Back to Journal Cover Page


CURRENT RESEARCH IN SOCIAL PSYCHOLOGY


Volume 7, Number 12
Submitted: April 26, 2002
Resubmitted: May 2, 2002
Accepted: May 6, 2002
Publication date: May 10, 2002

AN EXPERIMENTAL EXAMINATION OF STRATEGIES FOR RESISTING PERSUASION

Kimberly A. Cameron
University of North Carolina at Greensboro

Julia Zuwerink Jacks
University of North Carolina at Greensboro

Maureen E. O’Brien
University of North Carolina at Greensboro

ABSTRACT

The present research took an experimental approach to examining five strategies that may be effective in conferring resistance to persuasion (i.e., counterarguing, attitude bolstering, source derogation, negative affect, and assertions of confidence). Participants listened to a persuasive message then copied statements consistent with one of the resistance strategies under the ruse of providing handwriting samples for a lie-detection experiment. Compared to those who copied neutral statements, those who copied attitude bolstering statements, assertions of confidence, and negative affect statements were more resistant to change. Surprisingly, copying counterarguments and source derogations did not confer resistance. The results from a speech-only control group that was allowed to respond (i.e., resist) naturally suggested that the counterarguments condition prevented individuals from effectively counterarguing on their own. The implications of these results for each strategy were discussed.

[205]
---------------
[206]

INTRODUCTION

After more than 50 years of activity, persuasion researchers have produced an enormous literature concerning the factors that contribute to successful attitude change and the mechanisms that underlie such change (for a review, see Petty and Wegener 1998). Yet a remaining issue for contemporary persuasion research concerns understanding how people go about resisting attitude change when they encounter a persuasive message (Eagly and Chaiken 1993, 1995; Jacks and Cameron, in press). This issue is a critical one for basic researchers because a comprehensive theory of attitude change must be capable of explaining when and how attitudes change and resist change. Currently, the dominant models of persuasion (i.e., the Elaboration Likelihood Model, Petty and Cacioppo 1986, and the Heuristic-Systematic Model, Chaiken 1987) do a very respectable job of explaining attitude formation and attitude change for certain kinds of attitudes (e.g., "parochial college issues," Zanna 1993, p. 143), but they are less well suited to explain why some attitudes stubbornly resist change (e.g., attitudes toward abortion) and how one might overcome resistance for such hard to change attitudes (Abelson 1988; Zanna 1993). Understanding when and how individuals resist attitude change is also an important issue for applied reasons. Clearly there are times when the goal is to increase resistance to persuasion (e.g., increasing resistance to pro-smoking influences) and times when the goal is to increase vulnerability to persuasion (e.g., increasing the persuasiveness of anti-smoking influences). For both goals, understanding the mechanisms that underlie resistance to persuasion is essential. The purpose of the current research is to advance our understanding of the processes by which individuals resist persuasion by examining these processes experimentally.

Resistance Strategies
Theoretically, people can respond to a counterattitudinal communication in a number of ways that might facilitate resistance. In the present research, we examined five of these strategies: counterarguing, attitude bolstering, source derogation, negative affect, and assertions of confidence. Counterarguing involves the direct rebuttal of message arguments (e.g., Buller 1986; Eagly and Chaiken 1995; Festinger and Maccoby 1964; Jacks and Cameron, in press; McGuire 1964; Wright 1975). Attitude bolstering, on the other hand, involves support arguing—that is, generating thoughts that are favorable toward one’s original attitude without directly refuting message arguments (e.g., Jacks and Cameron, in press; McGuire 1964; Sherman and Gorkin 1980). Source derogation is a resistance strategy that involves insulting the source, dismissing his or her expertise or trustworthiness, or otherwise rejecting his or her validity as a source of information (Buller 1986; Festinger and Maccoby 1964; Jacks and Cameron, in press; Wright 1975). Negative affect involves responding to the persuasion attempt by getting angry, irritated or otherwise upset (Abelson and Miller 1967; Jacks and Cameron, in press; Jacks and Devine 2000; Zuwerink and Devine 1996). Finally, the assertions of confidence strategy involves explicitly asserting that nothing or no one could ever change one’s opinion (Jacks and Cameron, in press).

[206]
---------------
[207]

Each of these strategies for resisting persuasion was examined recently by Jacks and Cameron (in press). In their first study, participants were asked to explain in an open-ended essay how they go about resisting attitude change for personally important attitudes or beliefs. Each of the previously mentioned strategies was reliably identified in participants’ essays. Of particular interest was the discovery of the assertions of confidence strategy, which up to that point had not been explicitly identified in the literature. In their second and third studies, Jacks and Cameron asked participants to rate their likelihood of using each of these strategies to resist changing their attitudes toward abortion (Study 2) or the death penalty (Study 3). Participants overwhelmingly preferred the strategies of counterarguing and attitude bolstering, with source derogation and negative affect being their least preferred strategies. (Assertions of confidence fell in between.) In a final study, Jacks and Cameron exposed participants to a persuasive message concerning the death penalty and examined the actual frequency and effectiveness of these various strategies for resisting persuasion1. The most frequent responses to the persuasive message (obtained in an open-ended thought-listing task) were counterarguing, attitude bolstering, and source derogation—despite the self-report findings indicating that source derogation was not a preferred strategy. Source derogations and negative affect were also measured with rating scales, and both these data and thought-listing data were used to predict actual resistance to change. Results revealed that counterarguing and ratings of negative affect were effective in conferring resistance to attitude change (cf. Zuwerink and Devine 1996).

