Public Opinion

How do people’s views become public and acquire influence?

Suppose a city proposes changing the route of a bus service. Students welcome a shorter journey. Shopkeepers worry about access. Some residents favour the proposal but consider it a minor issue; others organise a meeting. A news report announces that “the public supports the change”.

Several questions now arise. Which public was studied? What exactly do people support? How strongly do they feel? Whose voices become visible, and which decision-makers respond?

A survey can estimate how views are distributed within a defined population. A discussion can reveal reasons, disagreements and conditions of support. A study of public influence can trace which expressions enter institutional decisions.

Key’s political definition

V. O. Key, Jr.’s Public Opinion and American Democracy (1961) connects public opinion with the views of private persons that governments find it prudent to heed. This directs attention towards the relationship between citizens’ views and governing institutions.

The question of influence extends beyond the presence of a majority. Decision-makers encounter elections, surveys, organised groups, public meetings, reporting and personal networks. Investigate how each signal becomes visible and how it is interpreted.

Follow three levels

• Individual judgement: what a person believes, prefers or feels about an issue.
• Public distribution and discussion: how those views are shared, disputed, expressed and measured.
• Institutional consequence: how public signals enter decisions, and whose views carry weight.

In my lecture, I distinguish quantitative and qualitative conceptions of public opinion. This chapter develops that distinction and connects it with media, social influence, measurement and democratic practice.

Carry the bus-route proposal through the chapter. It offers a concrete way to examine the movement from an individual preference to a public claim.

Foundational readings: Key (1961); Price (1992).

An issue connects people through shared consequences and communication.

A public can be identified through the people affected by an issue, those attending to it, those discussing it or the population whose views a study seeks to estimate. Explain which meaning applies before making a claim about “the public”.

One issue can gather several publics

The bus-route proposal affects passengers, residents, workers, businesses and people whose journeys connect with other services. Some follow the issue closely; others encounter it only when asked. These relationships can overlap.

An issue public brings sustained interest or involvement to a particular topic. An attentive public follows public affairs more broadly. Organised participants may have the time, resources or experience to maintain a visible presence.

Four questions from the lecture

The lecture directs attention towards an issue, people who hold and express views, some alignment among positions, and the possibility of influence. Use these as questions about the formation and force of public opinion.

Ask how the issue becomes recognisable, how people formulate positions, where agreement or conflict emerges, and how expression reaches others. A shared concern can gather a public even while its members disagree about the solution.

Subject matter and expression

A privately held view can concern a public issue. It can enter a survey anonymously, appear in a conversation or become part of a public statement. The content of the view and its mode of expression are distinct parts of the account.

Suppose a passenger privately opposes the route change but remains silent at a meeting. A confidential interview and an observation of the meeting may reveal different aspects of the same situation.

Plurality makes the analysis more precise. Identify which public, which issue, which period and which expression your evidence covers.

A position gains meaning through its strength, importance and conditions.

A favour–oppose measure identifies direction. It leaves other questions open: how strongly the view is held, how important the issue feels, what the person knows and which conditions would alter the preference.

Opinion, attitude, belief and value

An opinion is a judgement or position on a question, often articulated in a response. An attitude is an evaluative orientation towards an object, person or issue. A belief concerns what someone accepts as true. A value identifies a broader principle or goal regarded as desirable.

A passenger may value equitable access, believe that a proposed stop is difficult to reach, hold a favourable attitude towards public transport and oppose this particular plan. These elements can support a coherent account while leaving room for disagreement or revision.

Strength and importance deserve separate questions

Someone can strongly prefer one option while placing little priority on the issue. Another can regard the issue as urgent while remaining uncertain about the best solution. Ask about both relationships.

Suppose 60% support a proposal, but the opponents organise because they expect substantial costs. Investigate the distribution of views, the consequences people anticipate and the resources through which they express their concerns.

Information and experience can change a judgement

A revised timetable, a trial journey or a conversation may alter how a person understands the proposal. Track which consideration changed and how it entered the judgement.

The categories help identify what is changing. A shift in issue importance, a revised factual belief and a reversal of policy preference constitute different findings.

Counting positions and examining discussion reveal different relationships.

The quantitative conception treats public opinion as a distribution of individual responses within a defined population. The qualitative or discursive conception foregrounds how judgements develop, which reasons enter discussion and how positions acquire collective force.

These conceptions guide the object of inquiry. They can also inform a combined study that estimates a distribution and investigates the processes behind it.

