Warning: Declaration of thesis_comment::start_lvl(&$output, $depth, $args) should be compatible with Walker::start_lvl(&$output, $depth = 0, $args = Array) in /home2/leehopki/public_html/blog/wp-content/themes/thesis/lib/classes/comments.php on line 155

Warning: Declaration of thesis_comment::end_lvl(&$output, $depth, $args) should be compatible with Walker::end_lvl(&$output, $depth = 0, $args = Array) in /home2/leehopki/public_html/blog/wp-content/themes/thesis/lib/classes/comments.php on line 155

Warning: Declaration of thesis_comment::start_el(&$output, $comment, $depth, $args) should be compatible with Walker::start_el(&$output, $object, $depth = 0, $args = Array, $current_object_id = 0) in /home2/leehopki/public_html/blog/wp-content/themes/thesis/lib/classes/comments.php on line 155

Warning: Declaration of thesis_comment::end_el(&$output, $comment, $depth, $args) should be compatible with Walker::end_el(&$output, $object, $depth = 0, $args = Array) in /home2/leehopki/public_html/blog/wp-content/themes/thesis/lib/classes/comments.php on line 155
How innovation spreads through a community via communication

How innovation spreads through a community via communication

by Lee Hopkins on July 29, 2008

in academic research,nonverbal communication,second life & 3d virtual worlds,tools

the diffusion of innovation through communication is very dependent on who is telling the story to whom

"Can you back that up with empirical data?" asked one of the delegates.

I was running a Social Media Strategy workshop in Sydney for the PRIA and had been pontificating that early adopters of technology (such as Second Life residents) were also likely to be ‘influencers’ within their communities and social spheres.

So I was ‘called to task’ on my claim, asked to substantiate it with something more scientifically sound and rigorous than just my own hearsay-based opinion.

I was able to point to Sherry Turkle‘s work on the behavioural characteristics of computer users, showing that many were counter-intuitively also likely to be active in large social networks and to have some power of influence and persuasion. Indeed, another delegate (who has similarly looked at this question) was the first to answer the question and point to Turkle’s work.

And Turkle is not alone in looking at this.

The late Ev Rogers spent an academic lifetime looking at how innovations spread through communities and what role ‘communication’ plays in the process; along the way he influenced and guided a large number of doctoral students to do likewise.

Albert Bandura, noted social psychologist most famous for his work on Social Cognition, once commented that "The worth of a theory is ultimately judged by the power of the change it produces." Arvind Singhal and James Dearing at the School of Communication Studies at Ohio University pointed out that (as at 2006) Rogers’ book Diffusion of Innovations was the second most cited book in the social sciences. Singhal subsequently writes, in a foreword to a chapter written by Bandura, that using Bandura’s own measure of a theory’s importance, Bandura’s own Social Cognition work and Rogers’ work on diffusion would rank close to the top.

As part of a substantial body of academics who, since the late 1950s, have looked at the communication of innovation and its subsequent adoption, Thomas W. Valente PhD provided a fabulous review of the field in Singhal & Dearing’s 2006 collection of invited essays reflecting on the legacy of Rogers’ work and life.

According to Valente’s cursory review of communication network models, there have been four major types of model developed to explain how innovations diffuse through a community:

  1. Interpersonal influence: starting with the earliest research, interpersonal models of influence were supported by evidence that ‘leadership’ can strongly influence the adoption of innovations and that greater density and ‘connectedness’ within communities can help the innovation diffuse faster. Thus, those who were rated by their peers as ‘thought leaders’ and who their peers turned to for advice were early adopters of innovations if and when these thought leaders considered that the innovations were compatible with the culture of their community.
  2. Structural models: researchers found that key ‘bridges’ (individuals or organisations, depending on the population size) and the structural characteristics of a network can affect diffusion. Thus, thought leaders who adopted a behaviour were quickly mimicked by others in the community. However, this was mitigated or influenced by an individual’s ‘Adoption Threshold Level’ — the number of prior adopters needed for a person to themselves adopt. Additionally, a person of influence can be a ‘bridge’ between two communities not otherwise connected. Even though that person may be considered a ‘weak’ link between these two otherwise disparate communities, because of their influential position they are able to bridge them both and allow innovations from one cross over to the other.