One important means of identifying strategy use is the thought-listing task (Jacks and Cameron, in press), a very commonly used task in the attitude change literature (see Cacioppo, Harkins, and Petty 1981). For this task, respondents are asked to list whatever thoughts (or feelings) came to mind while reading the persuasive message, without concern for whether those thoughts or feelings are positive, negative, neutral, or irrelevant in regard to the message. These responses are then coded on a dimension of interest for the researcher (e.g., valence, affective tone, or the presence of specific resistance strategies). One problem with using the thought-listing task in the study of resistance strategies, however, is that by its very nature the task may be more likely to elicit counterarguing and attitude bolstering than other strategies. That is, when asked to write down what they were thinking, participants may assume that the researcher is interested in their thoughts about the message. Thus, they may be less likely to report reactions that are less message-relevant (e.g., "It doesn’t matter what this says, I’m not changing my mind.") Consistent with this observation, Jacks and Cameron (in press) found only six assertions of confidence out of 109 relevant cognitive responses, despite the fact that participants reported that they would be fairly likely to use this strategy as a means of resisting persuasion. Thus, measuring resistance strategies by coding thought-listing protocols may not always be the most desirable or informative way of studying these strategies.

[207]
---------------
[208]

Furthermore, treating resistance strategies exclusively as a measured variable limits the kinds of questions one might ask (as well as the answers one might get) concerning those strategies. In contrast, treating resistance strategies as an independent variable allows one to address new questions in informative new ways. For example, the independent variable approach allows one to ask how effective each strategy is in facilitating resistance. Consider the case of assertions of confidence. Using the measurement approach, assertions of confidence are rarely expressed in a thought-listing protocol. Does this result indicate that this strategy is never used and/or is ineffective in promoting resistance persuasion? Or does it mean that people just do not express such thoughts on a thought-listing task? If they were to express assertions of confidence, would such thoughts help them resist? As these questions illustrate, simply measuring the strategies that people typically use does not tell us everything we might want to know about the various strategies for resisting persuasion. If assertions of confidence turns out to be an effective means of resisting persuasion, for example, then people could be taught to use the strategy.

The Present Research
The purpose of the present research was to examine five resistance strategies (i.e., counterarguing, attitude bolstering, source derogation, assertions of confidence, and negative affect) experimentally. Our specific goals were (1) to develop a method for manipulating which strategy would be used by participants and (2) to examine the effectiveness of these strategies in conferring resistance to persuasion. As will be evident in the method section, we exposed individuals to a counterattitudinal message first and then made salient one of five types of responses to the message (e.g., attitude bolstering). To those familiar with McGuire’s (1964) classic research on the inoculation theory of resistance, this strategy may seem somewhat familiar. It is similar to McGuire’s procedures in the sense of giving participants thoughts that they might not have had on their own. It is different, however, in a number of critical ways. First, McGuire (e.g., McGuire 1961) gave individuals attitude supportive thoughts or refutational thoughts (i.e., counterarguments) prior to exposure to the persuasive message. This inoculation procedure was shown to both motivate and enable resistance to a subsequent (and different) counterattitudinal message. In the current research, individuals are given the responses (e.g., attitude bolstering or source derogations) after exposure to a single persuasive message. Second, McGuire only examined counterarguing and attitude bolstering, whereas we examine a variety of different strategies. Third, McGuire examined cultural truisms, which are defined as beliefs that are widely accepted and rarely challenged (e.g., it’s a good idea to brush one’s teeth after every meal), whereas we examine attitudes toward an important social policy issue (i.e., the death penalty) that has clearly been the subject of debate and controversy in this culture.

[208]
---------------
[209]

Because no prior research has experimentally examined these resistance strategies, we were reluctant to make predictions regarding which one(s) might be effective in conferring resistance to persuasion. Likewise, we made no predictions as to the relative effectiveness of the strategies. Our primary purpose was to develop a plausible method of manipulating strategy use and to examine which strategy or strategies would be effective in conferring resistance to persuasion in the context of our study. Future research employing this or a similar method is expected to further clarify why and under what conditions each strategy (or combination of strategies) is effective.

METHOD

Participants and Design
In a pre-testing session conducted approximately four weeks prior to the study, introductory psychology students indicated their overall attitude toward the death penalty (1 = unfavorable, 9 = favorable). Participants were randomly selected from among those who reported a favorable attitude (6 to 9) toward the death penalty. Sixty-six women and six men were recruited by telephone and participated in partial fulfillment of a research requirement in their introductory psychology class. Participants were randomly assigned to one of six conditions: counterarguing, attitude bolstering, source derogation, negative affect, assertions of confidence, or a control group.

