Back to top

Mass and Interpersonal Communication: Buzz for Behaviour Change

~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~

I Introduction

Publicists, marketers, public relations gurus, and producers all try to
establish "buzz" about their products by creating media campaigns
designed to generate discussion among peers.  Buzz can be an end
point on its own, but is also thought to encourage people to buy
products or adopt behaviors.  

Scholars have studied the interaction of mass media and interpersonal
communication since at least Lazarsfeld's pioneering work in the 1940s
[1].  We may not know much about how media and interpersonal
communications interact, but below are some of the theories and their
implications for designing health communication campaigns.

~ ~ ~ ~Social networks have been documented as important influences on
behavioral decisions ever since the classic 1943 study of factors that
influence farmers' adoption of new technology [2].  Rogers' [3]
diffusion of innovations research has shown that for a wide variety of
behaviours and populations, interpersonal influence continues to be the
most significant single factor influencing a person's decision to
accept new ideas and practices. Others have shown that creating "buzz"
about products helps to get them purchased [4, 5]. Indeed, after people
become aware of a new product, they'll probably discuss it with someone
to find out if they've tried it, and once they themselves try it, if
they like it, they'll tell others. For more evidence that interpersonal
communication is important in behaviour change, please see the
additional resources listed at the end of this article. ~ ~ ~ ~

~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~

II Opinion leaders

The first attempt to formulate a model was the two-step flow hypothesis
by Katz and Lazarsfeld [6, 7].  The two-step flow hypothesis
posited that the media influence opinion leaders who in turn influence
others who are less attentive to media communications. Usually, these
others are thought to be family, friends, co-workers and perhaps
acquaintances, people with whom they are close and have strong
credibility and trust.

Opinion leaders were found to consume more media and were more aware of
current events.  In order to persuade others to follow their
opinions, they used media communications to buttress their arguments.
According to Gladwell [5], these "mavens" use the media to stay
up-to-date on their favourite topics; they freely share this
information with others, and they are seen as credible sources of this

 ~ ~ ~ ~In his book The Tipping Point: How Little Things Can Make
a Big Difference, Malcolm Gladwell [5] talks about how ideas, trends or
behaviours can become "social epidemics" and the ways in which they
spread throughout a society.  He argues that social epidemics
"tip"--that is, get underway--when three circumstances are present: the
involvement of a few well-connected and persuasive individuals
("connectors," "mavens," and "salespeople"); a memorable and persuasive
message ("stickiness"); and favourable social contexts ("the power of
context").  Social trends, ideas or behaviours spread throughout
an audience and are adopted by that audience through this process of
"buzz."  [For reviews of Gladwell's book, see 8, 9, 10]. ~ ~ ~ ~
This, however, may be an oversimplification.  It may be that media
influence opinion leaders who influence others that influence others -
a two-step or even multi-step flow.  Further, it may be that some
opinion leaders influence one or a few others, while others have much
higher multiplier effects, influencing five, ten or even hundreds of

These models, however, neglect to consider a number of other factors regarding the media influence process.

First, it is likely that opinion leaders are influenced by others, as
much as others are influenced by them, and that the media shapes their
messages in accordance with what they think the audience wants to
hear.  In sum, to simply say that A influences B may be stretching
it a bit, when in fact B also influences A.

Second, individuals are embedded within complex social network
structures.  Some people have small networks, while others have
quite large ones.  Some social networks are dense (i.e., their
friends know each other) while others are sparse (i.e., their friends do
not know one another).  Also, the degree of similarity or
difference between a person and his/her social contacts affects the
flow of ideas and behaviors.

Finally, there is variation in risk taking and risk avoidance in the
population.  The amount of influence required for a person to
adopt a behaviour varies.  Some people will adopt new behaviours
when only a minority of their friends or the population has done so;
others wait until a practice is widely accepted before they are willing
to adopt it.

These factors show that the relationship between mass media and
interpersonal communication is complex.  Unlike the simple models
above, it is likely that people attend to media communication, then
interpret and talk about it in unanticipated ways.  