Quality of judgement and access to discussion

Some qualitative conceptions assign special authority to informed judgement. Their elite versions emphasise knowledgeable participants. Democratic approaches examine how information, discussion and fair opportunities to speak can broaden participation.

Investigate the conditions under which a person can contribute. Time, language, confidence, access to information and the organisation of a meeting all shape whose reasons receive attention.

Blumer’s challenge

Herbert Blumer’s 1948 essay asks whether polling captures public opinion as a social process. It emphasises interaction, organised interests and the unequal influence of voices entering collective decisions.

This opens a question beyond the percentage: how do individual views become positions that institutions encounter? A survey can estimate support, while a study of meetings and organisations traces the routes through which claims become consequential.

Conceptions and methods

A conception identifies what public opinion is understood to involve. A method identifies how evidence is collected or analysed. Interviews can reveal reasons, and their responses can also be coded and counted. A survey can include open questions as well as standard response categories.

Choose methods in relation to the question. Combining them becomes useful when it connects an estimate with an explanation of formation, expression or influence.

Read Blumer (1948) and Price (1992).

Modern democracy depends on how people learn about a world beyond direct experience.

Walter Lippmann and John Dewey examined the difficulty of public judgement in a complex society. Both took communication seriously. Their writings foreground different responses to the distance between everyday experience and the consequences of large institutions.

Lippmann: the world and its representations

In Public Opinion (1922), Lippmann examines how people act through pictures of a world they can encounter only partly. News, symbols, categories and stereotypes help organise that picture.

His pseudo-environment names the mediated representation through which people interpret and respond to circumstances. Stereotypes simplify perception and can protect familiar expectations. His argument directs attention towards organised intelligence and the conditions under which reliable accounts reach public decision.

For the route proposal, compare the plan, the news account and the image a reader forms. Ask what the representation foregrounds, which categories make it familiar and which important details remain difficult to see.

Dewey: discovering the public through consequences

In The Public and Its Problems (1927), Dewey examines publics emerging around the indirect consequences of associated action. People need ways to recognise those consequences, investigate them and communicate about their shared problems.

Inquiry, education and communication help develop the conditions of democratic judgement. Expertise contributes to that process through its relationship with the experiences and concerns of those affected.

In the bus-route case, residents’ accounts can reveal consequences that a technical plan leaves unexplored. Shared inquiry can connect those accounts with transport evidence and alternative arrangements.

Use the classics to investigate a contemporary problem

Analyse the information available, the public’s opportunities to question it and the institutions through which a response can occur. Compare expertise, experience and public communication within the same situation.

These questions connect media literacy with institutional design. A clear account of a proposal helps people understand it; an accessible forum allows them to test its implications and contribute reasons.

Primary texts: Lippmann (1922); Dewey (1927).

People build judgements through experience, relationships and available considerations.

Opinion formation brings together personal experience, family socialisation, group identities, education, media, trusted contacts and elite messages. Their importance varies with the issue and the person’s relationship to it.

Socialisation and reference groups

People acquire political language, expectations and evaluative habits through recurring relationships. Families, peers, workplaces and communities can provide cues about what matters and which accounts deserve trust.

A reference group supplies a standpoint against which people interpret their own position. A student may consider the experiences of classmates, a household or a professional community when judging a proposal.

Direct experience and mediated knowledge

A daily passenger has experience of the service’s delays and access points. A distant observer depends more heavily on accounts supplied by others. Both can encounter new evidence and competing interpretations.

Ask which experiences a person draws upon and how they connect them with the issue. The same event can enter different accounts of fairness, efficiency or responsibility.

Zaller’s receive–accept–sample framework

John Zaller’s The Nature and Origins of Mass Opinion (1992) explains preference statements through exposure to messages, acceptance in relation to predispositions and the considerations accessible when a response is given.

Political awareness affects which messages are received and how they are evaluated. A survey response can draw upon a balance of accessible considerations, including competing arguments.

Suppose a passenger recalls the shorter journey when first asked about the proposal. Later, a question about accessibility makes a different concern prominent. Investigate which considerations enter each response and whether the underlying evaluation has changed.

Elite cues and credibility

A trusted organisation or public figure can help people interpret an unfamiliar proposal. Examine the source, the cue and the reason it carries weight. Expertise, shared identity and a record of reliable guidance may contribute differently.