Procedure
Participants were run in groups of four to ten. Upon arriving to the laboratory, they were given a consent form to read and sign if they agreed to participate. No one refused to participate at this point. Participants were told that they were taking part in a pilot study for a study to be conducted the following semester on the analysis of handwriting for the sake of lie detection. They were led to believe that they would be providing confidential handwriting samples by copying typed statements that the experimenter would provide for them. Participants were also told that the statements were possible responses to a taped speech regarding the death penalty and that they would hear this speech before copying the statements. Ostensibly, students in a study the following semester would evaluate these handwritten statements to determine whether the individual who wrote the statements actually agreed with the statements or not. Participants were led to believe that the issue of the death penalty was chosen because its controversial nature insured that some would agree with the copied statements and others would not, thereby providing the experimenters with handwriting samples of both truths and lies. Upon being informed that the speaker was a professor of sociology at another university, in order to bolster the perceived credibility of the communicator, the participants listened to the audiotaped speech.

[209]
---------------
[210]

Following the speech, participants were provided with a page of typed statements and were instructed to read the statements carefully and copy them word-for-word as though they were their own truthful responses to the speech. After this task, participants completed a survey assessing their own attitudes toward the death penalty and the extent to which they agreed with each of the statements they had copied. At the end of the session, participants were fully debriefed and dismissed with credit. No participants indicated any suspicion regarding the procedures, and all data were retained for analysis.

Stimulus Materials
Message. The persuasive message contained five primary arguments against the death penalty. It was 692 words in length, was audiotaped by a 55 year-old male speaker, and lasted approximately four minutes. The five primary arguments were as follows: (a) with the use of the death penalty, there is always the risk of putting an innocent person to death, (b) the death penalty is the same as premeditated murder, (c) no one has the power or moral authority to decide when another person deserves to die, (d) the murder rates are no different in states with or without the death penalty, and so the death penalty does not deter crime, and (e) criminals should be required to work to pay for their life in prison rather than being put to death. The full text of this message can be found in Appendix A.

Eleven pilot participants were asked to listen carefully to a speech given by a sociology professor from another university because they would be evaluating it on a number of dimensions. After listening to the speech, they reported their attitudes toward the death penalty, their impression of the message, and their impression of the speaker. All of these participants were in favor of the death penalty (M = 7.45, SD = 1.13). The message was evaluated using eight 9-point semantic differentials (worth considering—not worth considering, persuasive—unpersuasive, unclear—clear, weak—strong, convincing—unconvincing, poor—good, unfamiliar—familiar, and understandable—not at all understandable). A message-strength index (a = .91) was calculated by averaging these eight items (after reverse scoring when necessary) so that high scores indicated greater perceived strength. Mean message-strength was 7.20 (SD = 1.51), indicating that the message was perceived as fairly strong and well-written.

These pilot participants also indicated their overall impression of the speaker by indicating the extent to which six adjectives (credible, unreasonable, knowledgeable, unintelligent, sincere, and biased) accurately described the speaker (1 = does not apply at all, 7 = applies very much). A speaker-impression index (a = .90) was calculated by averaging the responses to the adjectives (after reverse scoring when necessary) such that high numbers indicate a more favorable impression. The mean of this index was 6.21 (SD =1.49), indicating that the speaker was perceived as reasonably credible.

[210]
---------------
[211]

Resistance Strategies. Six sets of response statements were created, representing six experimental conditions (i.e., counterarguing, attitude bolstering, source derogation, negative affect, assertions of confidence, and a neutral sentence control). Each set of response statements contained four sentences, two of which were approximately 30 words in length and two of which were approximately 50 words in length. These response statements appear in Appendix B. Two of the four sentences for each condition referred directly to statements made in the persuasive message, and two did not. For example, the counterarguing response set contained two sentences mentioning arguments in the message and then directly countered these arguments, and two sentences contained counterarguments that did not refer directly to the message. As much as possible, the content of the response sentences was controlled across conditions. The content of these response statements is very similar in nature to the naturally occurring responses coded by Jacks and Cameron (in press).

Dependent Measures
Final Attitude. Participants indicated their attitudes toward the death penalty using three 9-point semantic differentials (favorable—unfavorable, negative—positive, and good—bad). An overall attitude index (a = .95) was computed (after reverse scoring when necessary), such that higher numbers indicated more favorable attitudes.

Agreement with Responses. Participants also reported the extent to which they agreed with each of the four statements they copied using a 9-point scale (1 = completely disagree, 9 = completely agree). An agreement index (a = .85) was calculated by averaging their agreement with each of the four statements.

