Content uploaded by Joann L. Atkin
Author content
All content in this area was uploaded by Joann L. Atkin on Mar 26, 2015
Content may be subject to copyright.
Content uploaded by Joann L. Atkin
Author content
All content in this area was uploaded by Joann L. Atkin on Mar 26, 2015
Content may be subject to copyright.
SPECIAL SECTION:
COMMUNICATION ON ALCOHOL
Are “Drink Responsibly” Alcohol Campaigns
Strategically Ambiguous?
SMITH, ATKIN, ROZNOWSKISTRATEGIC AMBIGUITY
Sandi W. Smith and Charles K. Atkin
Department of Communication
Michigan State University
JoAnn Roznowski
Department of Marketing and Advertising
Western Michigan University
This article applies the concept of strategic ambiguity in examining viewer responses to
brewer-sponsored “responsible drinking” television advertising campaigns. Strategically am-
biguous messages are designed to engender diverse interpretations between varied audience
segments, and these different selective perceptions should translate into relatively uniform pos-
itive corporate images. In this study, teenage and young adult respondents were shown a series
of television spots from two leading alcohol companies. As predicted, there was a high degree
of diversity in meanings of message content and campaign purpose derived by viewers, particu-
larly among less sophisticated teenagers. Moreover, evaluative ratings of messages and spon-
sors were generally favorable and more uniform than interpretive responses. The research dem-
onstrates how seemingly prohealth messages can serve to subtly advance both industry sales
and public relations interests.
Self-regulation and marketing communication tactics are
two strategies employed by national brewers and distillers in
an effort to curb the rising criticism of their product and its
advertising. The nation’s largest brewers, Anheuser-Busch,
Miller Brewing, and Coors, are involved in consumer aware
-
ness activities to fight alcohol misuse issues of drinking and
driving and underage drinking and to promote responsible
consumption of beer by adults who choose to drink.
Anheuser-Busch has run an advertising campaign promoting
“Know When to Say When,” and Coors ran a similar adver
-
tising campaign promoting “Not Now,” and “21 Means 21.”
These industry campaigns may seem similar to public ser
-
vice announcements (PSAs) sponsored by governmental
agencies or public interest groups (e.g., Mothers Against
Drunk Driving) in that they both promote a message on be
-
half of a cause. The “drink responsibly” ad campaigns devel
-
oped by the brewers, however, are usually designed with two
goals in mind: (a) to create a positive image (or reputation)
for the firm and (b) to communicate the firm’s view on a so
-
cial, business, or environmental issue (Belch & Belch, 2001).
Hence, these campaigns may reflect a hybrid of conventional
commercials and PSAs that can be termed private service
messages. On one hand the campaign may create a positive
image or reputation for the firm if target audiences—con
-
sumers (both current and potential), public health advocates,
or legislators—equate the campaigns with being socially re
-
sponsible. But at the same time, the ads may still communi
-
cate messages that promote brand preference and product
consumption.
One theoretical framework presented here for examining
audience responses to “drink responsibly” campaigns is de
-
rived from the strategic ambiguity approach for attaining or
-
ganizational goals. This report begins with a description of
the central ideas of the strategic ambiguity perspective and
HEALTH COMMUNICATION, 20(1), 1–11
Copyright © 2006, Lawrence Erlbaum Associates, Inc.
Correspondence should be addressed to Sandi W. Smith, Department of
Communication, Michigan State University, East Lansing, MI 48824.
E-mail: smiths@msu.edu
how it lends itself to the study of “drink responsibly” ad cam
-
paigns followed by presentation of findings from a study
measuring viewer responses to “drink responsibly” televi
-
sion commercials.
STRATEGIC AMBIGUITY
This perspective is characterized by the strategic and pur
-
poseful use of messages with high levels of abstraction to si
-
multaneously accomplish multiple, and often conflicting, or
-
ganizational goals. Strategically ambiguous messages are
designed to engender different interpretations of the same set
of symbols within and across different receivers (Eisenberg,
1984). Strategic ambiguity is a relational concept arising
through a combination of source, message, and receiver vari
-
ables; it takes into account the entire context of sender goals
and message content, as well as individual differences among
and within receivers. Strategic ambiguity is independent of
perceived ambiguity; in some cases, the message may seem
vague and abstract, but other executions that appear to be
quite clear and literal may nevertheless elicit diverse re-
sponses among various segments of the audience. In analyz-
ing the influence process, attention focuses on the differen-
tial interpretations drawn by receivers in processing the
symbolic representations. This can be differentiated from at-
titudinal ambiguity, which focuses on the meaning of a re-
ceiver’s neutral response on an attitude scale, and from attitu-
dinal ambivalence, which focuses on receivers’simultaneous
and conflicting positive and negative evaluation of an attitude
object (Priester, 2002). In the strategic ambiguity perspec-
tive, a source with multiple goals creates messages that can
engender differential interpretations among different audi
-
ence segments and can engender multiple interpretations of
sender goals within the individual.
There are two primary outcomes of strategic ambiguity:
(a) diversity of message interpretations within and between
audience segments, and (b) widespread consensus in bot
-
tom-line attitudinal outcomes across the overall audience. To
attain diverse responses, an ambiguous message facilitates
inference-making by receivers who “fill in” context and
meaning according to their attitudinal predispositions and
cognitive processing abilities (Bruner, 1964; Capella &
Street, 1989; Kellerman & Lim, 1989; Lord, 1985; Ortony,
1978; Smith, 1995; Wyer & Carlston, 1979; Wyer & Gordon,
1984).
Previous research examining strategically ambiguous
messages supports the idea of diverse message interpretation.
For example, Contractor and Ehrlich (1993) investigated the
impact of strategically ambiguous messages on the develop
-
ment of a multimillion dollar research organization at a major
university and found that the organization’s development was
enhanced by allowing for multiple interpretations among key
constituents. Markham (1996), while studying work commu
-
nication in a small graphics firm, found that management
used intentionally ambiguous messages to foster freedom,
creativity, and flexibility.
In examining persuasive impact on individuals, the out
-
comes of interest are acceptance of advocated positions and
formation of favorable corporate images and attitudes. Stra
-
tegic ambiguity can facilitate selective perception of message
content, preempt counterarguing, and minimize source dero
-
gation; indeed, source attributions may be amplified, particu
-
larly if the organization appears to be acting against its own
best interests (Eisenberg, 1984).