Compare responses to the same information from different sources where the question warrants it. Follow how people explain acceptance, doubt or revision.

Key reading: Zaller (1992).

Media shape attention, interpretation, evaluation and opportunities to speak.

Media influence can operate through several processes. A story can make an issue more prominent, organise its meaning, supply a standard for judging performance or offer an argument for a particular position.

Specify the process and the outcome. Then investigate how exposure, interpretation, social relationships and prior views enter the explanation.

Agenda-setting

McCombs and Shaw’s 1972 study developed the agenda-setting question through the relationship between media emphasis and the issues voters considered important. This foregrounds issue salience as an outcome.

A transport story may move an issue higher on a person’s list of concerns while leaving the preferred solution unchanged. Measure importance and policy preference separately.

Framing

Entman’s account of framing examines selection and salience within communication, including the definition of problems, interpretation of causes, evaluation and proposed remedies.

Suppose one report emphasises journey times and another foregrounds access for older residents. Compare how viewers understand the problem and which consequences enter their judgement.

Priming

Iyengar and Kinder’s News That Matters investigates how television news can affect the considerations used to evaluate political performance. Priming directs attention towards the weight carried by a consideration.

A study can ask whether transport performance becomes more strongly related to an overall evaluation after exposure to relevant coverage. Interpret the result through the design and the comparisons it supports.

Reinforcement, discussion and competing accounts

Media encounters can confirm an existing view, supply new reasons, prompt disagreement or lead to conversation. Follow the response and the person’s prior relationship to the issue.

Suppose a resident reads a story, checks the proposed route and discusses it with a daily passenger. The resulting judgement develops across media and interpersonal communication. A diary or follow-up interview can trace those connections.

Institutions also interpret public signals

Editors, campaigners and officials may use polls, letters, meetings and social media activity as evidence of concern. Investigate how they assess these signals and how their decisions shape subsequent visibility.

Core readings: McCombs and Shaw (1972); Entman (1993); Iyengar and Kinder (1987).

People act on what they believe others think as well as on their own views.

An opinion climate is a perceived pattern of prevailing views within a relevant setting. People infer it through conversation, visible expressions, reporting, polls and the reactions they anticipate.

Compare three relationships: what people privately think, what they believe others think and what they are willing to express. This helps reveal how an apparently shared position develops.

False consensus

Ross, Greene and House’s 1977 research examined an egocentric bias in estimates of how common one’s own responses are. A familiar network can make a position appear widely shared, while the wider distribution remains different.

Ask participants for their own view and an estimate of others’ views. Compare those estimates with an appropriate measure of the relevant group’s distribution.

Pluralistic ignorance

Prentice and Miller’s 1993 study of campus alcohol practices investigated a gap between students’ own comfort and their perceptions of peers’ comfort. It illustrates how a perceived norm can differ from privately held views.

For a meeting study, collect confidential individual responses and perceptions of the group. Observe how the meeting’s visible conduct helps sustain or challenge the perceived norm.

The spiral of silence

Elisabeth Noelle-Neumann’s theory links perceptions of the opinion climate, fear of social isolation and willingness to speak. Differences in expression can then alter the climate others perceive.

Investigate the topic, audience, setting and expected consequences. A person may speak in a trusted group and remain silent in a formal meeting. Supportive relationships and commitment to a cause can also sustain expression.

Media representations of the public

A montage of interviews or a prominent comment thread offers a visible selection of voices. Ask how the selection was made, whom it includes and what audiences infer from it.

A study can compare exposure to different presentations with estimates of the opinion climate and willingness to speak. Treat these as distinct outcomes and examine their relationship.

Read Ross, Greene and House (1977); Prentice and Miller (1993); Noelle-Neumann (1974).

A credible estimate connects a clear question with a defined population.

An opinion poll estimates responses to specified questions within a target population and period. Its interpretation depends on recruitment, coverage, wording, response, analysis and disclosure.

Begin with the population: adult residents, service users, students or another defined group. Explain the sampling frame, the selection or recruitment procedure and how the resulting data support inference to that population.

Sampling and recruitment

A probability sample uses a random selection procedure with known, nonzero selection probabilities for the units covered by the design. A nonprobability sample uses another recruitment route, such as voluntary participation or an opt-in panel.

Either approach can collect responses online. The mode of answering and the method of recruitment answer different questions about the study. Describe both.