RESULTS AND DISCUSSION

Attitude change scores were calculated by subtracting posttest attitudes from pretest attitudes. Higher numbers reflect greater attitude change. Data were analyzed in an analysis of covariance (ANCOVA)2. The agreement index was entered as a covariate, and the predictor variable was experimental condition (control, counterarguing, attitude bolstering, negative affect, assertions of confidence, source derogation). As expected, agreement was a significant covariate, F(1, 72) = 20.02, p < .001. The condition effect was also significant, F(5, 72) = 2.80, p < .03. Planned comparisons revealed that three out of five resistance strategies increased resistance to attitude change relative to the control group. As can be seen in Table 1, after adjusting for agreement, participants who copied negative affective sentences, assertions of confidence, or attitude bolstering sentences showed less attitude change compared to those participants who copied neutral sentences, ps < .05. Copying counterarguments or source derogations did not help participants resist attitude change significantly more than the control group (ps > .05).

[211]
---------------
[212]

Table 1. Mean Attitude Change Scores By Condition, Adjusted For Agreement With Copied Statements

Condition

M

SE

Control

3.43 a

.56

Counterarguing

2.66 a

.56

Source Derogation

2.37 a

.56

Attitude Bolstering

1.61 b

.54

Assertions of Confidence

1.40 b

.54

Negative Affect

.89 b

.55

Note. Means with different subscripts differ significantly at p < .05.

How are we to understand these results? To our minds, the results for counterarguing were initially the most puzzling. In the present study, counterarguing did not effectively confer resistance to persuasion relative to the control group. This result is counterintuitive and, in fact, contradictory to much previous research (e.g., Jacks and Cameron, in press; McGuire 1961; Romero, Agnew, and Insko 1996; Zuwerink and Devine 1996). The classic cognitive response approach to persuasion (e.g., Chaiken 1987; Greenwald 1968; Petty and Cacioppo 1986; Petty, Ostrom, and Brock 1981) suggests that negative cognitive responses to a persuasive message should result in resistance to that message. However, a key to this approach is that the responses are idiosyncratic or self-generated. We believe this observation is key to understanding the present null effect. That is, we suggest that providing individuals with prefabricated counterarguments disrupted the natural counterarguing process that participants may have engaged in on their own. Thus, by inhibiting their own idiosyncratic counterarguments, we may have inadvertently decreased their ability to resist the message3.

[212]
---------------
[213]

To test this explanation, we exposed thirty additional participants (a "speech only" control group) to the same death penalty message used in the main experiment and measured both their attitudes and their cognitive responses in a thought-listing task. These responses were then coded for the presence of counterarguments and other resistance strategies4. All thoughts were coded independently by each of the authors. Reliability was high (Cohen’s Kappa = .76), and disagreements were resolved through discussion. This sample, then, provides us with information as to the natural frequency of occurrence (in a thought-listing task) of the five different resistance strategies under consideration in this research. Interestingly, the most common resistance mode reported in the thought-listing task was attitude bolstering, with 88% of participants offering at least one attitude bolstering thought in response to the counterattitudinal message (range = 0 to 5). Counterarguing was also relatively common, with 50% of participants offering at least one counterargument (range = 0 to 4). Only 9.4% of participants offered any source derogations (range = 0-1) and only 6.2% of participants offered any negative affective statements (range = 0 to 2). No participants generated assertions of confidence.

To assess the effectiveness of these naturally occurring resistance responses, attitude change scores were calculated by subtracting final attitude from initial attitude (measured in a prescreening session, as was done for participants in the main experiment). More positive change scores indicate greater persuasion. The mean attitude change score for this sample was 0.90 (SD = 1.51). Post hoc comparisons involving this sample with the conditions of our main experiment revealed that when left to their own thoughts, participants in this speech only control group resisted persuasion to the same degree as experimental participants who had been given negative affect, assertions of confidence, or attitude bolstering thoughts to copy (ps > .34). This speech only control group was more resistant to persuasion, however, compared to individuals in the source derogation, counterarguing, and neutral sentence control conditions (ps < .01).

One additional way to address the specific question of whether the experimental counterarguing condition disrupted naturally occurring (and perhaps more effective) counterarguments is to examine the effectiveness of counterarguing for this speech only control group in conferring resistance. To that end, attitude change scores for the speech only control group were entered as the dependent variable in a multiple regression with the frequency of attitude bolstering statements and the frequency of counterarguing statements entered as predictor variables. Assertions of confidence, source derogations, and negative affective statements were not entered into the equation because their frequencies of occurrence were zero or negligible5. The overall model was significant, R2 = .59, F(2,27) = 7.196, p < .01. Attitude bolstering (b = -.51) significantly predicted resistance to attitude change, t(1,28) = -3.178, p = .004, independent of counterarguing. Likewise, counterarguing (b = -.46) significantly predicted resistance to attitude change, t(1,28) = -2.828, p = .009, independent of attitude bolstering6. This result suggests that participants’ naturally occurring counterarguments (and attitude bolstering thoughts) were effective in conferring resistance to attitude change. Thus, it seems probable that the counterarguments provided in the main experiment disrupted participants naturally occurring (and more compelling) counterarguments, which resulted in less resistance to the fairly strong persuasive message.