Very few studies have examined ambiguity in the context
of advertising or advertising-related communication. The ef
-
fect of perceived ambiguity, defined traditionally in terms of
unclear meaning, was investigated by Hitchon, Duckler, and
Thorson (1994) in the context of consumer responses to mu
-
sic video commercials for a variety of product categories.
Perceived incoherence decreased attitudes toward the brand
and the ad. If the ads had been able to engender individually
clear interpretations that varied within and across audience
segments (the premise of strategic ambiguity), then attitudes
toward the brands and the ads should have increased.
In addition, Ha and Hoch (1989; Hoch & Ha, 1986) exam
-
ined the role of ambiguity, as defined by the distinctiveness
of brands and the potential for multiple interpretations of
quality. They concluded that consumers are easily convinced
by ambiguous evidence in ads that can not be easily
disconfirmed by them. The findings from this study are more
consistent with the strategic ambiguity approach, as they al-
low for multiple interpretations of quality and uniform posi-
tive outcomes for the ads. Therefore, it appears that strategic
ambiguity can be applied to “drink responsibly” beer cam-
paigns as well.
PURPOSES OF ”DRINK RESPONSIBLY”
CAMPAIGNS
The goals of a “drink responsibly campaign” appear to be, as
previously asserted, a blending of conventional commercial
and public service messages. First and foremost, brewers
may seek to maximize sales of the company brand by in
-
creasing both specific brand share and overall product de
-
mand (Wallack, Dorfman, Jernigan, & Themba, 1993). Like
most marketers, the alcohol companies seek to create and re
-
inforce brand name preference and loyalty by projecting dis
-
tinctive brand imagery and extolling advantageous substan
-
tive attributes. In addition, the ad campaigns are designed to
promote generic benefits of alcohol consumption by portray
-
ing drinking as an attractive and rewarding practice.
Second, a beer company may seek to enhance its corpo
-
rate image and reputation by displaying sensitivity and con
-
cern about the well-being of its customers and by projecting
the company as respectable, prosocial, and commu
-
nity-minded (Atkin, 2002; DeJong, Atkin, & Wallack, 1992).
It also is important for advertising to help diminish drinking
2
SMITH, ATKIN, ROZNOWSKI
inhibitions by depicting alcohol as a socially acceptable and
physically harmless product, thereby attempting to dispel
consumer worries about the negative psychological, social,
moral, and health outcomes. In these ads the companies at
-
tempt to accomplish a selling function without undermining
the corporate reputation or industry responsibility goals.
Current research supports the idea that the establishment and
maintenance of a positive corporate evaluation made by all
corporate stakeholders hinges on corporate credibility
(Fombrun, 1996; Gregory, 1991). This goal is part of an in
-
dustry-wide effort to combat potential societal controls (i.e.,
government regulation) over the marketing of alcohol prod
-
ucts. It should be noted that these two goals are interrelated
and often mutually reinforcing. For example, an enhanced
corporate image tends to strengthen brand preferences and
may also deflect blame and diminish support for government
regulation.
A third, and most manifest, communication goal is to ac
-
tually diminish alcohol problems such as injuries, diseases,
and antisocial acts (Atkin, 2004; DeJong & Atkin, 1995).
Campaigns can do this by encouraging prevention of drunk
-
enness and drunk driving, and this reflects actual concern on
the part of the beer companies.
The brewers’ campaigns mounted in the past few years
can be analyzed using the strategic ambiguity approach that
simultaneously achieves diverse effects on multiple viewing
audiences by using universal appeals rather than narrowly
targeted messages in commercial campaigns. Most funda-
mentally, they may attempt to differentially influence the
mainstream general public and elites (who are most con-
cerned about societal drinking problems) versus younger and
heavier drinkers (who consume most of the product) versus
problem drinkers (who drink excessively or drive drunk),
while giving all audiences the impression that they are shar
-
ing the same interpretations. The campaigns have been criti
-
cized for being too ambiguous to be effective, for mollifying
critics, and serving a public relations function (Abramson,
1991; Beatty, 1997; Beaver, 1997; J. A. Brown, 1991). The
strategic ambiguity approach can show how the campaigns
could serve all of these functions, as well as combating dan
-
gerous drinking practices and selling the product.
Strategically ambiguous messages are designed to engen
-
der different interpretations within and between audience
segments but with widespread consensus in bottom-line out
-
comes. Therefore, it is expected that the ultimate outcome of
“drink responsibly” messages will be a widely shared posi
-
tive impression of the sponsoring companies and their cam
-
paigns. This is the pattern of responses that is predicted by
the strategic ambiguity approach (Eisenberg, 1984). Hence,
based on the preceding discussions, the following hypotheses
are offered:
H1: Interpretations of message content and campaign
purpose will be highly diverse, as manifested by an
evenly spread distribution of answers across the full
range of multiple response categories on the items
measuring ad appearance, drinking portrayals, spon
-
sor motives, numerical guidelines, and underage
policies.
H2: Interpretations of campaign purpose will be highly
diverse, as evidenced by simultaneous holding of
multiple perceptions across divergent dimensions of
sponsor motives and underage policies.
H3: Compared to interpretations, the multiple evaluations
of corporate image and attitudes toward sponsoring
companies will be more uniform and positive.
DIFFERENTIAL RESPONSE BY AGE LEVEL
Age level is a key individual difference variable that is ex
-
pected to mediate audience responses to “drink responsibly”
campaign messages because developmental progression dur
-
ing adolescence has maturational implications for both so
-
phistication of message processing and predisposition to
-
ward alcohol.
A teenage viewer who has not yet progressed to an ad-
vanced stage of development or attained several years of ex-
perience with the specific domain of alcohol products and
messages will not be sufficiently prepared to process these
campaign spots in a sophisticated manner. A more elaborate
array of media literacy skills for analyzing media messages
competently (J. A. Brown, 1998; Christ & Potter, 1998) is re-
quired to fully understand the motives and techniques of
“drink responsibly” advertisers (Boush, 2001; D. A. Brown,
2001; J. A. Brown, 1991; Martin, 1997; Paxson, 2003; Potter,
2001, 2004).
This study examines the age range from middle teenage
years through early twenties. During this period of develop
-
ment, young people are steadily advancing their information
processing skills and gradually adopting more responsible
alcohol predispositions (Atkin, 2004; Grube, 2004). Al
-
though these forms of progression are linear, comparisons
will be drawn between the younger segment of respondents
in high school (ages 15–18) versus the older segment in early
adulthood (ages 19–22).