Work through a distribution

Suppose a probability survey of 1,000 adult residents records 580 supporting the route proposal, 370 opposing it and 50 uncertain. In this unweighted example, support is 58%, opposition 37% and uncertainty 5% of all respondents.

Among only those giving a support or oppose response, support is 580 ÷ 950 × 100 ≈ 61.1%. These percentages use different denominators. State the base and retain the uncertain responses in the full report.

Sampling uncertainty

For an illustrative simple random sample of 1,000 from a large population, the maximum approximate 95% margin of sampling error for a proportion is 1.96 × √(0.5 × 0.5 ÷ 1,000) ≈ 0.031, or ±3.1 percentage points.

A repeated-sampling interpretation concerns the procedure: intervals constructed this way would cover the population proportion in about 95% of comparable samples under the assumptions. Real designs require the appropriate calculation for weighting, clustering and other design features.

Wording and the judgement being measured

Schuman and Presser’s research examines how question form, wording and context enter survey responses. Define the proposal clearly, ask one question at a time and offer response categories appropriate to the judgement.

Compare “Do you support changing Route A to use the new interchange?” with “Do you support the efficient new route that will improve journeys?” The second embeds a favourable evaluation within the question. A split-ballot experiment can investigate the difference produced by wording.

For a multilingual study, check that each version presents the same issue and response task. Cognitive interviews can reveal how participants understand the wording and which assumptions they bring to it.

Nonresponse, weighting and subgroups

Examine who participates and how respondents differ from the target population. Weighting adjusts the contribution of responses according to stated benchmarks and assumptions. Assess the resulting balance and uncertainty.

Report the sample supporting each subgroup estimate. An overall sample of 1,000 may contain a much smaller number of older daily passengers, so the precision of their estimate needs separate attention.

Change over time

Compare wording, population, recruitment, mode and analytical procedures across waves. Assess a difference with its appropriate uncertainty. A panel follows the same participants; repeated cross-sectional surveys draw new samples from a defined population.

Methodological guides: Schuman and Presser (1981); AAPOR’s best practices; AAPOR on uncertainty measures.

Individual responses and collective trends can follow different patterns.

Public opinion can display continuity while individuals revise their responses. Aggregate change can also emerge through new information, changing circumstances, shifts in who participates or changes in the composition of a population.

Converse and the organisation of belief

Philip E. Converse’s work on belief systems in mass publics examined differences in political knowledge, ideological organisation and the stability of responses. It raised influential questions about how survey answers relate to underlying beliefs.

Use those questions to investigate an issue: how much information does a person have, which considerations organise the view and how does the response behave across comparable occasions?

Page and Shapiro: collective patterns

Benjamin I. Page and Robert Y. Shapiro’s The Rational Public (1992) examined long-term trends in Americans’ policy preferences. They argued that collective opinion can show coherence and respond meaningfully to events and information even amid individual fluctuation.

For another context, examine comparable trend data and the information environment. Explain which population and period support the conclusion. Separate collective continuity from the behaviour of particular respondents.

Wlezien’s thermostatic model

Christopher Wlezien’s model examines how preferences for more or less policy activity can respond to policy itself. As provision changes, demands for further increases or decreases may adjust.

The thermostat analogy highlights feedback. Suppose a service expands and fewer residents ask for additional expansion. Investigate whether the desired level has been approached, whether needs changed and which other factors entered the response.

Responsiveness and unequal influence

Policy responsiveness concerns the relationship between public preferences and institutional action. Examine which preferences are heard, which changes occur and the time through which the relationship develops.

Gilens and Page’s 2014 analysis of U.S. policy cases examined the independent influence associated with average citizens, affluent citizens and organised interests. Their findings directed attention towards inequalities in representation within that dataset and setting.

A study of the route proposal can compare survey distributions with meeting participation, organisational access and decision records. Follow the reasons officials give and the revisions that enter the plan.

Read Converse; Page and Shapiro (1992); Wlezien (1995); Gilens and Page (2014).

Visible participation and informed judgement require different kinds of evidence.

Digital media provide routes for expressing views, sharing arguments, gathering support and challenging accounts. Platforms also organise visibility through interfaces, rankings, moderation and opportunities for repeated expression.

Study the relationship among participation, perceived opinion and the wider population. Define whether the evidence concerns posts, accounts, contributors, viewers or people in a target public.

Digital traces and the public they describe

Ruths and Pfeffer’s work on social media research examines the methodological problems of drawing behavioural conclusions from platform data. Selection, access to data and the organisation of the platform affect what becomes observable.