[213]
---------------
[214]

Other Strategies
We now proceed to a discussion of the other resistance strategies, beginning with source derogation. Source derogation is a classic resistance strategy that has long been assumed to contribute to resistance to persuasion (e.g., Festinger and Maccoby 1964). Jacks and Cameron (in press) found that source derogation is perceived to be a relatively ineffective strategy that individuals prefer not to employ, and individuals who scored high on a dispositional measure of social desirability in that research were especially averse to the strategy. Furthermore, in their fourth study, source derogation (as measured by both a thought-listing task and a rating scale) was not especially effective in conferring resistance, even though it was a much more common response than self-reported likelihood of using the strategy suggested. Thus, it seems that individuals are somewhat conflicted in regard to derogating the source as a means of resisting persuasion.

In the present study, making source derogations salient was ineffective in conferring resistance. One potential explanation for this result is the same as that for counterarguments: Perhaps the provision of source derogations disrupted the naturally occurring resistance process while not providing a sufficiently compelling or novel reason to reject the communication. To suggest that the source derogations we provided were not compelling in this context is reasonable give that the source was identified at the outset as a sociology professor. Therefore, the source derogations we provided may not have helped these college students resist because these derogations (e.g., ignorant, stupid, closed-minded) were inconsistent with the speaker’s identity as a sociology professor. Given these conditions, we are reluctant to conclude that source derogations do not work as a resistance tool. It is very likely that source derogations work well at least sometimes. Additional research will be necessary to determine the conditions under which this strategy will be an effective and/or likely means of resisting persuasion.

Regarding attitude bolstering, this research demonstrated that making attitude-supportive arguments salient to participants increased resistance to persuasion relative to the neutral sentence control group. That is, giving participants potentially novel reasons for retaining their original attitude was an effective strategy for enhancing resistance. This effect is consistent with classic inoculation research (McGuire 1964), which shows that giving participants attitude-supportive arguments before exposure to a counterattitudinal message also confers resistance (e.g., McGuire 1961). It is also consistent with the results of our speech only control group, for whom it was found that attitude bolstering facilitated resistance, over and above the effects of counterarguing. Thus, it seems that providing participants with potentially novel supportive arguments may be a particularly effective means of conferring resistance to persuasion immediately following a persuasive message.

[214]
---------------
[215]

Finally, this research also showed that making negative affect and assertions of confidence salient were effective ways of conferring resistance. As previously alluded to, individuals in Jacks and Cameron’s (in press) research were very unlikely to generate cognitive responses that could be coded as either negative affect or assertions of confidence. Likewise, the current speech only control group generated few, if any, assertions of confidence or statements of negative affect. Yet previous research has shown that negative affect can be an important mediator of resistance to persuasion (Jacks and Devine 2000; Zuwerink and Devine 1996). The current results show that making negative affective reactions to the message salient to participants is effective in heightening resistance to the message (cf. Abelson and Miller 1967). This result further corroborates the idea that negative affective reactions contribute to resistance to persuasion. Thus, encouraging individuals to get angry or irritated by an unwanted persuasive appeal may significantly bolster their resistance to that appeal.

Regarding assertions of confidence, the current research adds to our limited knowledge of this strategy. Jacks and Cameron (in press) were the first to identify this strategy as a potential response to persuasion attempts, and their research indicated that it is perceived to be a moderately likely and effective strategy. However, as noted, it is not very readily produced in the thought-listing task. Thus, prior to this research, it was virtually unknown what the true effectiveness of this strategy might be. The present results suggest that making statements such as "I would never change my mind" salient is an effective means of bolstering resistance to persuasion. Perhaps one reason for the effectiveness of this strategy is that it disrupts thinking about the issue and prevents any cognitive responses that might be favorable toward the persuasive message. (This could be said of the other strategies as well.) In addition, this strategy likely increases the perceived certainty with which an attitude is held, leading to greater resistance (cf. Tormala and Petty 2002). Of course, much more research is needed to develop our understanding of assertions of confidence as a strategy for resisting persuasion.

Caveats and Conclusion
This research highlights the fact that individuals have a number of different strategies for resisting persuasion at their disposal. Because of our interest in examining the efficacy of each strategy independently, we attempted to make one strategy salient above others. As we have argued, this procedure disrupts naturally occurring responses to persuasive attempts. Although this fact could be seen as a limitation, we think that our procedure nonetheless has advantages over a pure measurement strategy. Simply measuring how people resist tells us what they naturally do. However, it cannot inform us as to what kinds of strategies could be effective and could thus be taught and encouraged. Our methodology can be used to further investigate the conditions under which a given strategy might be effective. For example, future research could manipulate the expertise of the source and/or the strength of message arguments to see when source derogations might be effective in conferring resistance. Although they were ineffective in the present context, perhaps with a weak message and/or an inexpert source they would do nicely in facilitating resistance.