Most college-age young adults will have gained the requi
-
site combination of maturity, experience, savvy, and skepti
-
cism to see through superficial message attributes and ap
-
peals to detect the underlying strategies of “drink
responsibly” campaigns and to form more distinct and uni
-
form interpretations of these messages (however, their so
-
phistication may also lead to a greater likelihood of simulta
-
neously perceiving multiple meanings and purposes in the
ambiguous messages, which is a secondary indicator of di
-
versity). In terms of alcohol predispositions, a substantial
majority of this older age group can be characterized as mod
-
erately permissive; most of them regularly consume alcohol
and are generally favorable toward drinking, whereas rela
-
tively small segments exhibit either restrictive antidrinking
STRATEGIC AMBIGUITY 3
or extreme prodrunkenness orientations (Felsted, 1986;
Johnston, O’Malley, & Bachman, 1995). This also will pro
-
duce less divergent interpretations of ambiguous messages.
On the other hand, teenagers still in high school are still
relatively naive regarding these types of campaigns. Spe
-
cifically, they may treat the messages at face value by per
-
ceiving the public service component as the dominant theme.
Further, these adolescents bring more disparate sets of pre
-
dispositions regarding the subject of alcohol. There is greater
polarization within this age group, with substantial propor
-
tions of “conservatives” (who are predrinkers or disapprov
-
ing of alcohol) versus “permissives” (who mainly drink to
get drunk) and a smaller segment falling in the moderately
prodrinking range of the continuum (Felsted, 1986; Johnston
et al., 1995). In interpreting“drink responsibly” messages,
teenagers are more likely to exhibit highly diverse responses
because of these tendencies.
Based on this discussion, the following hypothesis is of
-
fered:
H4: There will be a greater diversity in perceptual re
-
sponses among the teenage audience segment than
among young adults.
STRATEGIC MESSAGE CONTENT
EXECUTIONS UNDER STUDY
In the Anheuser-Busch campaign, the primary slogan is
“Know When to Say When.” The messages also advise peo-
ple to: “Drink responsibly,” “Always be in control,” “Be re-
sponsible,” “When you’re having a beer you’ve got to make
the right call,” “Don’t drink and drive … it’s crazy driving
any car when you’ve had too much to drink,” “If you’ve been
drinking, pass your car keys to someone who hasn’t,” “If
you’ve been drinking, don’t take your show on the road,” and
“Please use a designated driver.” Visually, most messages
portray drinkers enjoying alcohol in a party setting (although
several are of the “talking head ” variety). The spots never de
-
pict the harmful consequences of excessive or unsafe drink
-
ing.
The content of these messages features an array of ambi
-
guities that enable the audience to draw differential interpre
-
tations: First, the spots do not clearly define when to stop
drinking, either in terms of quantity consumed or degree of
intoxication. Hence, light and moderate drinkers may inter
-
pret that vague stopping point conservatively, whereas heavy
drinkers may interpret it quite liberally, or even regard it as a
challenge to be reached or exceeded. Basically, the “Know
When” spots do not explain how a drinker can recognize that
moment of knowledge.
Second, the spots do not suggest the option of
nondrinking for certain situations or certain types of individ
-
uals (e.g., youth, pregnant women, alcoholics). In particular,
the message does not proscribe drinking by drivers; by rec
-
ommending “know when” in the context of driving, it as
-
sumes that the driver will be consuming some alcohol.
Third, the spots do not clearly apply the guideline to
nondriving situations. Some versions recommend “know
when” in the context of driving, implying that in other situa
-
tions one does not have to follow the recommendation.
The Coors “Not Now” campaign presents a set of spots
featuring a basic format: brief visual depictions of a series of
acceptable drinking settings (parties, campfires, and sports
events) for which song lyrics labeling Coors as the “right
beer now,” intercut with three obviously risky situations for
which the announcer progressively disclaims as “not now,”
“definitely not now,” and “absolutely, positively not now.”
Two forms of ambiguity can be seen in the Coors spots.
First, there is a manifest commercial element in each mes
-
sage, in direct juxtaposition with the distinct warnings about
unsafe drinking. Second, the ads do not clearly specify
whether “not now” means zero consumption, no additional
consumption, or a limited quantity of consumption. There
also is further confusion because some “not now” situations
include strong adjectives (e.g., “definitely,” “absolutely,” and
“positively”).
METHOD AND PROCEDURE
Stimuli and Independent Variables
This investigation measures the responses of teenagers versus
young adults to an array of brewer-sponsored television spots
dealingwithresponsibledrinkinganddrunkdriving. A totalof
nine ads were tested (six sponsored by Anheuser-Busch and
three by Coors). The messages were packaged into three sub-
setsofthreeadseachandwerepresentedtoasampleofrespon
-
dents (N = 326) drawn from classes at several Michigan col
-
legesand high schools (n= 174 young adults ages 19–22 and n
= 152 teenagers ages 16–18); approximately one third of each
age group was shown each of the three subsets of spots. Thus,
more than 100 respondents saw each specimen spot, and each
respondent saw three spots. In the overall sample, 44% were
male and 56% were female.
Dependent Measures
The distribution of responses to each of the following
items is the basis for deviation scores that will be described
later. The first two sets of interpretations described following
were measured immediately after exposure to each individ
-
ual ad:
Perceived appearance.
A pair of items assessed per
-
ceptions of the similarity of the “drink responsibly” spots to
standard ads and PSAs: These items were measured along a
four-step scale ranging from “very similar” to “not similar.”
Each item was analyzed separately to determine diversity of
responses.
4
SMITH, ATKIN, ROZNOWSKI
Perceptions of drinking portrayals.
To determine how
the respondents perceived the content of each beer company
spot, viewers responded to the following descriptive state
-
ments on an agree–disagree scale: “This spot suggests drink
-
ing beer is fun,” “This spot encourages people to drink beer,”
and “Some people in this spot look like they’ve had too much
to drink.” These items were measured along a three-step
scale ranging from “agree” to “neutral” to “disagree.” Each
item was analyzed separately to determine diversity of re
-
sponses.
Following exposure to a full set of spots, measures were
taken of three types of variables encompassing interpreta
-
tions of the broader campaign:
Perceived motives.
Viewer perceptions of the pur
-
poses of brewer TV ad campaigns were measured with a
checklist of 10 briefly defined potential motivations. Respon
-
dents were asked whether or not they thought each reason had
prompted the beer company to sponsor the messages, thus
their responses to these 10 items were either “yes” or “no.”
