Can public priorities be manipulated by media?

Yes, media coverage influences which issues people rank as important, and a priority ranking in My Vote Forward would reflect that influence. Maxwell McCombs and Donald Shaw argued in 1972 that by choosing what to cover and how prominently, news organizations may set the “agenda” of a campaign, and later research has built a large literature on that idea. The framework cannot separate “real” priorities from media-shaped ones, because people’s sense of importance always comes partly from what they see and hear. What it can do is date every result, show change over time, and refuse to treat a snapshot as a permanent public will.

The objection in its strongest form

The objection to My Vote Forward here is that public priorities are not independent facts waiting to be measured; they are produced. A crime story that leads the local news for a week, a viral video, or a coordinated social media campaign can move an issue to the top of people’s minds without any change in underlying conditions. If the framework publishes “the public’s top priority is X” a week later, it launders media salience into the language of democratic will. Worse, whoever can most effectively shape coverage, whether a well-funded campaign, a partisan network, or a foreign government, gains a new lever: move the coverage, move the ranking, and then cite the ranking as public demand.

A sharper version notes the feedback loop. Journalists report rankings as news, which increases coverage of the top-ranked issue, which reinforces its ranking. The framework becomes another amplifier.

Why the objection matters

The objection matters for My Vote Forward because priorities are the framework’s most public output. If rankings mostly track coverage, they tell institutions little they could not learn from reading the news, and they risk giving coverage-driven swings an air of authority. It also matters because media influence is not evenly distributed: groups with resources to shape coverage gain influence over rankings. See agenda capture and misinformation.

Defined terms

• Agenda setting: the influence of media coverage on which issues the public considers important. See agenda-setting.
• Salience: how prominent an issue is in people’s minds at a given time. See salience.
• Persuasion: a change in people’s positions or votes, as distinct from a change in what they consider important.
• Influence operation: a coordinated effort, often concealed, to shape public perception, sometimes by a foreign actor.
• Priority: in the framework, a judgment of relative importance across subjects. See priorities.

Evidence

The founding study is McCombs and Shaw’s “The Agenda-Setting Function of Mass Media,” published in Public Opinion Quarterly in 1972.⁠¹ Its abstract argues that editors and broadcasters shape political reality through choices about what to present and how to display it, that readers learn how much importance to attach to an issue from the amount of coverage and its position, and that the media may thereby set the “agenda” of a campaign.⁠¹ That study concerned one campaign and one community; agenda-setting has since become a large research field, and this page does not summarize the whole of it.

Survey design shows a closely related effect at the moment of measurement. Pew Research Center reports that when the economy was offered as an option, 58% chose it as the most important issue, compared with 35% who volunteered it unprompted.⁠² If a list of options can move answers by more than 20 points, so can whatever made an issue feel urgent before the question was asked.

Media can also move votes, not only attention. Gregory Martin and Ali Yurukoglu used cable channel positions as a source of variation in viewing and estimate that “Fox News increases Republican vote shares by 0.3 points” among viewers induced to watch 2.5 additional minutes per week.⁠³ That is a study of persuasion through one network; similar questions apply to other outlets, and the authors’ model is designed to assess media-driven polarization generally.

Evidence on online influence operations is more mixed than popular accounts suggest. A study of the Russian Internet Research Agency’s Twitter campaign in the 2016 United States election found that 1% of users accounted for 70% of exposures, that exposure was smaller by an order of magnitude than exposure to posts from national news media and politicians, and that there was “no evidence of a meaningful relationship” between exposure and changes in attitudes, polarization, or voting; the authors note the estimates are observational and cover only Twitter.⁠⁴ That finding does not show influence operations are harmless; it shows their measured effect on attitudes in that case was small relative to domestic media and politicians.

Trust in media is low and divided, which shapes how any media-driven priority shift is received. Gallup’s September 2025 survey found that 28% of Americans had a great deal or fair amount of trust in mass media, compared with 68% to 72% when Gallup began asking in the 1970s; trust was 8% among Republicans, 27% among independents, and 51% among Democrats.⁠⁵

How comparable systems have handled it

Programs that measure priorities handle media influence through time, structure, and information.

• Repeated measurement. A priority question asked the same way at regular intervals can be read as a time series, so that changes can be compared with events and coverage. A single reading is much harder to interpret.
• Deliberation before measurement. Deliberative polls measure views after participants have discussed the issues in depth, rather than at whatever moment coverage peaks. See deliberative poll.
• Fixed agendas set by accountable bodies. Ireland’s Citizens’ Assembly had its topics set by parliamentary resolution rather than by what was in the news. See who chooses the ten?.

Competing interpretations

One interpretation holds that media influence on priorities is not manipulation but how democracy works. People learn about problems from news; if coverage of a crisis raises its priority, the system is functioning. On this view, the objection proves too much: it would make every measure of public concern illegitimate.

A second interpretation distinguishes legitimate from illegitimate influence. Coverage of real events is legitimate; concealed coordination, fabricated content, or paid amplification is not. On this view the framework should focus on detecting inauthentic activity, not on removing media influence. See won’t bots flood it?.

A third interpretation holds that priorities measured between elections are especially vulnerable because they lack the stabilizing force of a campaign in which all sides compete for attention. On this view, results should be treated as volatile readings rather than settled preferences.

What the framework answers

My Vote Forward accepts that priorities are shaped by information and focuses on making that visible.

• Every result is dated. Releases state the participation window, so readers can compare results with events.
• Results are shown over time. A single ranking is presented alongside earlier releases using the same list version where available. See participation over time.
• Priorities can be revised. People can update rankings as they learn, and releases use stated cutoff rules.
• Separate priorities from positions. Even if coverage raises an issue’s priority, the framework does not infer what people want done about it. See priorities.
• Coordinated activity is flagged under published rules. See won’t bots flood it?.

Hypothetical example. A county release in March shows “safety and justice” rising from fifth to first after weeks of coverage of a high-profile case. The release shows the earlier rankings, the dates of participation, and the number of people who changed their ranking during the window. It does not claim the shift reflects a change in crime rates or a durable change in priorities.

Where the objection still wins

The objection wins on its core claim: the framework cannot separate media-shaped priorities from any other kind, and there is no neutral baseline to compare against. Showing results over time helps readers interpret them but does not stop a dated snapshot from being quoted as “what the public wants.” The framework also has no capacity to detect sophisticated influence operations, and it has no data showing how much its rankings would track coverage.

Unresolved questions

• Should releases include a media context note, and who would write it neutrally?
• Would a minimum participation window, such as several weeks, reduce sensitivity to short news cycles, and at what cost to responsiveness?
• How could the framework study the relationship between its rankings and coverage without itself becoming a media monitor?
• Is there any meaningful sense of “underlying” priorities independent of information?

Related

• Agenda capture
• Misinformation
• Local news collapse
• Salience
• Won’t this increase polarization?
• Hard questions