Anti-drug campaigns are a good case in point: one study, for example,
found that talking about anti-marijuana ads led youth to report more
pro-marijuana beliefs than did youth who did not discuss the ads with
others [11].  Similarly, prolonged exposure to anti-drug ads and
discussion of the ads with peers may lead youth to perceive the
messages as boring or laughable [12-14].  

In other words, the effect of media communications on individuals is a
function of how the messages are interpreted within the context of
people's social networks - how, with whom, and in what ways the
messages are discussed. As David et al. [11] put it, if "messages
encourage discussion among peers, and such discussion in turn leads to
negative effects, then a media campaign can result in substantial
deleterious effects, perhaps especially among the segment of the
population most likely at risk" [p. 136].

So if the relationship between media and interpersonal communication is
so complicated, how can we design health communication messages that
create not just buzz, but the right "buzz"?

The answer probably lies in

  • mapping social networks to identify truly trusted and credible opinion leaders,
  • conducting formative research, particularly with opinion leaders, and
  • rigorous pre-testing with social networks.

~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~

III Validated Techniques to Identify Opinion Leaders

As discussed above, opinion leaders can be important in creating
"buzz."  The trick is identifying true opinion leaders - those
that are trusted and credible.  Many opinion leaders are
self-selected or identified by project staff with no knowledge of
whether they actually act as opinion leaders for members of the
intended audience.

 ~ ~ ~ ~Evidence shows that who delivers the message, and to whom,
is just as important as the actual message itself. For example, a
recent meta-analysis was carried out on ninety-four studies related to
substance abuse prevention in school settings [15].  It showed
that programs which used peers to deliver health messages were far more
effective than those which did not use peers, or those which used a
combination of teachers and peers to deliver the messages. ~ ~ ~ ~

Valente and Davis have begun work on a system that matches audience
members to the leaders they nominated [16].  Figure 1 shows an
example of how, in a small network, leaders can be identified as those
who receive the most votes as a leader and then groups are constructed
by assigning people to the leaders they nominated.* After asking
community members to identify opinion leaders (normally in the context
of a specific topic), the process has three steps [16]:

1.    Identifying the 10% of individuals who received
the most nominations by the community members. These individuals become
the opinion leaders.
2.    Matching the leaders with the community members,
with each individual being assigned to the leader he or she nominated,
or to the leader she is most closely connected to.
3.    Assigning community members who did not nominate
any leaders or those whom no one nominated in a random fashion or
assigning them proportionally to the more popular opinion leaders.

This process provides "an optimal matching of opinion leaders to the
community members who look to each of them for advice and thus can be
used to accelerate the diffusion process" [16, p. 60].  

A relatively recent study validated this approach: in a peer-led
smoking prevention school program, it was found that students who were
matched with their nominated opinion leaders enjoyed the program more,
had improved attitudes and self-efficacy, and reported decreased
intention to smoke than did students randomly assigned to leaders or
those assigned to leaders by their teachers [17].  The
intervention consisted of eight fifty-minute workshop sessions, which
included discussions, role-playing exercises and interactive games.

An alternative approach, also shown to be effective, is to define
groups first, then select leaders within these groups [18].  While
it may be more efficacious for participants themselves to nominate peer
leaders [16], this approach may not always be appropriate or feasible
for the intervention and audience in question.  Thus yet another
approach is to have "key informants" identify opinion leaders. For
example, Kelly and colleagues [19-21], in their peer-led HIV
intervention, asked bartenders in gay bars to observe their bar patrons
for 10 days and indicate the names of whom they thought were the most
popular with the other men.  The lists were then cross-checked and
names appearing on more than one list were recruited for training as
peer opinion leaders.  