They were also asked to review the list and mark which one
was the most important motive. The list contained randomly
ordered items representing three basic motivational dimen-
sions: public relations (“To improve the company’s image,”
“To make it appear they’re concerned,” “So the government
won’t regulate alcohol,” and “To avoid getting blamed for al-
cohol problems”); public service (“To preventdrunk driving,”
“To persuade people to avoid getting drunk,” and “Because
they actually are concerned”); and commercial selling (“To
makedrinkinglook like fun,” “To make more money,” and “To
sell their beer”). These items were analyzed separately.
Numerical interpretation of drinking guidelines.
A
pair of items sought to ascertain the meaning that the audi
-
ence derived from each campaign slogan in terms of recom
-
mended quantity of consumption. Respondents were asked
to write down the exact number of drinks that they believed
the slogans suggested were appropriate to consume: “These
messages recommend that drinkers should ‘know when to
say when.’ What is your interpretation of the meaning of
‘when’—after how many cans of beer should a 150-pound
adult stop drinking?” or “These messages depict several ‘not
now’situations. For these situations, how many beers do you
think the Coors company wants people to drink?”
Perceived policies.
To assess perceptions of the brew
-
ers’ position on underage drinking practices, respondents
were asked to infer whether the companies wanted teens to
think that they “should wait until they’re 21 to drink” or
whether it is appropriate for teens to “drink beer in modera
-
tion” or for older teens to “get drunk on occasion.” These
items were answered either as “yes” or “no.”
Finally, the questionnaire contained two sets of outcomes
associated with strategic ambiguity dealing with evaluative
ratings of message substance and attitudinal ratings of the
sponsoring companies.
Brewer evaluations.
To assess whether viewers be
-
come more positive toward the sponsoring companies, attitu
-
dinal favorability and corporate reputation/image, as well as
ratings of each company’s campaign message qualities, were
measured. Corporate images were measured on a series of
five-step semantic differential rating scales anchored by pairs
of polar-opposite adjectives: “Caring–Uncaring,” ”Responsi
-
ble–Irresponsible,” “Respectable–Not Respectable,” “Trust
-
worthy–Untrustworthy,” “Sincere–Dishonest,” and “Good–
Bad.” Attitude change was assessed with a self-report item:
“As a result of these messages, do you feel more favorable or
less favorable toward the Anheuser-Busch (Coors) com
-
pany?” To assess the campaign messages, respondents were
asked to rate their “personal reaction” on a scale from 0 to 10
on these key qualities listed and briefly defined on the ques
-
tionnaire: “sensible (wise, reasonable advice),” “useful
(helpful and valuable information),” and “convincing (agree
with ideas).”
Degree of diversity.
The degree of diversity was oper-
ationalized in several ways. First, the flatness of the distribu-
tion of responses was computed with an average deviation
score (AvDev) summing the absolute difference between the
percentage of individuals falling in each response category
versus the percentages that would be expected if there was an
even spread across all categories and divided by the number
of categories. An array of fairly equal percentage figures (in-
dexed by an average deviation approaching 0%) indicates
that individuals are deriving differential interpretations. For
example, maximum diversity would be manifested by 25%
of respondents in each of the four categories (AvDev = 0%),
whereas minimal diversity would be manifested by 100% of
the respondents scoring in one category (AvDev = 44%).
A more conventional measure of dispersion across multi
-
ple response categories is kurtosis; zero kurtosis indicates
that the data are distributed normally around a single central
point, positive kurtosis indicates a higher degree of cluster
-
ing, whereas negative kurtosis indicates a lesser degree of
clustering. A significantly negative level of kurtosis can be
interpreted as representing a relatively flat distribution
spread across response categories. In the analyses, kurtosis
provides a compact and comparable summary statistic (k)
and a basis for significance testing to determine whether the
overall distribution differs from normality or whether the
older versus younger subgroups differ to a statistically signif
-
icant degree.
A different operationalization of diversity is applied to the
measures of perceived motives and corporate policies, where
certain answers are at least partially inconsistent with other
answers: perception of company policy regarding teenage
drinking as promoting abstinence versus moderation versus
drunkenness and perception of the sponsor motives as both
STRATEGIC AMBIGUITY 5
prevention-oriented and selling-oriented. If results show that
a sizable percentage of respondents derive multiple interpre
-
tations from the same sets of campaign spots, it provides an
-
other indication of diversity.
RESULTS
H1
H1 posited that interpretations of message content would be
highly diverse and divergent. First, evidence is presented re
-
gardingthe manifestationof evenlyspread distributionsof an
-
swers across response categories on perceptual categories
(H1). In addition, age-related differences predicted in H4 will
be reported for each dependent measure in H1 and summa
-
rizedinthesectiononH4. Theinitialanalysesof perceptualre
-
sponses focus on the two most ambiguous spots featuring ma
-
jor elements that promote drinking: Anheuser-Busch’s
“Friends” (showing couples enjoying a noisy bar), and
Coors’s “NotNow”(showingdrinkers ata club and a cookout,
intercut with “not now” driving and jet-skiing scenes).
Perceived appearance.
As seen in Table 1, there is
fairly high diversity in judgments regarding the similarity of
these messages to standard beer commercials. The average
deviation is 8% for “Not Now” (k = –.97, p < .05), and 12%
for “Friends” (k = –.62, ns). Moreover, teenagers are almost
evenly split in perceiving differing degrees of similarity of
these two spots to a typical commercial designed to sell beer;
the average deviation is just 3% (k = –1.35, p < .05). Young
adults exhibit less diversity, with a 17% average deviation (k
= +.57, ns).
Respondents tend to perceive that these spots are dissimi
-
lar to a typical PSA, but a moderate degree of diversity is still
apparent for both “Not Now” (AvDev = 10%, k = –.52, ns)
and “Friends” (AvDev = 12%, k = –.46, ns). The more uni
-
formly distributed perceptions tend to be concentrated
among young adults (AvDev = 18%, k = +.33, ns). Teenager
responses are more diverse (AvDev = 11%; k = –.81, ns), and
the kurtosis values for teenagers versus young adults differ
significantly (–.81 – .33 = –1.14, p < .05).
Perceptions of drinking portrayals.
Table 1 also shows
that about two thirds of all respondents agree that the spots
suggest that beer drinking is fun; there is a high average devia
-
tion of 23% for “Not Now” and 24% for “Friends,” with posi
-
tive kurtosis values for overall sample. This concentrated dis
-
tribution is primarily due to the answers of young adults
(AvDev = 34%, k = +5.16, p < .01), as there is considerable di-
versity among teenagers (AvDev = 12%, k = –1.32, p < .05).