[215]
---------------
[216]

Another caveat to the present research is the fact that individuals were given a fairly elaborate cover story designed to reduce the likelihood that they would perceive the study as being concerned with persuasion processes. Would the results have been different if we had instructed individuals to resist persuasion? Very probably. Our goal, however, was to examine the effectiveness of each strategy in isolation. If individuals had been explicitly motivated to resist persuasion, they would probably have had a number of different resistant responses to the message (e.g., attitude bolstering, counterarguing, and negative affect). While acknowledging this as a potential limitation of the present study, we also see this as an invitation to further research that examines the effectiveness of various combinations of strategies in conferring resistance to persuasion.

In conclusion, the present research contributes to a growing body of literature indicating that individuals have a number of different tools in their resistance toolbox. In the present context (i.e., for participants who were not explicitly motivated to resist and who heard a fairly strong message by an expert source), attitude bolstering, assertions of confidence, and negative affect each were effective means of resisting the persuasive message. Source derogations and prefabricated counterarguments were ineffective. In other contexts, however, it seems that individuals may be willing to employ whatever tools are necessary in order to protect their attitude. This is particularly true when the attitude is personally important in some way (Ahluwalia 2000; Jacks and Devine 2000; Zuwerink and Devine 1996). We look forward to additional research that explicates when, why, and for whom each resistance strategy, alone or in combination, is effective in conferring resistance to persuasion.

NOTES

1. A source expertise manipulation was also employed, but without any significant effects.

2. Data were also analyzed using pretest and posttest scores as a repeated measures factor in a mixed-model ANOVA. The effects of these analyses were essentially the same as the ANCOVA on change scores. Change-score analyses are reported for ease of presentation.

3. This argument may also explain the increased effectiveness of negative affect and assertions of confidence. These responses are not commonly generated in a thought-listing task (Jacks and Cameron, in press). Thus, providing these responses helped to bolster resistance because participants are less likely to generate such responses on their own.

[216]
---------------
[217]

4. Thoughts were coded as message favorable, message neutral, message irrelevant, or as one of five possible resistance modes (i.e., source derogations, assertions of confidence, negative affect, attitude bolstering, counterarguing). Each thought received only one code, such that a thought that seemed to be expressing negative affect and counterarguing was forced into the single most representative category. Source derogations were thoughts that expressed a negative evaluation of the person who gave the speech (e.g., "That man is an idiot"). Assertions of confidence were thoughts that expressed certainty or confidence in one’s favorable view toward the death penalty, without mentioning any direct support for one’s attitude (e.g., "I’m not going to change my mind about supporting the death penalty"). Negative affective thoughts were those that simply expressed a negative emotional reaction to the speech or the issue (e.g., "That speech made me so angry"). Attitude bolstering thoughts were those that provided a reason for one’s favorable attitude toward the death penalty without directly countering any arguments in the message (e.g., "Without the death penalty, murderers could escape and kill again"). Finally, counterarguments were thoughts that disagreed with specific points made in the message (e.g., "Innocent people are not at risk for being executed wrongly, unlike what he said").

5. Including source derogation and negative affect as predictor variables does not change the effects reported.

6. An analysis of the relative effectiveness of these modes in conferring resistance was not possible because the same participants who counterargued also attitude bolstered.

APPENDIX A

Death Penalty Speech
Thank you for having me here today to present my views concerning the use of the death penalty in this country. As we all know, this issue is controversial and there are many good reasons to be for or against the death penalty. In this talk I will briefly discuss just the key issues that have led me to the conclusion that the use of the death penalty is wrong.

First, consider that with the use of the death penalty there is always the risk of putting an innocent person to death. Unfortunately, as we are all aware, our justice system is not perfect. Guilty people sometimes go free, while innocent people get convicted. In fact, Newsweek recently published an article reporting that over the past decade, several wrongly convicted people on death row proved their innocence during the appeals process and were set free. What if these people had been put to death before they could prove their innocence and obtain real justice? Once a wrongly convicted person has been killed, the horrible injustice can never be undone.

[217]
---------------
[218]

Second, I believe that it is morally wrong to kill a person except in immediate self-defense of one’s own life. The laws of the land clearly establish that premeditated, planned murder is a heinous crime that deserves severe punishment. But I don’t believe that death is an appropriate punishment for murder—because what is the death penalty if not systematic, planned killing? There is obviously a double standard in saying that murder is wrong but the death penalty is not.

A third and related reason for being against the death penalty is that it is not my place or anyone else’s to decide who should live and who should die. Who on this earth is really qualified to determine the value of someone’s life—no matter what they’ve done. I don’t believe that anyone has the authority or the wisdom to judge who should be allowed to live and who should die. That judgment should be left up to God alone. No one has the power or moral authority to decide when another person deserves to die. On these grounds alone I am against the death penalty.

But, as a fourth point against the death penalty, I would also like to point out that the National Institute for Crime Prevention recently reported that murder rates are no different in states with or without the death penalty. These statistics prove that the death penalty does not deter crime. Obviously, we want to do whatever we can to prevent crimes from ever taking place. But the threat of the death penalty simply isn’t working as a deterrent. I don’t believe that the thought of having to die for a crime ever enters people’s minds—that’s why the threat of death doesn’t prevent them from committing horrible acts of violence, especially crimes of passion. Thus, the death penalty is not defensible as a deterrent to crime.