Suppose a proposal attracts 10,000 comments. Identify the number of distinct contributors, the collection rules, repeat participation and the relevant population. Analyse the arguments and expressions within that dataset, then justify any wider inference.

Sentiment analysis can classify language under a model. Inspect quotation, irony, local usage and multilingual variation through validation suited to the question. A negative expression can oppose the proposal, criticise its implementation or challenge another contributor.

Deliberation and considered judgement

James S. Fishkin’s work examines public judgement under conditions that provide information, discussion and opportunities to question competing arguments. Deliberative polling connects recruitment with a structured process of consideration and repeated measurement.

For a civic proposal, examine how materials are prepared, how different positions are represented and how participants can speak and question. Compare preferences, knowledge and reasons before and after the exercise.

Democratic evaluation

Ask who can enter the conversation, whose experience is recognised, how disagreement is handled and how institutions explain their decisions. A numerical majority, a reasoned discussion and an accountable response each answer part of the democratic question.

Public opinion research becomes useful when it makes these relationships visible. It can show where views converge, where they remain divided and which changes would allow a broader public to contribute.

Read Ruths and Pfeffer (2014) and Fishkin (2018).

Take a claim about “the public” back to its evidence.

For each case, define the issue, the relevant public and the relationship being claimed. Develop an answer before opening the discussion.

Case 1 · Two headlines from one survey

Suppose a survey records 580 supporters, 370 opponents and 50 uncertain respondents among 1,000 adult residents. One headline reports 58% support. Another reports about 61% support. A third announces “strong public enthusiasm”.

Your task: examine the denominator behind each percentage and the evidence required for the claim about enthusiasm.

Case 2 · A room that sounds unanimous

Suppose a college meeting appears to support a proposed timetable. In confidential interviews afterwards, many students express reservations and say they believed others were satisfied. Several also describe concern about appearing uncooperative.

Your task: distinguish the private distribution, perceptions of the group and the process of public expression.

Case 3 · A campaign enters a policy decision

Suppose residents organise around the bus-route proposal. A poll indicates broad support for improving access but divided views on the plan. A business association receives several meetings with officials. A revised route appears, and campaigners claim that public opinion determined the decision.

Your task: propose evidence about responsiveness, unequal access and alternative explanations.

Listen for the movement from individual views to public judgement.

Watch my lecture on Public Opinion. Follow the contrast between quantitative and qualitative conceptions, the discussion of Lippmann and Dewey, and the questions about media, opinion climates and political influence.

Watch my Public Opinion lecture on YouTube ↗

After watching, write a 200-word explanation of one public opinion claim. Identify the issue, the public, the evidence and one additional relationship that would deepen the account.

Choose a claim that the evidence can support.

Choose one answer for each question before opening the explanation. Return to the relevant section to clarify any distinction that remains uncertain.

Connect the distribution of views with the process you want to explain.

A strong study begins with a bounded question about a particular issue and public. Define the judgement, the relevant population, the period and the relationship under investigation.

Activity 1 · Audit a public opinion claim

Choose a published claim about a civic, cultural or educational issue. Locate its underlying evidence and record what the claim concerns.

1. Name the public. Identify the target population and the people actually represented in the evidence.
2. Specify the judgement. Distinguish direction, intensity, salience and conditions of support.
3. Inspect the method. Record recruitment, wording, dates, analysis and uncertainty where available.
4. Compare claim and evidence. Identify what the evidence establishes and which additional questions remain.
5. Write a supported account. Preserve the denominator, disagreement and scope of inference.

Suggested output: an 800-word audit and a short rewritten report of the finding. Include a direct link to the underlying study or methodological disclosure.

Activity 2 · Compare private views with perceived views

Choose a low-risk campus or community issue and obtain informed consent for a small exploratory exercise. Ask participants privately for their own view, their estimate of the group’s views and their willingness to speak in a specified setting.

Compare the three measures and interview a few participants about their estimates. Explain the recruitment route and the limits of the sample. Protect responses that could identify individuals or expose a private disagreement.

Suggested output: a comparison table and a 700-word interpretation. Treat the exercise as evidence about the participants studied and justify any wider claim.

A combined design for the bus-route proposal

Use a survey to estimate views and their dimensions in a defined population. Interview people with different experiences of the service. Analyse media accounts and observe relevant meetings with appropriate access and consent.

