran and housing costs each receive 15%.

Because multiple responses were accepted, those shares are not slices of a pie that must add to 100%. Someone concerned about corruption could also choose groceries. Nor is the question asking respondents to explain how much each issue will influence their actual ballot. A ranking of concerns and a measurement of vote choice answer different questions, even when both appear in the same survey.

The study reports an overall sampling margin of about 3.1 percentage points. That figure is not a special significance test of the difference between two answers given by the same respondents. The reasonable conclusion is that corruption and groceries are both prominent. Calling the two-point gap a clear national mandate would go further than the topline establishes.

Affordability is broader than one price

A question that divides economic concerns into groceries, gasoline, housing and jobs can produce a different-looking ranking from one that uses a broad category such as the economy. That is a feature of question design, not proof that voters abruptly stopped caring about household budgets. It is also why simply adding the individual percentages would be misleading: respondents can appear in more than one category.

For a family evaluating competing campaign claims, affordability is best translated into concrete proposals. Does a candidate describe a change to taxes, benefits, health costs, housing supply or energy policy? Who would qualify, when would the change begin and how would it be funded? A promise to lower costs becomes more informative when a voter can connect it to an identifiable policy and a realistic timetable.

Prices also differ from the rate at which prices rise. Slower inflation does not by itself mean a family's grocery bill returns to an earlier level. A campaign may point to improvement in the inflation rate while a household remains dissatisfied with the amount it pays. Those statements can coexist. Neither one, on its own, explains whether a particular legislative proposal would improve that family's finances.

What AP-NORC measured separately

The AP-NORC primary topline reports interviews with 1,119 adults October 8–12. Its future-Congress question asks how high a priority each listed action should be. Among all adults in that published table, lowering the cost of living receives 80% in the highest-priority category, and ending the war with Iran receives 68%. These are ratings of individual actions, not a forced competition for one top spot.

The same table puts investigating the Trump administration at 36% and supporting its agenda at 18% in the highest-priority category. These findings should not be translated into a claim that accountability is unimportant. Investigating one administration is a more specific proposition than concern about government corruption generally. Someone can favor stricter ethics rules, spending audits or conflict-of-interest restrictions without making a particular investigation their first priority.

AP-NORC used NORC's probability-based AmeriSpeak panel, with online and telephone interviews. Its overall reported sampling margin is 3.9 percentage points for adults. Some accompanying releases focus on registered voters, a subset of the sample. This article identifies the all-adult denominator when using the published topline table; the full-sample margin should not be casually assigned to smaller partisan or voter subgroups.

Corruption needs a specific policy question

Concern about corruption can mean different things to different people: conflicts of interest, favoritism in contracting, misuse of public money, unethical conduct or illegal activity. A poll response is evidence of concern, not a finding that a named person committed a crime. Voters should ask candidates which conduct they propose to address and which standards they would apply to members of their own party.

Useful tests include whether a proposal requires public disclosure, independent review, enforceable penalties and consistent treatment across administrations. Another is whether the proposal would prevent a conflict before a decision is made, or merely require a report afterward. Those are different forms of accountability. A candidate's willingness to explain the difference is more informative than a general promise to clean up Washington.

Congress has actual tools in this area. The House's account of its investigative authority explains that oversight is connected to the legislative function. Committees can examine how laws and programs operate and seek information relevant to legislation. Oversight is not the same as a criminal trial, and a public hearing does not itself establish criminal guilt.

The Iran issue connects foreign and domestic choices

The AP-NORC finding about ending the war measures a desired priority, not support for a detailed peace agreement or a particular military withdrawal timetable. Respondents were not approving identical terms concerning security guarantees, nuclear restrictions, regional partners or enforcement. Treating the result as agreement on every possible route out of the conflict would erase the choices that policymakers still have to make.

For voters, the useful follow-up is how a congressional candidate would evaluate military funding, diplomatic proposals and the risks of further escalation. Congress controls appropriations and can conduct oversight, while the president directs the executive branch. The House's explanation of the power of the purse sets out why spending decisions matter. A campaign promise should be judged against those institutional powers rather than against the assumption that one newly elected lawmaker could settle a war alone.

There can also be tension between priorities. A voter may want lower prices, reduced federal spending and strong national security simultaneously. Candidates should explain tradeoffs rather than imply that every objective can be achieved immediately at no cost. Asking for the likely timing, funding source and practical limits of a proposal is a way to test seriousness without assuming that either party has a simple solution.

What this means for the final campaign stretch

The two surveys were conducted at different times and among different populations. They should not be presented as a before-and-after experiment showing that one week's news caused a measurable shift. They also do not identify how every competitive district will vote. A national concern can be widespread while its electoral effect varies with local candidates, turnout and the issues raised in a particular race.

The strongest reading is therefore about the questions candidates still need to answer. What would they do to ease household costs? How would they prevent and investigate conflicts of interest? What powers would they use to address the Iran war, and what would require cooperation with other officials? Specific answers would help voters compare policies instead of merely comparing slogans.

As Election Day approaches, readers can use the underlying questionnaires to check campaign claims about what the public wants. Look for the interview dates, the people included, the exact question and whether respondents could choose more than one answer. That small amount of context separates a useful piece of evidence from a dramatic statistic stripped of its meaning. The polling points toward concerns that demand substantive answers; it does not substitute for the election itself.