. 39
( 115 .)


effect in persuasion. When
attitudes are measured
immediately, a message 5
from a low-credibility
communicator is not 0
persuasive. However, after
a delay, the low-credibility
communicator becomes “10
more persuasive.
Immediate Delayed
From data provided by Gruder and colleages
Time of Attitude Measurement
et al. (1979).
Chapter 6 Persuasion and Attitude Change 191

What can we say happens to a persuasive message after several weeks? When the
discounting cue occurs before the message, the effect of the message diminishes. When
the discounting cue occurs after the message, the power of the message is reinforced.
The lesson for persuaders, then, is that they should attack their adversary before he or
she makes a case or conveys a rebuttal.

Gender of the Communicator and Persuasion
Does it matter whether the communicator of a persuasive message is male or female?
Unfortunately, there is not a great deal of research on this. Early research produced
inconsistent results (Flanagin & Metzger, 2003). Sometimes, males were more persua-
sive, and sometimes, females were more persuasive. In fact, the relationship between
gender of the communicator and persuasion is not simple, as we shall see next.
In one experiment male and female participants evaluated information on a per-
sonal Web site attributed to either a male or female author (Flanagin & Metzger, 2003).
Participants visited a Web site that was specially designed for the experiment. On the
Web site participants read a passage on the harmful effects to pregnant women of radia-
tion exposure during pregnancy. Participants rated the credibility of the source of the
message. The results showed that male participants rated the female author as more
credible than the male author. Conversely, female participants rated the male source as
more credible than the female source.
In another study (Schuller, Terry, & McKimmie, 2005) male and female partici-
pants evaluated expert testimony (simple or complex) that was presented by either a
male or female expert witness. The results, shown in Figure 6.2, showed that the male
expert witness was more persuasive (resulting in higher dollar awards) than the female
expert witness when the evidence was complex. However, the female expert witness
was more persuasive when the evidence was less complex. The male expert has an
advantage when the content of the message requires more cognitive effort to process,

Gender of Expert
Male Female



Mean Award



Figure 6.2 The
relationship between
the gender of an expert
witness and the complexity
of trial testimony.
Low High
Based on data from Schuller, Terry, and
Testimony Complexity McKimmie (2005).
Social Psychology

and the female expert has an advantage when the message does not require such effort.
Gender, then, is used differently depending on the nature of the cognitive processing
required (Schuler et al., 2005).
There is also some evidence for a gender-domain effect, meaning that a male
communicator may be more persuasive for male-oriented issues, and a female com-
municator may be more persuasive for female-oriented issues (McKimmie, Newton,
Terry, & Schuller, 2004; Schuller, Terry, & McKimmie, 2001). McKimmie et al. (2004)
found that a male expert was more persuasive than a female expert when the case was
male-oriented (a case involving an automotive service company). When the case was
female-oriented (a case involving a cosmetics company), the female expert was more
persuasive. They also found that jurors evaluated the expert witness more favorably
when he or she testi¬ed about a gender-congruent case.

The Message and the Audience
Thus far, we have seen that the characteristics of the communicator can in¬‚uence the
degree to which we modify our attitudes in response to a persuasive message. But what
about the message itself? What characteristics of messages make them more or less
persuasive, and how do these elements interact with the characteristics of the audience?
We address these questions next.