As a fifth and final point, I believe that instead of putting criminals to death, this country should require them to pay for their life in prison by working on federal and community projects. In fact, the New York Times recently reported that several maximum-security prisons have already instituted such work programs with great success. It is an essential component of our justice system that criminals pay back society for their crimes. What better way to have them pay than by putting them to work? That way we get a return on our tax dollars instead of paying for their incarceration while they sit around all day on death row.

In sum, the death penalty is morally wrong. It is ineffective as a deterrent to crime. It poses a serious risk to those who are wrongly accused and convicted. And there are reasonable alternative punishments for those who might otherwise receive the death penalty. For all of these reasons, I oppose the use of the death penalty in this country. I appreciate the invitation to come and speak to you today, and I thank you for your attention. Now I would be glad to answer questions…

[218]
---------------
[219]

APPENDIX B

Response Statements
Counterarguments
I’m in favor of the death penalty because it is much less expensive than keeping convicted murderers in prison for life, even if they are working under guard to pay for their life in prison.

He argued that instead of putting criminals to death, we should require them to work; but there are no jobs a violent criminal could do because of the risk of using work-related materials as weapons or tools for escape.

He mentioned the risk of putting an innocent person to death, but that risk is negligible because the death penalty is only given to criminals who have been convicted of murder by a unanimous jury that has seen the evidence and concluded that the person is guilty beyond a shadow of a doubt.

I recently read an article in the New York Times reporting that in the past 10 years no one on death row has proved his innocence during an appeal and been released from prison.

Attitude Bolstering
I’m in favor of the death penalty because I believe that people who intentionally murder others deserve to die themselves.

He said that the death penalty is wrong because there is always the risk of putting an innocent person to death; I say that it’s more important to consider that the use of the death penalty communicates that in the United States we will not let convicted murderers just get away with their violent crimes.

He said that instead of putting criminals to death, this country should require criminals to work to pay for their life in prison; but it’s much more important to realize that the death penalty establishes justice and closure for the victim’s family and friends.

I believe that if we do not use the death penalty with first degree murderers, there is a good chance that they could escape or be released later and commit murder again.

[219]
---------------
[220]

Source Derogation
I’m in favor of the death penalty and I know that the man who gave the speech was not worth listening to because he was totally biased—I mean he really came off as incredibly closed-minded.

When he argued that the death penalty is wrong because there is always the risk of putting an innocent person to death, I realized just how narrowminded and manipulative he really was.

The speaker was obviously stupid because no intelligent person would argue against the use of the death penalty by saying that instead of putting criminals to death, this country should require criminals to work to pay for their life in prison.

When I heard the beginning of the speech, I began to think that the speaker was actually rather ignorant, but I continued to listen to him anyway, and by the end of the speech I knew that the man was a complete idiot.

Negative Affect
I’m in favor of the death penalty, and I get really irritated when I listen to arguments against the death penalty that I just don’t agree with.

I was quite annoyed when he said that the death penalty is wrong because there is always the risk of putting an innocent person to death—that kind of argument just really annoys me a lot.

I became very agitated and angry when he suggested that instead of putting criminals to death this country should require criminals to work to pay for their life in prison.

I really get disgusted in response to the whole debate over the use of the death penalty because I guess I have a hard time being nice and calm when people are expressing views that are the exact opposite of my own strongly held views.

Assertions of Confidence
I’m in favor of the death penalty and there is no way I would ever change my mind about the merits of the death penalty for our society as a whole.

[220]
---------------
[221]

Even when the man argued that the death penalty is wrong because there is always the risk of putting an innocent person to death, I was not convinced to change my views on this issue because I know that my opinion is right.

I listened when he said that instead of putting criminals to death, this country should require criminals to work to pay for their life in prison, but I’m not about to change my mind about this issue because I am very confident in my opinion.

I just can’t see myself changing my views because I am certain that I have thought through all the arguments and I know that I am in favor of the use of the death penalty.

Control
The man who gave the speech sounded like he was older (probably in his 50’s) and from the southwest; I can imagine that it cost a lot of money for him to fly here in order to give the speech.

The man who gave the speech said that he believed the death penalty is wrong because there is always the risk of putting an innocent person to death.

The man who gave the speech argued that instead of putting criminals to death, this country should require them to work to pay for their life in prison.

Although this issue has been in the news a lot, I think it will probably be discussed in the news a lot more before any new laws are passed because people in this country typically debate a great deal about an issue before they come to any conclusion.

REFERENCES

Abelson, R. P. 1988. "Conviction." American Psychologist, 43:267-275.

Abelson, R. P. and J. C. Miller. 1967. "Negative persuasion via personal insult." Journal of Experimental Social Psychology, 3:321-333.

[221]
---------------
[222]

Ahluwalia, R. 2000. "Examination of psychological processes underlying resistance to persuasion." Journal of Consumer Research, 27:217-232.

Buller, D. B. 1986. "Distraction during persuasive communication: A meta-analytic review." Communication Monographs, 53:91-114.

Cacioppo, J. T., S. G. Harkins and R. E. Petty. 1981. "The nature of attitudes and cognitive responses and their relationships to behavior." In R. E. Petty, T. M. Ostrom and T. C. Brock (Eds.), Cognitive responses in persuasion (pp. 31-54). Hillsdale, NJ: Erlbaum.

Chaiken, S. 1987. "The heuristic model of persuasion." In M. P. Zanna, J. M. Olson and C. P. Herman (Eds.), Social influence: The Ontario symposium (Vol. 5, pp. 3-39). Hillsdale, NJ: Erlbaum.

Eagly, A. H. and S. Chaiken. 1993. The psychology of attitudes. New York: Harcourt Brace Jovanovich.

Eagly, A. H. and S. Chaiken. 1995. "Attitude strength, attitude structure, and resistance to change." In R. E. Petty and J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences (pp. 413-432). Hillsdale, NJ: Erlbaum.

Festinger, L. and N. Maccoby. 1964. "On resistance to persuasive communications." Journal of Abnormal and Social Psychology, 68:359-366.

Greenwald, A. G. 1968. "Cognitive learning, cognitive response to persuasion, and attitude change." In A. G. Greenwald, T. C. Brock, and T. M. Ostrom (Eds.), Psychological foundations of attitudes (pp. 147-170). New York: Academic Press.

Jacks, J. Z. and K. A. Cameron. (in press). "Strategies for resisting persuasion." Basic and Applied Social Psychology.

Jacks, J. Z. and P. G. Devine. 2000. "Attitude importance, forewarning of message content, and resistance to persuasion." Basic and Applied Social Psychology, 22:19-29.

McGuire, W. J. 1961. "The effectiveness of supportive and refutational defenses in immunizing and restoring beliefs against persuasion." Sociometry, 24:184-197.

[222]
---------------
[223]

McGuire, W. J. 1964. "Inducing resistance to persuasion: Some contemporary approaches." In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 1, pp. 191-229). New York: Academic Press.

Petty, R. E. and J. T. Cacioppo. 1986. "The elaboration likelihood model of persuasion." In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 19, pp. 123-205). New York: Academic Press.

Petty, R. E., T. M. Ostrom and T. C. Brock (Eds.). 1981. Cognitive responses in persuasion. Hillsdale, NJ: Erlbaum.

Petty, R. E. and D. T. Wegener. 1998. "Attitude change: Multiple roles for persuasion variables." In D. T. Gilbert, S. T. Fiske and G. Lindzey (Eds.), The handbook of social psychology (Vol. 1, 4th ed., pp. 323-390). New York, NY: McGraw-Hill.

Romero, A. A., C. R. Agnew and C. A. Insko. 1996. "The cognitive mediation revisited: An empirical response to methodological and theoretical criticism." Personality and Social Psychology Bulletin, 22: 651-665.

Sherman, S. J. and L. Gorkin. 1980. "Attitude bolstering when behavior is inconsistent with central attitudes." Journal of Experimental Social Psychology, 16:388-403.

Tormala, Z. L. and R. E. Petty. 2002. "Resisting persuasion and attitude certainty: A meta-cognitive analysis." Chapter to appear in E. Knowles and J. Linn (Eds.), Resistance and Persuasion. Mahwah, NJ: Erlbaum.

Wright, P. 1975. "Factors affecting cognitive resistance to advertising." Journal of Consumer Research, 2:1-9.

Zanna, M. P. 1993. "Message receptivity: A new look at the old problem of open- versus close-mindedness." In A. A. Mitchell (Ed.), Advertising exposure, memory, and choice. Advertising and consumer psychology (pp. 141-162). Hillsdale, NJ: Erlbaum.

Zuwerink, J. R. and P. G. Devine. 1996. "Attitude importance and resistance to persuasion: It’s not just the thought that counts." Journal of Personality and Social Psychology, 70:931-944.

[223]
---------------
[224]

AUTHORS’ NOTE

We would like to thank Mark Choate, Lisa Harbison, Jennifer Jones, Lawrence Nicholson, and Anders Selhorst for their dedicated research assistance.

AUTHORS’ BIOGRAPHIES

Kimberly A. Cameron received her Ph.D. in social psychology from The University of North Carolina at Greensboro. She is currently working for a market research firm in Philadelphia, PA. E-mail: Kim_Cameron@msn.com.

Julia Zuwerink Jacks received her Ph.D. in social psychology from the University of Wisconsin at Madison. She is currently an Assistant Professor at The University of North Carolina at Greensboro. Her main research interests are in the area of attitudes and resistance to persuasion. E-mail: jrjacks@uncg.edu.

Maureen E. O’Brien is a Ph.D. student in social psychology at The University of North Carolina at Greensboro. Her main research interests involve attitude functions and prejudice. E-mail: meobrien@uncg.edu.

[224]
---------------
[225]