Trace submissions and decision records, then follow later travel experience. This connects distribution, interpretation, expression and institutional response within one inquiry.

Strengths and limits of the main methods

Surveys compare standardised responses across a sample; their interpretation depends on design and measurement. Interviews reveal reasons and experience; the interview situation shapes the account. Observation follows interaction; access and researcher presence delimit what can be seen.

Experiments can strengthen inference about a defined manipulation; the setting and participants shape how far the findings travel. Records help trace decisions; their completeness and purpose need examination. Combine methods where they address a clear relationship.

Reflexive research note

Record your own position on the issue and the assumptions it brings. Include perspectives that challenge the initial account. Explain how participants were selected and how decisions about coding, categories and interpretation were made.

Examination and seminar prompts

• Compare quantitative and discursive conceptions of public opinion.
• Discuss Lippmann and Dewey through a contemporary media problem.
• Distinguish direction, intensity, salience, stability and ambivalence.
• Explain false consensus, pluralistic ignorance and the spiral of silence.
• Evaluate a poll’s claim to represent a public.
• Examine how media coverage and public preferences can shape one another.
• Assess the evidence needed to establish policy responsiveness.

Ask whose views are counted, whose reasons are heard and whose preferences matter.

Key terms

Public opinion

The distribution, formation and expression of views about issues of shared concern.

Public

People defined through a relationship with an issue, its consequences, discussion or a study’s population.

Issue public

People with sustained interest or involvement in a particular issue.

Attentive public

People who follow public affairs with relatively sustained attention.

Opinion

A judgement or position on a question, often articulated in a response.

Attitude

An evaluative orientation towards an object, person or issue.

Belief

Something a person accepts as true.

Value

A broader principle or goal regarded as desirable.

Direction / valence

The position taken, such as support, opposition or a qualified response.

Intensity

The strength with which a position is held.

Salience

The prominence or importance of an issue for a person or public.

Stability

Consistency across time and comparable measurements.

Ambivalence

The coexistence of competing evaluations within a judgement.

Conditionality

The circumstances upon which a preference depends.

Quantitative conception

Understanding public opinion through the distribution of individual responses.

Discursive conception

Understanding public opinion through discussion, judgement and the formation of collective positions.

Pseudo-environment

Lippmann’s concept for the represented world through which people interpret circumstances and act.

Socialisation

The acquisition of orientations, expectations and practices through social experience.

Reference group

A group supplying a standpoint for comparison and judgement.

Elite cue

A signal from a prominent actor or organisation that helps people interpret an issue.

Receive–accept–sample

Zaller’s framework connecting exposure, acceptance and accessible considerations with preference statements.

Agenda-setting

A process through which media emphasis can shape issue prominence or importance.

Framing

Selection and emphasis that organise the interpretation of a problem, cause, evaluation or remedy.

Priming

A process affecting the weight of considerations used in an evaluation.

Opinion climate

A perceived pattern of prevailing views in a relevant setting.

False consensus effect

Overestimation of the relative commonness of one’s own response or position.

Pluralistic ignorance

A shared misperception of others’ prevailing views, often sustained by public conduct.

Spiral of silence

A process linking perceived support, anticipated isolation and willingness to express an opinion.

Target population

The people or units about whom a study seeks to make claims.

Sampling frame

The list or structure through which units can be identified and selected for a sample.

Probability sample

A sample selected by random procedures with known, nonzero selection probabilities under the design.

Nonprobability sample

A sample recruited through another procedure, such as voluntary participation or an opt-in panel.

Coverage error

Error associated with differences between the population and the units covered by the sampling frame.

Nonresponse error

Error arising when participation patterns leave relevant differences inadequately represented.

Weighting

Adjustment of responses’ contributions according to stated benchmarks and assumptions.

Margin of sampling error

A measure of sampling uncertainty under a specified probability design and calculation.

Total survey error

A framework examining sampling and nonsampling sources of error together.

Panel study

A study following the same participants across occasions.

Repeated cross-section

Successive samples drawn from a defined population to examine distributions over time.

Deliberation

Consideration of reasons and competing arguments through an organised communicative process.

Policy responsiveness

The relationship between public preferences and institutional action.

Policy feedback

The influence of policy and its consequences on subsequent experience, preferences or participation.

Further reading

Begin with the classic questions about public judgement. Move through opinion formation, media influence, measurement and the relationship between preferences and policy.