On the question of whether the spot encourages drinking,
the average deviation scores for the two spots are 15% and
16%, with nonsignificant kurtosis values. The responses of
6
SMITH, ATKIN, ROZNOWSKI
TABLE 1
Viewer Responses to Two Specimen “Drink Responsibly” Spots
Message Age Group
Not Now %
a
Friends %
a
16–18 %
b
19–22 %
c
Perceived appearance
Similarity of spot to beer commercials
Very similar 33 49 30 53
Fairly similar 32 22 24 30
Slightly similar 21 18 26 13
Not similar 14 11 20 4
Similarity of spot to standard PSAs
Very similar 9 11 13 6
Fairly similar 26 14 27 13
Slightly similar 45 32 40 38
Not similar 20 43 20 43
Perceptions of drinking portrayals
Spots suggest drinking beer is fun
Agree 68 69 52 84
Neutral 21 13 23 11
Disagree 11 18 25 5
Spots encourage beer drinking
Agree 56 58 36 77
Neutral 25 26 37 15
Disagree 19 16 27 8
Spot characters had too much to drink
Agree 45 24 26 51
Neutral 14 25 23 10
Disagree 39 51 51 39
a
n = 109.
b
n = 102.
c
n = 116.
young adults cluster in the “agree” category (AvDev = 26%, k
= +2.09, p < .01), whereas teenage responses result in a fairly
even distribution in perception (AvDev = 4%, k = –1.36, p <
.05).
Perceptions of overdrinking by characters in the “Not
Now” spot are polarized, with 45% agreeing and 39% dis
-
agreeing (AvDev = 12%; k = –.57, ns); the “Friends” ad has
the greatest spread on this item (AvDev = 11%). Teenagers
have a lower average deviation and significant kurtosis value
(+12%; k = –1.14, p < .05) but do not differ significantly
compared to young adults (+16%; k = –.74, ns).
Perceived motives.
Table 2 presents results for the
perceived purpose of the campaigns (H2), when respondents
were asked whether or not the set of spots reflected each of
10 motivations. There is little diversity on the five most fre
-
quently cited motivations; more than 80% answer “Yes”
rather than “No” in each of these cases. It appears that several
reasons for sponsoring the campaign are quite obvious to
most viewers. Nevertheless, there are four other results that
are quite close to a 50–50 split with very low deviation
scores: “Actual concern” “Prevent drunkenness,” and
“Drinking as fun” for the Anheuser-Busch campaign, and
“Forestall regulation” for the Coors campaign.
Respondents also were asked to rank which of the 10
listed reasons is the primary purpose. Combining the three
subsets of items representing each basic motivational dimen-
sion, there is a fairly even three-way split in interpretation.
For the Coors campaign, 39% give a top-ranking to one of the
public relations motives, and 43% of those viewing the
Anheuser-Busch spots rank a public relations item as most
important. For the selling motives, the figures are 38% for
Coors and 31% for Anheuser-Busch. Finally, 23% of the
Coors viewers and 26% of the Anheuser-Busch viewers per
-
ceive that the primary purpose is public service. The average
deviation figures across these three basic motives are 7% for
the Coors campaign and 6% for Anheuser-Busch campaign
(k = –1.29, p < .05, and –1.40, p < .05, respectively).
The perceptions of the primary brewer motive are quite
distinct between the two age groups. Focusing on the motiva
-
tion that ranks No. 1 out of the 10 listed, 58% of young adults
versus 26% of teenagers give top ranking to public relations,
but for public service motives the figures are reversed, with
36% of teenagers versus 13% of young adults ranking this as
most important. A relatively small difference is found on the
selling dimension (29% of young adults vs. 38% of teenag
-
ers). Examining the overall pattern, there is much greater di
-
versity of perceived motives among teenagers, with a fairly
flat distribution of 26%, 36%, and 38%; the average devia
-
tion is 5% (k = –1.41, p < .05), compared to 17% for young
adults (k = +.60, ns).
Numerical interpretation of drinking guidelines.
A
key question assessed how respondents interpreted the exact
STRATEGIC AMBIGUITY 7
TABLE 2
Percentage of Yes Interpretations of Campaign Motives and Company Policies
Company Age Group
Coors %
a
Anheuser-Busch %
b
16–18 %
c
19–22 %
d
Selling motives
1. Sell beer 86 93 82 94 Yes
2. Make more money 84 72 73 80 Yes
3. Show drinking as fun 71 45 49 58 Yes
Public relations motives
1. Improve company’s image 97 93 91 97 Yes
2. Appearance of concern 90 91 85 98 Yes
3. Avoid blame for problems 84 86 80 91 Yes
4. Forestall regulation 47 64 61 56 Yes
Address alcohol problems motives
1. Prevent drunk driving 92 91 92 91 Yes
2. Actual concern 63 50 51 58 Yes
3. Prevent drunkenness 16 48 44 31 Yes
Perceived policies
1. Shouldn’t drink 37 33 45 24 Yes
2. Moderate drinking OK 54 60 49 66 Yes
3. OK to get drunk 44 39 39 42 Yes
Quantitative interpretation of drinking recommendation
Number of “Not Now”and “Know When” beers 14 10 17 8 Six or more
13 28 20 20 Four or five
16 29 18 27 Three
14 24 18 21 One or two
25 01 13 12 Zero
18 08 14 12 Don’t know
a
n = 109.
b
n = 217.
c
n = 102.
d
n = 116.
quantity of alcohol to be consumed under the “Not Now” and
“Know When” guidelines promoted by each company. The
degree of diversity is remarkable in each case, as shown in
Table 2. One-quarter of the sample thinks that Coors wants
people to consume no drinks in a “Not Now” situation,
whereas one-quarter thinks the company is recommending
four or more drinks; the average deviation is 6%, and a fur
-
ther indication of ambiguity is the finding that almost one
fifth have no idea what is meant by the recommendation. In
interpreting the meaning of “Know When,” estimates are
evenly spread across three levels of consumption quantity
ranging from “one or two” drinks up to “four or five” drinks;
overall, the average deviation is 10%. The kurtosis value for
Coors is –1.34 (p < .05), and the Anheuser-Busch value is
–.89 (p < .05). The distribution of answers is particularly dis
-
persed for teenagers (AvDev = 3%; k = –1.13, p < .05), with
fairly equal proportions of responses in all six response cate
-
gories. For young adults, degree of dispersion does not reach
significance (AvDev = 6%; k = –.74, ns).
