Measuring Influence: 3 Paths That Research Supports

This is a summary of a paper presented at this year’s International PR Research Conference. It received a “Top Three Paper for Practical Significance” award from the Institute for Public Relations.

I’m a public relations pro trying to run his own business and help clients achieve their objectives.  Social media companies that claim to have a secret formula for influence have me frustrated on behalf of my clients, who are wanting some answers about influence online. I decided to find out what the academics had to say about it. In a paper I presented at the International PR Research Conference, I lay out the base themes. Here’s what every public relations practitioner ought to know. 

The mechanics of influence online fit broadly into these three categories: Influence as a function of strength, immediacy and number of sources (social impact); influence as a function of membership in a group (group dynamic); and influence related to a person’s position in a social network (network position.)

Without getting too deep into the academic weeds, let’s take closer look at each.

Social Impact

Strength, “the salience, power, importance, or intensity of a given source to the target;” is a huge factor in whether we get influenced, and in identifying who might influence us. Define “strength” in terms of position in society — supervisors, parents, professors, and clergy hold power over us, and are important to us – or by subject matter expertise.  People with similar jobs or academic majors as ours, or colleagues bearing information we desire have great salience to us. A panic-stricken person imploring us to get out of a burning building does so with great intensity, as might a public speaker moving us with eloquence. 

We could measure influence or its potential by examining strength alone, according to power, importance, salience and intensity, either through surveys focusing on influencer, or the influenced.

Immediacy, “closeness in space or time and absence of intervening barriers or filters,” is another factor. Your neighbor, your parent, your friend are likely more influential to you than someone else’s. Research shows that your direct supervisor influences you more than someone higher up the organizational ladder, that people close to you influence you more than others. 

Number of sources, “how many people there are” carrying out influence behavior is also important, but it comes with a big caveat. Someone online with a lot of followers would be more influential than someone with fewer followers, right? Not so fast.  There’s a plateau of influence online, and influence declines even as scale increases – bad news if you believe a lot of followers makes you automatically influential.

Social impact is a good theory for explaining influence, and for figuring out how to measure it, because strength and immediacy might unlock how we discover opinion leaders.

Group Dynamic

Research shows that social pressure brought by members of a group on its members affects adoption of innovation, a good proxy for influence. The extent to which social identity is tied to the group is a marker for influence. Think of a group you’re active in – say, a neighborhood. Ever looked at your lawn and compared it to your neighbor’s? Do you feel compelled to take holiday decorations down about the same time as your neighbors do?

Professionals could measure the strength of identification with a group (for example, in internal communication) as an indicator of influence.

Network Position

Social network analysis, whether in medicine or computer science, looks at “nodes” – intersections – in the network and describes how information (or viruses) naturally flows or naturally stops flowing.  The nodes with the most connections aren’t the most efficient spreaders, the nodes closest to the center of the network are (because there are fewer steps between nodes.)

Think of a gatekeeper—perhaps the CEO’s secretary—who is one step removed from the most powerful individual in a firm. The secretary also is one step removed from the CEO’s direct reports. The capacity for influence is very strong in the secretary, as she is a broker between nodes and groups of nodes.  We could target her as an influencer who could affect both the CEO and the rest of the leadership team, and measure her influence.

I found two studies operating under the same rules, however, that reached opposite conclusions.  In one, the node did a better job of stopping information flow than spreading it more efficiently. In the other, a virus moved quite efficiently from node to node.

Think of opinion leadership – or stakeholder analysis – defining closely each of our audiences and evaluating them as potential influencers based on their position in an organization may prove helpful.

Nothing about this research says you can identify influencers with a single number.  Social Impact, Group Dynamic and Network Position each offer a potential window to understanding how influence works online, and how to measure it.

Sean Williams is owner of Communication AMMO, Inc., and is both a Masters’ student and adjunct faculty member at Kent State University.