What Kind of Message Is Most Effective? The Power of Fear
An important quality of the message is whether it is based on rational or emotional
appeals. Early research showed that appeal to one emotion in particular”fear”can
make a message more effective than can appeal to reason or logic. Psychologists found
at ¬rst that an appeal containing a mild threat and evoking a low level of fear was more
effective than an appeal eliciting very high levels of fear (Hovland et al., 1953). Then
research suggested that moderate levels of fear may be most effective (Leventhal, 1970).
That is, you need enough fear to grab peopleʼs attention but not so much you send them
running for their lives. If the message is boring, people do not pay attention. If it is too
ferocious, they are repelled.
However, persuaders need to do more than make the audience fearful; they also
need to provide a possible solution. If the message is that smoking cigarettes results
in major health risks, and if the communicator does not offer a method for smokers to
quit, then little attitude or behavior change will occur. The smoker will be motivated
to change behavior if effective ways of dealing with the threat are offered. This prin-
ciple is in keeping with the Yale groupʼs notion that people will accept arguments that
bene¬t them.
Of course, individuals often avoid messages that make them uncomfortable. This
simple fact must be taken into account when determining a persuasion strategy. For
example, a strong fear appeal on television is not very effective. The message is there
only by our consent; we can always change the channel. This is why the American
Cancer Societyʼs most effective antismoking commercial involved a cartoon character
named “Johnny Smoke,” a long, tall cowboy cigarette. He was repeatedly asked, as he
blew smoke away from his gun: “Johnny Smoke, how many men did you shoot today?”
That was it: no direct threat, no explicit conclusion about the harm of smoking. It was
low-key, and the audience was allowed to draw their own conclusions.
Despite evidence that high-fear messages tend to repulse people, fear appeals are
widely used in health education, politics, and advertising. The assumption is that making
people afraid persuades them to stop smoking or to vote for a certain candidate or to buy
a particular product (Gleicher & Petty, 1992). Does fear work? Sometimes it does.
Chapter 6 Persuasion and Attitude Change 193

In one study of the effect of low versus high fear, Gleicher and Petty (1992) had
students at Ohio State University listen to one of four different simulated radio news
stories about crime on campus. The broadcasts were either moderate in fear (crime
was presented as a serious problem) or only mildly fearful (crime was not presented
as a serious problem). Besides manipulating fear, the researchers varied whether the
appeals had a clear assurance that something could be done about crime (a crime-watch
program) or that little could be done (i.e., the crime-watch programs do not work). The
researchers also varied the strength of the arguments; some participants heard strong
arguments, and others heard weak ones. In other words, some participants heard pow-
erful arguments in favor of the crime-watch program whereas others heard powerful
arguments that showed that crime-watch programs did not work. In the weak argument
condition, some participants heard not very good arguments in favor of crime-watch
programs whereas others heard equally weak arguments against the effectiveness of
crime-watch programs. In all these variations of the persuasive message, the speaker
was the same person with the same highly credible background.
The researchers found that under low fear conditions, strong persuasive arguments
produced more attitude change than weak arguments, regardless of whether the pro-
grams were expected to be effective. In other words, if crime did not appear to be a crisis
situation, students were not overly upset about the message or the possible outcome
(effectiveness of the crime-watch program) and were simply persuaded by the strength
of the arguments.
However, people who heard moderately fearful broadcasts focused on solutions to
the crime problem. When there was a clear expectation that something could be done
about crime on campus, weak and strong arguments were equally persuasive. If stu-
dents were con¬dent of a favorable outcome, they worried no further and did not thor-
oughly analyze the messages. But when the effectiveness of crime-¬ghting programs
was in question, students did discriminate between strong and weak arguments. In other
words, when there was no clear assurance that something effective could be done, fear
motivated the participants to carefully examine the messages, so they tended to be per-
suaded by strong arguments. Again, concern for the outcome made them evaluate the
messages carefully.
What we know from the Petty and Gleicher (1992) study is that fear initially moti-
vates us to ¬nd some easily available, reassuring remedy. We will accept an answer
uncritically if it promises us that everything will be okay. But if no such promise is
there, then we have to start to think for ourselves. So, fear in combination with the
lack of a clear and effective solution (a program to ¬ght crime, in this case) leads us
to analyze possible solutions carefully. Note that Petty and Gleicher were not dealing
with really high fear. Ethical considerations prevent researchers from creating such a
situation in the laboratory. It may be that very high fear shuts off all critical thinking
for most of us.
What do we know, then, about the effectiveness of using fear to persuade? The
¬rst point is that if we do scare people, it is a good idea to give them some reassurance
that they can protect themselves from the threat we have presented. The protection“
motivation explanation of how fear appeals work argues that intimidation motivates
us to think about ways to protect ourselves (Rogers, 1983). We are willing to make the
effort to evaluate arguments carefully. But, in keeping with the cognitive miser strategy,
if we donʼt need to analyze the arguments, we wonʼt.
What is the bottom line on the effectiveness of fear appeals? Based on the available
research we can conclude that fear appeals are most effective when four conditions are
met (Pratkanis & Aronson, 1992):
Social Psychology

1. The appeal generates relatively high levels of fear.
2. The appeal offers a speci¬c recommendation about how to avoid any dire
consequences depicted in the appeal.
3. The target of the appeal is provided with an effective way of avoiding the dire
consequences depicted in the appeal.
4. The target of the appeal believes that he or she can perform the recommended
action to avoid the dire consequences.