Perceived policies.
Table 2 also presents results for
items dealing with perceived underage drinking policies of
each brewer. The inferences drawn by viewers are highly di-
vergent, with a distinct lack of unanimity in responses to the
three partially contradictory statements. The deviation score
averages 8% across the three policies (AvDev = 13%, 4%,
and 6%) for those seeing Coors campaign spots, and 13% in
the case of Anheuser-Busch (17%, 10%, and 11%). Thus, the
pattern of responses across the policy statements indicates
that the subgroups derive diametrically opposed views of
company policy after viewing the ads. This is particularly
pronounced for the teenager deviation scores, which average
6%, whereas the young adult scores average 16%.
Therefore, with regard to the H1 prediction that interpre
-
tations of message content in terms of ad appearance, drink
-
ing portrayals, sponsor motives, numerical guidelines, and
underage policies are highly diverse, as manifested by an
evenly spread distribution of answers, ample support is pro
-
vided.
H2
H2 predicted that respondents would hold multiple percep
-
tions of sponsor motives and underage policies.
Multiple interpretations of motives and policies.
H2 predicted diversity in the form of multiple interpretations
among substantial proportions of the respondents, especially
instances in which the response options are divergent or even
contradictory. This can be examined for each campaign on
the question of sponsor motivation attribution. Table 2 shows
that 90% believe that the companies are creating these spots
to “sell their beer,” while at the same time more than 92% be
-
lieve that the spots are intended to “prevent drunk driving.”
A more clear-cut case of partially conflicting percep
-
tions occurs among the 48% who believe that the
Anheuser-Busch campaign seeks to prevent drunkenness;
almost all of these individuals simultaneously believe that
the company is also trying to sell beer. Likewise, most of
the Coors viewers who think that the company seeks to pre
-
vent drunk driving (92%) also think that the spots are de
-
signed to make drinking look like fun (71%). An interesting
contrast by sponsor occurs for prevention motivation: re
-
spondents interpret the two sets of ads as indicating that
each company is equally motivated to “prevent drunk driv
-
ing” (92% Coors vs. 91% Anheuser-Busch), but there is a
significant difference of 16% Coors versus 48%
Anheuser-Busch in perceived motivation to “prevent drunk
-
enness,” χ
2
= 29.8, p < .001). Moreover, there are divergent
perceptions of the two companies on perceived motive to
“show drinking as fun,” with a 71% Coors versus 45%
Anheuser-Busch difference, χ
2
= 29.8, p < .001).
Examining the pattern of responses to the set of state
-
ments about brewer policies regarding teenage drinking,
many viewers also derive somewhat contradictory percep
-
tions. An average of 40% simultaneously infer that each
company is promoting both moderation and drunkenness to
teenage audiences, and almost 20% draw the implication that
both moderation and abstinence are being promoted in the
spots. Hence, support is provided for H2 in that diversity is
present in the form of multiple interpretations among sub-
stantial proportions of the respondents, especially in in-
stances in which the response options are divergent or even
contradictory.
H3
H3 posited that the evaluative ratings will be more uniform
and more positive than the perceptual responses. First, the
campaign message ratings are described: averaging across
the Coors and Anheuser-Busch ads, the mean score on the 0
to 10 scale is 7.2 for sensible, 6.9 for convincing, and 6.7 for
useful. Most of the scores were above the midpoint of the
scale: 85% for sensible, 81% for convincing, and 75% for
useful. The vast majority of the respondents displayed posi
-
tive rather than negative evaluations the two sets of ads.
Table 3 shows that the three aspects of company image are
quite positively evaluated, with moderately low levels of di
-
versity. For Coors, a majority gave a positive evaluation on
each image dimension, with large positive to negative ratios
on responsible (62% to 18%), respectable (59% to 14%), and
caring (51% to 15%); the average margin was 57% to 16%.
The pattern is similar for Anheuser-Busch, with an average
margin of 54% positive to 15% negative. Compared to the
perception variables, the deviation scores are consistently
higher across the three images: 19%, 17%, and 12% for
Coors and 17%, 17%, and 11% for Anheuser-Busch.
On the attitudinal measure, most respondents who re
-
ported a change in favorability became more positive rather
8
SMITH, ATKIN, ROZNOWSKI
than negative. After seeing the Coors spots, 39% were more
favorable and 9% were less favorable, and respondents’ atti-
tudes toward Anheuser-Busch improved by a 25% to 10%
favorability margin. As predicted, the average deviation
scores were relatively high, with 17% for Coors and 21% for
Anheuser-Busch.
In sum, the evaluations of sponsors were generally posi-
tive and fairly uniform on each outcome measure, providing
support for H3. The relatively narrower spread of responses
is reflected by the higher deviation scores; across the eight
percentage distribution values reported above, the average
deviation was 17%. This compares to an average deviation of
11% across 20 perception measures.
H4
H4 predicted a general pattern of divergence of perceptual re
-
sponses between the teenage and young adult samples. In ev
-
ery comparison, the average deviation for the youthful view
-
ers is lower, indicating more diversity in interpretations:
similar appearance to commercials (3% teenagers vs. 13%
young adults) and to PSAs (11% vs. 18%); perceived por
-
trayals of fun (12% vs. 34%), encouragement to drink (16%
vs. 26%) and overdrinking by characters (12% vs. 16%); per
-
ceived predominant motive (5% vs. 17%); perceived com
-
pany policies (6% vs. 16%); and interpretation of numerical
guidelines (3% and 6%).
Therefore, substantial evidence supports the H4 predic
-
tion of a differential pattern of responses between the teenage
and young adult subsamples on the interpretation of message
content and campaign purpose; the difference is always in the
direction of greater diversity for the more unsophisticated
and polarized teenagers.
DISCUSSION
From the sponsor’s perspective, strategically ambiguous
“drink responsibly” campaigns are advantageous in three re
-
spects. First, “responsible use” messages provide an effective
complement to conventional advertising that glamorizes con
-
sumption; the “soft sell” style offers unique and influential
persuasive incentives for favoring the product or brand. Sec
-
ond, this type of message is more pragmatic than a hard-hit
-
ting prevention-oriented public service message; the “soft
service” appeal partially combats the misuse problem with
-
out undermining most forms of product consumption. Third,
the apparent good faith effort that is ambiguously symbol
-
ized in these messages serves a subtle public relations func
-
tion that may disarm critics, impress opinion leaders, and en
-
gender good will with the general public.