1. Key, V. O., Jr. (1961). Public opinion and American democracy. Alfred A. Knopf. Read ↗Begin with the relationship between public views and government.
2. Lippmann, W. (1922). Public opinion. Harcourt, Brace and Company. Read ↗Study mediated representations, stereotypes, news and organised intelligence.
3. Dewey, J. (1927). The public and its problems. Henry Holt and Company. Read ↗Examine consequences, inquiry, communication and public formation.
4. Blumer, H. (1948). Public opinion and public opinion polling. American Sociological Review, 13(5), 542–549. Read ↗Investigate public opinion as an interactive and organised social process.
5. Price, V. (1992). Public opinion. Sage. Read ↗Compare the intellectual traditions and concepts of public opinion research.
6. Converse, P. E. (2006). The nature of belief systems in mass publics (1964). Critical Review, 18(1–3), 1–74. Read ↗A reprint of the 1964 essay on belief organisation, political knowledge and response stability.
7. Zaller, J. R. (1992). The nature and origins of mass opinion. Cambridge University Press. Read ↗Connect political messages, predispositions and accessible considerations.
8. Page, B. I., & Shapiro, R. Y. (1992). The rational public: Fifty years of trends in Americans’ policy preferences. University of Chicago Press. Read ↗Examine the organisation and change of collective preferences in a defined historical setting.
9. McCombs, M. E., & Shaw, D. L. (1972). The agenda-setting function of mass media. Public Opinion Quarterly, 36(2), 176–187. Read ↗Study the relationship between media emphasis and issue salience.
10. Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51–58. Read ↗Analyse how selection and emphasis organise public interpretations.
11. Iyengar, S., & Kinder, D. R. (1987). News that matters: Television and American opinion. University of Chicago Press. Read ↗Read experimental research on agenda-setting, priming and political judgement.
12. Noelle-Neumann, E. (1974). The spiral of silence: A theory of public opinion. Journal of Communication, 24(2), 43–51. Read ↗Examine perceived opinion climates and willingness to speak.
13. Ross, L., Greene, D., & House, P. (1977). The “false consensus effect”: An egocentric bias in social perception and attribution processes. Journal of Experimental Social Psychology, 13(3), 279–301. Read ↗Investigate estimates of how common one’s own responses are.
14. Prentice, D. A., & Miller, D. T. (1993). Pluralistic ignorance and alcohol use on campus: Some consequences of misperceiving the social norm. Journal of Personality and Social Psychology, 64(2), 243–256. Read ↗Compare private views with perceptions of others’ views.
15. Schuman, H., & Presser, S. (1981). Questions and answers in attitude surveys: Experiments on question form, wording, and context. Academic Press. Read ↗Study how a survey question and its context enter the response.
16. American Association for Public Opinion Research. (n.d.). Best practices for survey research. Read ↗Connect research objectives with design, measurement, analysis and reporting.
17. American Association for Public Opinion Research. (2012). Understanding a “credibility interval” and how it differs from the “margin of sampling error” in a public opinion poll. Read ↗Distinguish design-based sampling uncertainty from model-based uncertainty.
18. Wlezien, C. (1995). The public as thermostat: Dynamics of preferences for spending. American Journal of Political Science, 39(4), 981–1000. Read ↗Examine feedback between policy provision and preferences for more or less.
19. Gilens, M., & Page, B. I. (2014). Testing theories of American politics: Elites, interest groups, and average citizens. Perspectives on Politics, 12(3), 564–581. Read ↗Investigate unequal influence in the U.S. policy cases analysed.
20. Fishkin, J. S. (2018). Democracy when the people are thinking: Revitalizing our politics through public deliberation. Oxford University Press. Read ↗Examine considered judgement and the design of public deliberation.
21. Ruths, D., & Pfeffer, J. (2014). Social media for large studies of behavior. Science, 346(6213), 1063–1064. Read ↗Assess inference from platform data to human behaviour and wider publics.

My lecture: Public Opinion Explained ↗

Six statements to retain

1. A public opinion claim needs a defined public, issue, period and relationship.
2. Direction, intensity, salience and conditions of support reveal different aspects of judgement.
3. Opinions form through experience, information, relationships and accessible considerations.
4. Private views, perceived opinion and public expression can differ.
5. Polls estimate responses under a design; interpretation depends on measurement, recruitment and uncertainty.
6. Democratic inquiry examines both the distribution of views and the unequal routes through which they enter decisions.