Two-step Flow of Information

While Gatekeeping is relatively easy to implement in traditional media such as newspapers and television, it is much harder to do so in social networks like Twitter. In the former, the editors and desk chiefs are a choke point for all disseminated information, and in the latter, the information source is a large number of people.

It is interesting to explore the information flow in such a setting. In general, we observe two groups – content producers and consumers. Those groups are not clearly segregated. A producer can also be a consumer and vice versa. To generalize, a small group of people, the “influencers,” generate much of the content, and a large group of people consumes it, the “consumers.” This leads us to an interesting discovery – namely that to be influenced or form an opinion. A consumer will take the information from somebody they trust and follow – the “influencer.” The “influencer” themselves may have been influenced by another “influencer,” mass or other specialized media.

As they say, “everything new is just well forgotten old,” this Two-step Flow of information was first hypothesized by Lazarsfeld in the late forties and developed further in the fifties (Katz & Lazarsfeld, 2006). In this theory, what we called “influencers” is called “opinion leaders.” It stipulates that most people do not form their opinion by direct influence by mass media, but instead, they are influenced by a limited number of opinion leaders whom they trust. In this case, the opinion leaders convey information they have acquired and put it in context and provide their interpretation and opinion. Those can be family, friends, spiritual leaders, and the like. As opinion leaders appear to be more exposed to mass media, they not only draw information from it but also have other “special” sources, like professional or specialized media. Particularly interesting is the “drug study” discussed by Lazarsfeld, where a highly coherent group of opinion leaders – in this case, medical doctors, formed the same medical opinion and started prescribing the same medication at approximately the same time without any prior coordination, and solely as a result of being exposed to publications in a narrow set of specialized medical journals.

Going back to Influence Operations, it appears that it would be relatively easy to influence a group of doctors on any medical issue by ensuring the “right” articles are published in a couple of the most prestigious magazines. And even if those articles are later retracted, their influence will persist because of some of the other media and cognitive effects (like Cultivation), which we will discuss later.

As it is easy to conduct the aforementioned operation with a group of opinion leaders who are highly coherent, it is not easy to do the same on social media with people from all clothes of life, different education levels, and types.

There is a big difference in how their constituents receive the opinion leaders. While it is universal around most of the physical world that a medical doctor's direction is good for your health, this is not the same with your favorite influencer online. Different people need a different approach, so in order to influence people online, it is far more efficient to have a larger number (Bastos et al., 2013) of less popular influencers(Bakshy et al., 2011; Bastos et al., 2013).

In summary, the information from mass and social media appears not to be directly ingested by most of the influenced. Instead, a Two-step Flow is discovered where information flows from an original source through intermediary opinion leaders, and then this information is consumed by the majority of the subjects mainly through the power of social connections. It is important to note that other opinion leaders may influence the opinion leaders themselves, and those structures may be hierarchical or peer.

Because of this highly distributed system, one-to-many approaches like Gatekeeping and Agenda Setting are challenging to implement, and this is when there is a need for a decentralized or many-to-many approach – meet the Spiral of Silence.


Recommended Reading

Bakshy, E., Hofman, J., Mason, W., & Watts, D. (2011). Everyone’s an Influencer: Quantifying Influence on Twitter. In Proceedings of the 4th ACM International Conference on Web Search and Data Mining, WSDM 2011 (p. 74). https://doi.org/10.1145/1935826.1935845
Bastos, M. T., Raimundo, R. L. G., & Travitzki, R. (2013). Gatekeeping Twitter: Message diffusion in political hashtags. Media, Culture & Society, 35(2), 260–270. https://doi.org/10.1177/0163443712467594
Katz, E. (1957). The Two-Step Flow of Communication: An Up-To-Date Report on an Hypothesis. Public Opinion Quarterly, 21(1, Anniversary Issue Devoted to Twenty Years of Public Opinion Research), 61. https://doi.org/10.1086/266687