Of the four main models of media effects that developed predominantly in the latter part of the 20th century and are conventionally taught in a-level sociology, three have in their different ways managed to carve-out varying degrees of theoretical relevance in the 21st century:
The 2-Step Flow model, however, seems irredeemably trapped in a media and cultural landscape – late 20th century modernity – that seems to have consigned it to the scarp-heap of irrelevancy in the 21st century digital world.
New research into how information can originate and spread across social media, however, may just have given the general model a new lease of life for a-level sociologists…
Going with the (2-Step) Flow…
The conventional way of seeing the 2-Step Flow model is as an example of a diffusion theory of media, one that broadly argues messages that originate in the media are received by audiences in two main ways:
In this respect, Katz and Lazarfield (1955) argued media messages flowed in two distinct steps:
1. From the media to opinion formers: people who directly received a message, were interested enough to want to relay it to others and influential enough for others to take the message on board.
2. From opinion formers to a mass audience: most people, in other words, received the original media message in a form mediated through influential people in the primary groups to which they belonged (such as family, friends and co-workers).
While the broad ideas underpinning 2-Step Flow – a theory of communication that stressed the significance of active audiences, allied to a recognition of the “importance of informal, interpersonal relations” in understanding media effects – still had some basic resonance at the end of the last century, significant cultural shifts in the early 21st century seemingly served to render the basic model somewhat redundant.
Although the theory was interesting for its time, it was also of its time: a theoretical representation of a media and wider-cultural landscape that was broadly unchanging, hierarchical and closed to all but a relatively small, wealthy and privileged, elite.
In other words, the theory made some sense when both access to and control over the media was tightly-controlled and hugely-restricted – in Britain in the 1950’s, for example, there were two national TV stations, one owned by the State, that broadcast for a few hours each evening, a small number of radio stations operated by the State-run BBC, a dozen or so newspapers that broadly reflected a similar, consensual, view of British society and so forth – but made much less sense in a contemporary mediascape where media access was both loosely-controlled and widely-available thanks to the development and growth of the Internet and various forms of social media.
Somewhat ironically perhaps, more-recent shifts in the political and cultural landscape – particularly in countries like America and, to a slightly-lesser extent, Britain – may have given the 2-Step Flow model a new lease of critical life as far as a-level media sociology is concerned.
In the face of contemporary developments in media technology, one line of criticism of the 4 traditional models of media effects has been their tendency to rest on an over-differentiation between “the media” and “the audience”, such that the two are both separate and distinct in terms of structure and role. The media, for example, broadcasts messages and the audience receive and interpret them passively or, in some cases, actively. Critics of these conventional models argue this no-longer holds true for a mediascape dominated by social media where broadcaster and receiver are frequently interchangeable.
The basic argument here is that because modern social media platforms, such as Facebook or Twitter, provide free and open access to all content generators this creates flat user spaces that are broadly “democratic”, in the sense that no one user-voice is unduly privileged over any other; each user-voice, in other words, competes in a democratic market place to be heard.
This does, however, beg the question of the extent to which such characterisations are themselves guilty of under-differentiating media and audience, in the sense of arguing they are one and the same. If social media, for example, is reconceptualised as a hierarchical rather than flat space it follows that some users will be able to generate and spread content in ways that mimic more-traditional forms of media / audience differentiation. If this is the case it’s possible to theorise how a broad modification of the basic 2-Step Flow model can be applied to understand a particular, distinctive and contemporary form of media effect.
In their original characterisation, Katz and Lazerfield argued the key element in the concept of a 2-Step Flow of mediated information was an audience’s involvement in primary social groups where media messages were discussed.
The key difference between a traditional 2-Step Flow and a modified version is that “the media organisation” (such as a newspaper) is replaced by a “new media agent” – an influential originator of media messages on, say, a social media platform, that are then disseminated to a highly-receptive audience of followers. Receptive followers, in turn, disseminate such messages to a less-involved, but still broadly receptive, wider audience and to understand how and why this works, consider the following example:
Dwoskin (2021) details a study undertaken by Facebook looking at the prevalence and spread of “vaccine hesitant beliefs” among users on the platform. It involved “dividing users, groups and pages into 638 ‘population segments’”, where each could potentially cover 3 million users. This meant “the study could examine the activity of more than 1 billion people” – a huge sample size by any measure.
The platform could accurately follow how information was published and republished (“shared”) by specific users / groups because of the way it tracks user behaviour (both on and off the platform – Facebook likes to know where users go once they leave): each and every “like” or “share” made by any given user is, for example, recorded and tracked which means it was possible to understand how information about, in this instance, Covid-19 vaccines, was liked and shared.
One of the most interesting aspects of the research was the finding that 50% of the “vaccine hesitant” content (a euphemism for varying degrees of vaccine-antagonism, from conspiracy theories about mind control to simply being unsure if the vaccine had been properly tested) shared on the platform originated from “just 10 of the 600+ population segments”.
Within this small subset “just over 100 users contributed 50% of all vaccine hesitant content”.
The significance of this research is that it suggests a process very similar to the conventional 2-Step Flow of media information, albeit one modified to take account of the fact that those originating / initially sharing content were platform users themselves, rather than media organisations:
In this instance, therefore, the process involves Step 1 influencers / opinion formers selecting information about covid-19 vaccines from the vast amount – both true and false – generated around the world and presenting it to their followers as “factual” (even where it is objectively false).
Step 2 involves followers taking and generally accepting this (highly-partial) information and then passing it on to others by sharing the information.
While Facebook is just one of a number of platforms (albeit the largest), the Election Integrity Project found a similar process at work across a range of other social media (such as Twitter, Instagram, YouTube and TikTok).
As Dwoskin notes: “The results from Facebook’s research track with findings from disinformation researchers, who have pointed that a small minority of people, particularly influencers and people who post frequently…can have an outsize impact on the conversation and act as super-spreaders of…information.”
What this research tentatively suggests, therefore, is that although the traditional 2-Step Flow model, as elaborated by writers such as Katz and Lazerfield, may have had its day, a similar 2-Step (Digital) Flow may have some limited (exam) currency in highly-specific media contexts. This is particularly true for hierarchical social media platforms such as Twitter, that allow verified users (so-called “blue tick” users) to display the fact they “are who they say they are” to others on the platform, thereby increasing their status as opinion formers in the eyes of different types of follower.