The Importance of Timing: Primacy Versus Recency
The effectiveness of any persuasive attempt hinges on the use of an effective strategy,
including the timing of the messageʼs delivery. When is it best to deliver your message?
If you were given the option of presenting your message before or after your opponent
in a debate, which should you choose? Generally, persuasive situations like these are
governed by a law of primacy (Lawson, 1969). That is, the message presented ¬rst has
law of primacy The law of more impact than the message presented second. However, the law of primacy does not
persuasion stating that the ¬rst always hold true. It depends on the structure of the situation. A primacy effect occurs
persuasive argument received when the two messages follow one another closely, and there is a delay between the
is more persuasive than later
second message and the audience response or assessment. In this situation, the ¬rst
persuasive arguments.
message has the greater impact. But when there is a delay between the two messages,
and a response or assessment is made soon after the second message, we see a recency
effect”the second message has a greater impact (Figure 6.3).
The primacy and recency effects apply most clearly under certain conditions”when
both sides have equally strong arguments and when listeners are reasonably motivated to
understand them. If one side has a much stronger argument than the other side, listeners
Figure 6.3 Conditions
are likely to be persuaded by the strong argument, regardless of whether it is presented
that favor either a primacy
¬rst or last (Haugtvedt & Wegener, 1993). When listeners are very motivated, very
effect (top) of recency
interested in the issue, they are more likely to be in¬‚uenced by the ¬rst argument (the
effect (bottom). Primacy or
primacy effect) than by those they hear later on (Haugtvedt & Wegener, 1993).
recency depends on when
a delay is introduced.
Chapter 6 Persuasion and Attitude Change 195

Fitting the Message to the Audience
The Yale group also was interested in the construction and presentation of persuasive
messages. One of their ¬ndings was that messages have to be presented differently to
different audiences. For example, an educated or highly involved audience requires a dif-
ferent type of persuasive message than an uneducated or uninvolved audience. Rational
arguments are effective with educated or analytical audiences (Cacioppo, Petty, & Morris,
1983). Emotional appeals work better with less educated or less analytical groups.

One-Sided Versus Two-Sided Messages
The nature of the audience also in¬‚uences how a message is structured. For less edu-
cated, uninformed audiences, a one-sided message works best. In a one-sided message
you present only your side of the issue and draw conclusions for the audience. For a
well-educated, well-informed audience, a two-sided message works best. The more
educated audience probably is already aware of the other side of the argument. If you
attempt to persuade them with a one-sided argument, they may question your motives.
Also, well-educated audience members can draw their own conclusions. They probably
would resent your drawing conclusions for them. Thus, a more educated audience will
be more persuaded by a two-sided argument (Hovland, Janis, & Kelley, 1953).
One-sided and two-sided appeals also have different effects depending on the initial
attitudes of the audience. Generally, a one-sided message is effective when the audi-
ence already agrees with your position. If the audience is against your position, a two-
sided message works best. You need to consider both the initial position of audience
members and their education level when deciding on an approach. A two-sided appeal
is best when your audience is educated, regardless of their initial position. A one-sided
appeal works best on an uneducated audience that already agrees with you.

Inoculating the Audience
When presenting a two-sided message, you donʼt want to accidentally persuade the
audience of the other side. Therefore, the best approach is to present that side in a weak-
ened form to “inoculate” the audience against it (McGuire, 1985). When you present
a weakened message, listeners will devise their own counterarguments: “Well, thatʼs
obviously not true! Any fool can see through that argument! Who do they think theyʼre
inoculation theory The
kidding?” The listeners convince themselves that the argument is wrong. Inoculation
theory that if a communicator
theory is based on the medical model of inoculation. People are given a weakened
exposes an audience to
version of a bacterium or a virus so that they can develop the antibodies to ¬ght the
a weakened version of
disease on their own. Similarly, in attempting to persuade people of your side, you give
an opposing argument,
them a weakened version of the opposing argument and let them develop their own the audience will devise
defenses against it. counterarguments to that
weakened version and avoid


. 39
( 115 .)