Two-Step Flow Theory

 

Origins of the Theory: The 1940 Election Study

Aim: Lazarsfeld, Berelson, and Gaudet (1944) set out to discover how and why voters made up their minds during a presidential campaign. They tested whether media alone drove this.

Method: the researchers studied Erie County, Ohio, during the 1940 campaign between Franklin D. Roosevelt and Wendell Willkie. Using a panel design, they interviewed the same voters seven times between May and November. This tracked how each person’s views changed over time.

Results: most voters had settled on a candidate early, based on stable factors like religion and social status. Conversation mattered more than the media, when minds did change.

A minority of voters, more attentive to campaign coverage than their peers, were turned to by others for information; the researchers called these individuals opinion leaders.

Conclusion: Lazarsfeld, Berelson, and Gaudet (1944) concluded that mass media influence works indirectly, in two steps: media to opinion leaders, then opinion leaders to the public. The pattern held throughout the campaign.

Evaluation: the panel design, tracking the same voters over seven months, let the researchers separate media-driven change from socially-driven change in a way a one-off survey could not. But Erie County was one homogeneous county in a print-and-radio era, so the findings may not generalise elsewhere.

Katz and Lazarsfeld’s Elaboration: The Decatur Study

Katz and Lazarsfeld did not treat the 1940 election finding as a finished theory. Over the next decade, they tested it directly.

Aim: to establish whether the two-step pattern held beyond politics, in everyday decisions. Who exactly were the opinion leaders?

Method: reported in their book Personal Influence (1955), the study centred on women in Decatur, Illinois. It examined decisions across four areas: groceries, fashion, cinema, and public affairs.

Respondents were asked whom they turned to for advice, or who turned to them. This self-report approach was later called the self-designating technique.

Results: interpersonal influence again outweighed direct media influence, across all four areas. But opinion leadership itself proved domain-specific, not one general trait.

A woman turned to for fashion advice was not necessarily consulted on public affairs.

Conclusion: Katz and Lazarsfeld (1955) concluded that the two-step pattern was a general feature of everyday life, not an artefact of one election. Opinion leadership, they argued, is domain-specific.

Evaluation: Personal Influence strengthened the model’s claim to generality, showing the pattern held for mundane consumer choices as well as political ones. Its main limitation is scope: the sample was women in one Mid-Western town in the 1940s.

How does it work (media → opinion leaders → wider audience)

The two-step flow model explains that mass media messages don’t usually reach people directly in a powerful way.

Instead, information from newspapers, radio, TV, or today’s social media is first picked up by opinion leaders – individuals who pay close attention to news, trends, and campaigns.

These opinion leaders interpret the content, form their own views, and then share it with people around them through conversations, recommendations, or social influence.

In this way, information flows in two steps: from the media to opinion leaders, and from opinion leaders to the wider audience.

For example, during an election, many voters might not read detailed political coverage themselves but instead rely on a friend, community leader, or influencer who explains the key points.

This process highlights how personal interaction often has a stronger effect on people’s attitudes and decisions than mass media alone.

Not everyone becomes an opinion leader. Research across both founding studies points to a consistent set of traits:

• Media exposure: opinion leaders follow coverage of their topic more closely than their peers.
• Social location: they tend to be well connected and easy to talk to, so their knowledge actually reaches other people.
• Domain expertise: their influence is usually anchored in real or perceived know-how in one specific area.
• Trust and similarity: people turn to them partly because they feel like a credible peer, not a distant authority.

Most opinion leaders are monomorphic: influential in one narrow area, like fashion, but not others, like politics. A rarer polymorphic leader is influential across several unrelated domains at once.

Weimann (1982) later complicated this picture. He found that a marginal position bridging several separate circles predicted opinion leadership better than simple popularity did.

Opinion leaders vs. influencers – are they the same?

Opinion leaders and influencers share similarities, but they are not identical.