In the work by Kelly and colleagues [19-21], opinion leaders worked
their risk reduction messages into natural conversation with their
peers.  The peer leaders personally endorsed these messages,
thereby lending their familiarity, popularity and credibility as peer
leaders to the messages themselves and helping to influence norms about
safer sex behaviour.  The programs were sustained by training
successive waves of peer leaders.  Their interventions, based in
gay bars in small US cities, helped reduce risky sexual behaviour by
almost 30% from their baseline levels [22, p. 140].  This approach
is in contrast to matching leaders to persons who nominated them as
leaders, which may not be possible in less structured environments,
such as a gay community as opposed to a classroom setting.  

~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~

IV Formative Research

Health communication practitioners already know the importance of
formative research, i.e. analyzing the target audience to secure an
in-depth understanding of how they view the issue at hand.  At a
minimum, it involves assessing their barriers and incentives for
behaviour change as well as their language, cultural cues, and beliefs
about the world.  

If designing messages meant to stimulate "buzz," this research should
include focus groups with opinion leaders, identified by the methods
noted above, to help design messages that engage opinion leaders in a
way that inspires them to talk about the message and that are correctly
understood by the opinion leaders.  

~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~

V Pre-testing With Social Networks

Like formative evaluation, pre-testing draft health communication
materials to determine whether they resonate with audiences, and in the
right way, is not a new concept to seasoned practitioners. 
However, to ascertain the potential for the right "buzz," pre-testing
should be done by exposing people to communications with members of
their social networks and encouraging them to discuss what they've seen
or heard.  

 ~ ~ ~ ~The ideas of "strong" and "weak" ties are typically traced
to the work of sociologist Mark Granovetter [23, 24].  He reasoned
that "strong" ties are those with whom we spend much time, have strong
emotional bonds, and do things for each other; in short, those we might
consider close friends and who are likely to know one another. 
"Weak" ties, on the other hand, refer to those people we consider
acquaintances, and whom are unlikely to know one another.  He
argued that weak ties act as bridges between dense social networks
(such as our own social networks and those networks that our
acquaintances are part of).  These bridges are extremely
important, he argued, because they permit the flow, or "diffusion," of
information between social networks.  In other words, information,
ideas and knowledge can be transmitted between dense social networks
through the bridges between members of each network~ ~ ~ ~

Focus groups are a particularly useful method of investigating exactly
how people discuss a health message [25].  Focus groups with
opinion leaders and their "ties," as identified by mapping, would also
be useful to see whether they talk about the messages in the right
ways.  In cases where such a mapping has not been conducted, at a
minimum, focus groups should be conducted with people who know one
another to determine how they discuss the topic, what they say, and to
whom.  In either case, it is important to understand the ways that
people discuss the health message, to ensure that the message brings
about the intended and expected discussion.  

~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~

VI Conclusion

As we have discussed, the effectiveness of all health messages depends
on how and with whom they are discussed.  Creating the right
"buzz" requires a few important steps:

  1. Mapping social networks to identify truly trusted and credible opinion leaders,
  2. Conducting formative research, particularly with opinion leaders, and
  3. Pre-testing messages with social networks, or at least with groups of people who know each other.

In the commercial marketing world, Silverman [25] recommends an
approach like this. He talks about using focus groups comprised of
experts, opinion leaders and peers to investigate the nature of "word
of mouth" among them.  In reviewing the health promotion
literature for this article, however, we did not find any examples of
formative research that had been conducted with opinion leaders and
their "ties" to see how health messages were discussed. We are very
interested in hearing about examples of this from our readers!

[*Editor's note: because the OHPE is plain text, we cannot include
this image; however, it is available online in the PDF file at (p 60) or by emailing for a Word version of this article]
Figure 1.  Social network of physicians in one community in Illinois in
the mid-1950s from Coleman and others [26].  Optimal assignments
based on leaders being assigned to those who nominated them or are
"closest to" in the network [16].  
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~
~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ * ~ *~
VII References