On the surface, the Anheuser-Busch and Coors messages
present appeals that discourage misuse or promote respon
-
sibility by consumers; these themes are occasionally inter
-
woven with positive portrayals of product consumption.
The focus of this investigation was the responses of teenag
-
ers and young adults to individual spots and the overall
campaigns.
The first and second hypotheses informed by the strategic
ambiguity approach predicted highly diverse interpretations
of the message content and campaign purposes. Consistent
support was obtained on four key diversity criteria. First,
there was a considerable spread across answer categories in
perceptions of two specimen spots and a fairly even split
among the three basic perceived purposes, along with a
highly disparate set of interpretations regarding company
drinking policies and recommendations. Second, large pro-
portions of respondents drew divergent multiple implications
of campaign purposes and drinking policies. Third, percep
-
tions were highly varied across a set of messages from the
same campaign. These data suggest that different segments
in the overall audience are deriving highly diverse interpreta
-
tions of the ambiguous messages due to varied drinking pre
-
dispositions and information processing skills.
In addition, the findings supported the third prediction—
that evaluative ratings would be generally positive and more
uniform than the interpretive responses. Evaluations of both
the message content and the company image are clearly posi
-
tive, and this was reflected in greater attitudinal favorability
toward the sponsors. There was a much lower degree of di
-
versity in these bottom-line responses, despite the disparate
meanings associated with the message content and campaign
purposes. This indicates that individuals arrived at the same
basic positive outcomes through differential interpretive
pathways.
The last hypothesis, which predicted that teenagers versus
young adults would display relatively greater diversity in in
-
terpretations, also was supported. In almost every compari
-
son, these younger viewers had lower deviation scores (indi
-
cating greater diversity) than did the older viewers. It appears
STRATEGIC AMBIGUITY 9
TABLE 3
Evaluations of Sponsoring Companies
Company Age Group
Coors %
Anheuser-
Busch % 16–18 % 19–22 %
Corporate image
Respectable 62 59 52 69
Neutral 20 28 29 19
Not respectable 18 13 19 12
Responsible 59 58 49 67
Neutral 27 28 35 22
Irresponsible 14 14 16 11
Caring 51 45 40 54
Neutral 34 37 42 31
Uncaring 15 18 18 15
Company favorability
More favorable 39 25 23 36
About the same 52 65 64 57
Less favorable 9 10 13 7
that their lack of sophistication and attitudinal polarization
led the teenagers to draw particularly varied implications
from the messages.
Thus, the evidence indicates that beer companies achieved
advantageous outcomes to a large extent with these “drink re
-
sponsibly” campaigns. The evaluative responses were pre
-
dominantly favorable, and the interpretations tended to be
mostly prodrinking; there is little indication of any attitudinal
boomerang or anticonsumption sentiment.
The question of whether these campaigns benefit the pub
-
lic is more doubtful (DeJong et al., 1992). There is no solid
evidence of message effectiveness in encouraging responsi
-
ble drinking behavior; it is likely that unambiguous public
service messages are more influential. However, there is
some evidence to suggest that the overall effectiveness of
PSAs also is limited (Andsager, Weintraub-Austin, & Pin
-
kleton, 2001; Hatch, 2002). Regardless, the ambiguity in the
“drink responsibly” advertisements enables the audience to
draw primarily reinforcing implications that will not substan
-
tially reform improper drinking patterns. Furthermore, the
appearance of addressing the problem may preempt more
persuasive campaign efforts from government agencies and
prevention organizations.
Many measures were custom-created for this study, and
some variables relied on single-item measurement. Future
researchers should borrow from the advertising and market-
ing literature traditionally well-accepted scale measures of
attitudinal outcomes, such as attitude toward the ad and/or at-
titude toward the brand measures, as well as possibly use the
newly developed measure of corporate credibility suggested
by Newell and Goldsmith (2001). To enhance reliability,
more elaborate scales employing multiple items to tap into
relevant constructs should be developed. Additional mea
-
sures of individuals’ information processing ability would
enhance the precision of age-based differences in response to
these messages. Of course, tracing the impact to actual drink
-
ing behavior would heighten the validity of research examin
-
ing these prevention messages.
Because this topic is highly relevant to a number of stake
-
holders, including public policy officials, public health advo
-
cates, and the alcohol marketers and their advertising part
-
ners, a larger and more representative sample should be used
in future studies. In particular, it would be advantageous to
include high school students from more diverse backgrounds
and to include noncollegiate young adults.
Subsequent studies might explore other modalities of
moderation ads, such as print or billboard executions, and ex
-
tend the scope of responsible drinking ads to include cam
-
paigns sponsored by wine or distilled spirits companies.
In conclusion, this study supports the use of the strategic
ambiguity perspective in understanding the interpretation of
alcohol moderation advertisements. The research demon
-
strates that although young viewers derived diverse interpre
-
tations of ad message content and ad purpose, they still ar
-
rived at positive evaluative outcomes. Due to the exploratory
nature of this study, the findings should be applied cautiously
to the policies regarding drinking prevention messages. Ad
-
ditional investigations are needed to provide definitive evi
-
dence addressing the controversial issues associated with in
-
dustry-sponsored moderation campaigns.
ACKNOWLEDGMENT
This research was supported by a grant to Charles K. Atkin
from the AAA Foundation for Traffic Safety, Washington,
DC.
REFERENCES
Abramson, J. (1991, May 21). Selling moderation: Alcohol industry is at
forefront of efforts to curb drunkenness. Wall Street Journal, p. A1.
Andsager, J. L., Weintraub-Austin, E., & Pinkleton, B. E. (2001). Question
-
ing the value of realism: Young adults’ processing of messages in alco
-
hol-related public service announcements and advertising. Journal of
Communication, 51(1), 121–142.
Atkin, C. (2002). Promising strategies for media health campaigns. In W.
Crano & M. Burgoon (Eds.), Mass media and drug prevention: Classic
and contemporary theories and research (pp. 35–64). Mahwah, NJ: Law-
rence Erlbaum Associates, Inc.
Atkin, C. (2004). Media intervention impact: Evidence and promising strat-
egies. In R. J. Bonnie & M. E. O’Connell (Eds.), Reducing underage
drinking: A collective responsibility (pp. 565–597). Washington, DC: Na-
tional Academies Press.