• Opinion leaders are individuals who naturally gain trust and authority within their social circles. They are often highly knowledgeable about a topic, follow media closely, and shape how others interpret information. For example, a teacher, community activist, or even a well-read friend could act as an opinion leader without being widely known.

• Influencers, in the modern sense, are people who build audiences on digital platforms like Instagram, YouTube, or TikTok and intentionally leverage their visibility to shape attitudes and behaviors. While many influencers function as opinion leaders, the relationship is more commercialized and can reach a much larger scale than the face-to-face networks described in the original two-step flow model.

Recent research adds real evidence to this comparison. A U.S. consumer survey found that a brand’s corporate social responsibility messaging landed better, and cut scepticism more, when delivered by a social-media influencer (Cheng, Hung-Baesecke, & Chen, 2024).

The influencer had to be seen as a genuine leader in “opinion and taste.” Trust, in other words, still does the work.

That is the same mechanism Katz and Lazarsfeld described for grocery shopping and fashion in Decatur, seven decades earlier, just running through a different channel.

Bause (2021), reviewing the political-communication literature, reached a compatible conclusion from a different angle. Political social-media influencers, he argued, are best understood as a further development of classic opinion leadership, not a wholly new phenomenon.

Both studies point to the same underlying dynamic.

In short: opinion leadership is a social role, while being an influencer is often a profession. Influencers are today’s digital-era version of opinion leaders, but not all opinion leaders are influencers.

Examples of the Model

The two-step flow model can be seen in many areas of everyday life:

• Politics: During election campaigns, most voters don’t read policy documents directly. Instead, they hear simplified interpretations from journalists, community leaders, or friends who follow the news more closely.

• Advertising: Brands often target influential figures – such as trendsetters or celebrities – knowing that their endorsement will shape the opinions and buying decisions of wider audiences.

• Social Media: Today’s influencers on platforms like Instagram, TikTok, and YouTube act as modern opinion leaders. They interpret content, share reviews, and guide how millions of followers think about products, fashion, or lifestyle choices.

• Health Communication: Public health campaigns often rely on trusted figures, such as doctors, teachers, or local leaders, to interpret and spread information in a way that communities will accept and act on.

Three of these applications have been studied directly. Norris and Curtice (2008) found the same two-step pattern held for online campaign information around the 2005 UK election. Engaged internet users passed material on to peers who had not gone online themselves.

Carpenter and Sherbino (2010) documented it in clinical medicine. Locally respected “educationally influential” clinicians, identified by their peers, speed up how quickly other doctors adopt new evidence-based guidance.

Nisbet and Kotcher (2009) found a similar pattern in climate communication. Recruiting and training already-trusted community members spread accurate science more effectively than a mass-media campaign alone, though reach and depth of engagement trade off against each other.

Critical Evaluation

Strength: Highlights the role of social influence.

The two-step flow model is praised for recognizing that people are not passive recipients of media but are influenced by social interactions.

This matters because research during the 1940 U.S. election showed voters were more likely to be swayed by trusted opinion leaders than by direct media exposure.

By shifting focus from a “top-down” media effect to interpersonal communication, the model provided a more realistic view of how ideas spread.

Trust matters more than the message.

This makes the theory useful for understanding political campaigning, advertising, and influencer marketing (paid promotion of products or ideas through people with large social-media followings).

It works through intermediaries, not the media alone.

Limitation: Oversimplifies communication flow.

Critics argue that the two-step flow model reduces the complexity of communication to just two stages: media → opinion leaders → general public.

In practice, information can pass through multiple steps, circles, and feedback loops, and people may access media directly without mediation.

Troldahl’s (1966) field test confirmed this: many people got information straight from the media, bypassing any opinion leader entirely.

The model, then, is not that neat.

This simplification overlooks the diversity of communication channels, especially in today’s digital environment.

Because of this, the model may underestimate the direct influence that media—particularly television and social media—can have on audiences, limiting its explanatory power in modern contexts.

Strength: Supported by empirical research.

The theory is grounded in detailed field studies, most notably The People’s Choice (1944) and Personal Influence (1955).

These studies systematically investigated how voting decisions were shaped and consistently found that interpersonal communication mediated media effects.

This empirical base distinguishes the model from more speculative theories like the hypodermic needle model.

The reliance on real-world data gives the theory credibility.

It helped establish communication research as an academic field, making it a cornerstone for later media-effect models.

Limitation: Context-dependent findings.

The original studies were conducted in the 1940s–1950s, in a media environment dominated by radio and newspapers.

The dominance of opinion leaders in filtering information may reflect that specific historical context.

Deutschmann and Danielson (1960) found an early limit to this: fast-breaking news often reached people directly, skipping any opinion leader entirely.

Not every case fits the model.

In today’s media landscape, audiences have direct, instant access to diverse sources of information, from news apps to social feeds.

This raises doubts about how far the two-step flow theory can be generalized across time.

While still useful, it may not fully capture the complexity of 21st-century communication where information spreads simultaneously through mass media, peer networks, and algorithm-driven feeds.

Relevance: Anticipates modern influencer culture.

Despite its limitations, the theory remains highly relevant in explaining how information spreads on social media.

Influencers, bloggers, and online community leaders act as contemporary opinion leaders, interpreting and shaping messages for their followers.

Direct media effects still exist.

But interpersonal influence plays the central role.

The persistence of this pattern means the model still shapes strategy in politics, marketing, and public health.

It remains a foundational theory, even now.

Contemporary Research

Social media lets researchers watch a whole communication network directly, not just what people say they read.

Aim (Hilbert, Vásquez, Halpern, Valenzuela, & Arriagada, 2017): to test whether the two-step flow model still describes how information moves on a digital platform.

Method: the researchers gathered over 150,000 tweets about nine environmental protest movements in Chile. They combined network analysis with qualitative interpretation. The goal: classify how messages moved between ordinary participants, “amplifier” accounts, and official media accounts.

Results: a substantial minority of mentions (39%) passed through amplifier accounts. That is a clear two-step pattern. But media and official accounts received most mentions (80–90%) directly from participants, with no intermediary at all.

Conclusion: the two-step flow model remains genuinely valid on digital platforms. It is not the only pattern, though.

Evaluation: this is an unusually strong test. It tracks who actually communicated with whom, not self-reported recall. Its main limit is scope: one fast-moving protest movement, on one platform. Calmer topics might show a different balance.

Paradigm Shift in Mass Communication Theory

Before the two-step flow model, the dominant idea was the hypodermic needle theory (also called the “magic bullet” model).

That theory assumed media messages were like injections—directly entering people’s minds and producing an immediate, powerful effect on everyone in the same way.

The two-step flow model challenged this view.

Instead of audiences being passive recipients, it showed that people are influenced indirectly through opinion leaders who filter and interpret media content.

This means media effects are not uniform or automatic, but depend on social networks and conversations.

In short, the hypodermic needle model emphasizes direct impact, while the two-step flow model highlights the importance of human interaction in shaping opinions.

Two other traditions sit alongside this comparison. Uses and gratifications theory asks a different question: not what media do to people, but what people do with media.

Katz developed it with Blumler and Gurevitch (1973), treating audiences as active users who satisfy their own needs.

The two ideas are not rivals. Uses and gratifications explains why someone chose the content in the first place; two-step flow explains what then happens to it as it moves through their social network.

Rogers’s (1962) diffusion of innovations shares the same basic premise: socially positioned early adopters, much like opinion leaders, accelerate how a new idea spreads through a population. It applies over months or years, though, not the course of a single campaign.

Cultivation theory offers a third contrast. It concerns the slow, cumulative effect of heavy television viewing on how people see the world, largely setting interpersonal communication aside altogether.