1.    Lazarsfeld PF, Berelson B, Gaudet H. The people's
choice: How the voter makes up his mind in a presidential election. New
York: Columbia University Press; 1944.
2.    Ryan R, Gross N. The diffusion of hybrid seed corn in two Iowa communities. Rural Sociology 1943, 8:15-24.
3.    Rogers EM. Diffusion of innovations. 5th ed. New York: Free Press; 2005.
4.    Rosen E. The anatomy of buzz: How to create word-of-mouth marketing. New York: Doubleday/Currency; 2000.
5.    Gladwell M. The tipping point: How little things
can make a big difference. Boston, MA: Back Bay Books; 2002.
6.    Katz E, Lazarsfeld P. Personal influence: The part
played by people in the flow of mass communication. New York: Free
Press; 1955.
7.    Katz E. The two-step flow of communication: An
up-to-date report on a hypothesis. Public Opinion Quarterly 1957,
8.    Denham BE. [Review of the book The tipping point:
How little things can make a big difference]. Journalism & Mass
Communication Educator 2001, 55:82-83.
9.    Goldsmith RE. [Review of the book The tipping
point: How little things can make a big difference]. Journal of
Consumer Marketing 2003, 20:71-73.
10.    O'Shaughnessy J. [Review of the book The tipping
point: How little things can make a big difference]. Journal of
Macromarketing 2000, 20:198-201.
11.    David C, Cappella JN, Fishbein M. The social
diffusion of influence among adolescents: Group interaction in a chat
room environment about antidrug advertisements. Communication Theory
2006, 16:118-140.
12.    Cohn J. Don't just say no - Berkeley researcher
Dr. Joel Brown questions the effectiveness of the federal government's
antidrug education policies. San Francisco Bay Guardian. 1998, August
5. Retrieved August 30, 2006, from
13.    Hastings G, MacFadyen L. The limitations of fear messages. Tobacco Control 2002, 11:73-75.
14.    Stewart DW, Martin IM. Intended and unintended
consequences of warning messages: A review and synthesis of empirical
research. Journal of Public Policy and Marketing 1994, 13:1-19.
15.    Gottfredson DC, Wilson DB. Characteristics of
effective school-based substance abuse prevention. Prevention Science
2003, 4:27-38.
16.    Valente TW, Davis RL. Accelerating the diffusion
of innovations using opinion leaders. Annals of the American Academy of
Political and Social Science 1999, 566:55-67.
17.    Valente TW, Hoffman BR, Ritt-Olson A, Lichtman K,
Johnson CA. Effects of a social-network method for group assignment
strategies on peer-led tobacco prevention programs in schools. American
Journal of Public Health 2003, 93:1837-1843.
18.    Buller DB, Morrill C, Taren D, et al. Randomized
trial testing the effect of peer education at increasing fruit and
vegetable intake. Journal of the National Cancer Institute 1999,
19.    Kelly JA, St Lawrence JS, Diaz YE, et al. HIV
risk behavior reduction following intervention with key opinion leaders
of population: An experimental analysis. American Journal of Public
Health 1991, 81:168-171.
20.    Kelly JA, St Lawrence JS, Stevenson LY, et al.
Community AIDS/HIV risk reduction: The effects of endorsements by
popular people in three cities. American Journal of Public Health 1992,
21.    Kelly JA, Murphy DA, Sikkema KJ, et al.
Randomised, controlled, community-level HIV-prevention intervention for
sexual-risk behaviour among homosexual men in US cities. Lancet 1997,
22.    Kelly JA. Popular opinion leaders and HIV
prevention peer education: Resolving discrepant findings, and
implications for the development of effective community programmes.
AIDS Care 2004, 16:139-150.
23.    Granovetter M. The strength of weak ties. American Journal of Sociology 1973, 78:1360-1380.
24.    Granovetter M. The strength of weak ties: A network theory revisited. Sociological Theory 1983, 1:201-233.
25.    Silverman G. The secrets of word-of-mouth
marketing: How to trigger exponential sales through runaway word of
mouth. New York: AMACOM; 2001.
26.    Coleman JS, Katz E, Menzel H. Medical innovation: A diffusion study. New York: Bobs-Merrill; 1966.