Beatty, S. G. (1997, August 14). Nipped in the bud: How the beer industry
uses TV ads to mollify critics, buff its image. Wall Street Journal, p. A1.
Beaver, W. (1997). What to do about alcohol advertising. Business Horizons,
40, 87–91.
Belch, G. E., & Belch, M. A. (2001). Advertising and promotion: An inte
-
grated marketing communications perspective (5th ed.). New York:
McGraw Hill.
Boush, D. M. (2001). Mediating advertising effects. In J. Bryant & J. A.
Bryant (Eds.), Television and the American family (2nd ed., pp. 397–412).
Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Brown, D. A. (2001). Media literacy and critical television viewing in edu
-
cation. In D. G. Singer & J. L. Singer (Eds.), Handbook of children and
the media (pp. 681–697). Thousand Oaks: Sage.
Brown, J. A. (1991). Television “critical viewing skills” education: Major
media literacy projects in the United States and selected countries.
Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Brown, J. A. (1998). Media literacy perspectives. Journal of Communica
-
tion, 48(1), 44–57.
Bruner, J. (1964). On going beyond the information given. In R. J. C. Harper,
C. Anderson, M. Christensen, & S. M. Hunka (Eds.), The cognitive pro
-
cesses (pp. 293–311). Englewood Cliffs, NJ: Prentice-Hall.
Capella, J., & Street, R. L. (1989). Message effects: Research on mental
models of messages. In J. Bradac (Ed.), Message effects in communica
-
tion science (pp. 24–51). Newbury Park, CA: Sage.
Christ, W. G., & Potter, W. J. (1998). Media literacy, media education, and
the academy. Journal of Communication, 48(1), 5–15.
Contractor, N. S., & Ehrlich, M. C. (1993). Strategic ambiguity in the birth
of a loosely coupled organization: The case of a $50-million experiment.
Management Communication Quarterly, 6, 251–281.
DeJong, W., & Atkin, C. (1995). A review of national television PSA cam
-
paigns for preventing alcohol-impaired driving, 1987–1992. Journal of
Public Health Policy, 16, 59–80.
10 SMITH, ATKIN, ROZNOWSKI
DeJong, W., Atkin, C., & Wallack, L. (1992). A critical analysis of “modera
-
tion” advertising sponsored by the beer industry: Are “responsible drink
-
ing” commercials done responsibly? The Milbank Quarterly, 70,
661–678.
Eisenberg, E. (1984). Ambiguity as strategy in organizational communica
-
tion. Communication Monographs, 51, 227–239.
Felsted, C. (1986). Youth and alcohol abuse. Phoenix, AZ: Oryx Press.
Fombrun, C. J. (1996). Reputation: Realizing value from the corporate im
-
age. Boston: Harvard Business School Press.
Gregory, J. R. (1991). Marketing corporate image: The company as your
number one product. Lincolnwood, IL: NTC Business Books.
Grube, J. W. (2004). Alcohol in the media: Drinking portrayals, alcohol ad
-
vertising, and alcohol consumption among youth. In R. J. Bonnie & M. E.
O’Connell (Eds.), Reducing underage drinking: A collective responsibil
-
ity (pp. 597–604). Washington, DC: National Academies Press.
Ha, Y. W., & Hoch, S. J. (1989). Ambiguity, processing strategy and adver
-
tising-evidence interactions. Journal of Consumer Research, 16,
354–360.
Hatch, D. (2002). Report: PSAs air in worst day part. Electronic Media, 2, 6.
Hitchon, J., Duckler, P., & Thorson, E. (1994). Effects of ambiguity and
complexity on consumer response to music video commercials. Journal
of Broadcast and Electronic Media, 38, 289–306.
Hoch, S. J., & Ha, Y. W. (1986). Consumer learning: Advertising and the
ambiguity of product experience. Journal of Consumer Research, 13,
221–232.
Johnston, L. D., O’Malley, P. M., & Bachman, J. G. (1995). National survey
results on drug use from the monitoring the future study, 1975–1994 (NIH
Publication No. 93-3598). Rockville, MD: National Institute on Drug
Abuse.
Kellerman, K., & Lim, T. S. (1989). Inference-generating knowledge struc-
tures in message processing. In J. J. Bradac (Ed.), Message effects in com-
munication science (pp. 102-128). Newbury Park, CA: Sage.
Lord, R. G. (1985). An information processing approach to social percep-
tions, leadership, and behavioral measurement in organizations. In L. L.
Cummings & B. M. Shaw (Eds.), Research in organizational behavior:
Vol. 7 (pp. 87–128). Greenwich, CT: JAI.
Markham, A. (1996). Designing discourse: A critical analysis of strategic
ambiguity and workplace control. Management Communication Quar
-
terly, 9, 389–422.
Martin, M. C. (1997). Children’s understanding of the intent of advertising:
A meta-analysis. Journal of Public Policy & Marketing, 16, 205–216.
Newell, S. J., & Goldsmith, R. E. (2001). The development of a scale to mea
-
sure perceived corporate credibility. Journal of Business Research, 52,
235–247.
Ortony, A. (1978). Remembering, understanding, and representation. Cog
-
nitive Science, 2, 53–69.
Paxson, P. (2003). Media literacy: Thinking critically about advertising.
Portland, ME: J. Weston Walch.
Potter, W. J. (2001). Media literacy (2nd ed.). Thousand Oaks, CA: Sage.
Potter, W. J. (2004). Theory of media literacy: A cognitive approach. Thou
-
sand Oaks, CA: Sage.
Priester, J. (2002). Sex, drugs, and attitudinal ambivalence: How feelings of
evaluative tension influence alcohol use and safe sex behaviors. In W.
Crano & M. Burgoon (Eds.), Mass media and drug prevention (pp.
145–162). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Smith, S. W. (1995). Perceptual processing of nonverbal relational mes
-
sages. In D. Hewes (Ed.), The cognitive bases of interpersonal communi
-
cation. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Wallack, L., Dorfman, L., Jernigan, D., & Themba, M. (1993). Media advo
-
cacy and public health. Newbury Park, CA: Sage.
Wyer, R. S., & Carlston, D. E. (1979). Social cognition, inference, and attri
-
bution. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Wyer, R. S., & Gordon, S. F. (1984). The cognitive representation of social
information. In R. S. Wyer & T. K. Srull (Eds.), Handbook of social cogni-
tion: Vol. 2 (pp. 74–149). Hillsdale, NJ: Lawrence Erlbaum Associates,
Inc.
STRATEGIC AMBIGUITY
11