    Stanley Milgram is most well known for his work on how people respond to and with authority — for example in ordering others to give life-threatening electric shocks — but less well known but equally respected academically for his ‘Small World’ experimental work where, as a social psychologist at Harvard University, he found that, around the world, informal social networks are structured in such a way that only six steps separate everyone from everyone else — the infamous "six degrees of separation." or six pixels of separation if you are my colleague Mitch Joel :-)

  3. Critical points: researchers further refined their data to conclude that there are key ‘moments’ and ‘thresholds of adoption’ that influence when and how innovations and behaviours are taken up by the wider community. Moreover, once that ‘tipping point’ is reached, either for the individual or the system, diffusion is likely to be self-sustaining.
  4. Dynamic interplay: there is argued to be a dynamic interplay between where an individual or an ‘influencer’ sits within a hierarchically-arranged network and whether and when an innovation or new behaviour is adopted. Thus the higher the social influence position an individual holds, the more likely others are to mimic their innovation-driven behaviour; at the same time, the higher an individual’s social influence the more likely they are to adopt an innovative behaviour.
  5. Valente also offers a fifth, nascent, model: that often the ‘message’ itself is of less importance than the messenger. By targeting those individuals who are influential within their community of interest, and supporting and encouraging them in supporting and encouraging others to change their behaviours, the more likely the changes are to diffuse and ‘stick’, and the faster that diffusion takes place.

    Of course, this is the classic Change Management model as taught at every Change Management programme that I have ever willingly attended or else unwillingly been subjected to. Find an enthusiastic evangelist who is also (a key point) influential in the community of interest, give them the tools and training to modify their own behaviours and belief systems, then give them tremendous levels of support and encouragement as they actively work to change the behaviours and belief systems of their peers. Done well it can be very successful in bringing innovative change to a community, as we all know. Done badly, as it usually is in my experience, and the adoption of innovation is a difficult labour with a highly-exposed risk to sabotage and failure.

    Valente argues that there are diagnostic tools that can identify who these change agents are, whether in a community adoption occurs best by one of the aforementioned four models, whether certain individuals display more or less susceptibility to innovations, and who displays more or less ‘infectiousness’ or the ability to influence others. As he says,

"These diagnostics will help promoting organizations tailor their promotions more effectively and focus their activities more on who gets the message and who delivers it rather than what the message says."

This is because the ‘song’ will never remain the same: at each interaction and at each step in a network the messag
e will change — sometimes blatantly, sometimes subtly.

The bottom line

To answer the question posed at the beginning of this post, and during my workshop, there IS a whole body of research on the power of early adopters to influence their wider communities.

So next time you look at innovations like Second Life (as we did) and say to yourself, "That will never take off", take a look at who is actually playing with the innovation — if they are influential and if they consider that the innovation will be of benefit to and likely to ‘fit’ within the culture of their community the chances are very strong that that adoption of that innovation will spread and spread and spread, until it reaches Gladwell’s ‘tipping point’ and becomes an almost unstoppable force.



Further information

For those keen to dig deeper into this, Valente showed the lineage of these diffusion network models in a table, replicated below.

Concept Description Publication
Social factors Social factors such as media exposure, discussion with friends are more important than economic ones such as wealth Ryan and Gross, 1943
Integration People well connected to the social system adopt innovations earlier than those on the periphery Coleman, Katz and Menzel, 1966
Opinion leaders People who are sought out by others for advice adopt earlier and influence others to adopt Rogers and Cartano, 1962; Rogers, 1962
Norms Community norms affect whether opinion leaders will be early or later adopters Becker, 1970
Weak ties Weak ties — links that connect otherwise disconnected groups — facilitate discussion Granovetter, 1973
Thresholds People adopt based on how many others they see adopting Granovetter, 1978
Structural equivalence People are influenced in their adoption decisions by others who occupy similar positions in the network Burt, 1987
Structural holes Gaps in the network Burt 1992
Small worlds Networks characterised by high levels of clustering yet overall short distances between people Travers and Milgram, 1969; Pool and Kochen, 1978; Watts, 2002
Critical mass, tipping point Diffusion reaches a distinct point where behaviour is self-propelling and is difficult to stop Marwell, Oliver and Prahl, 1988; Markus, 1987; Schelling, 1978; Gladwell, 2000
Network thresholds Thresholds can be calculated at the social network level Valente, 1995
Dynamic models Using event history analysis, susceptibility to the influence of others, and infectiousness, the ability to influence others, can be measured Strang and Tuma, 1993; Myers, 2000
Interventions Using network data to identify change agents to promote behaviour change Valente et al., 2003


And if you get out Singhal, A. and Dearing, J.W. (2006). Communication of Innovation. Sage, London you can find Valente’s article (pages 61-82) and investigate his bibliography.

I also commend other articles/chapters within the book, such as Bandura’s On Integrating Social Cognitive and Social Diffusion Theories (pp. 111-135), marketing guru Philip Kotler’s Social Marketing and the Broadening of Marketing Movement (pp. 136-145), and Adhikarya’s Implementing Strategic Extension Campaigns: Applying Best Practices and Lessons Learned from Ev Rogers, (pp.172-198).

In fact, read the whole thing and be amazed at how on just about every page you learn something new!

Previous post:

Next post: