Master Thesis
Resistance to political persuasion: What it is, where it comes from and how to counter it
Submission: 18 June 2020
Word count: 8’000
Graduate School of Communication Master Program Communication Science
Chiara Valli 12053643
valli_chiara@ymail.com Supervisor:
Abstract
While the concept of persuasion is anything but new to students of public opinion, there seems to be a lack of literature investigating why and through what processes voters resist persuasive messages in the political arena. It is crucial, however, to understand and manage resistance in order to achieve persuasion. By incorporating contemporary theories of related disciplines, this study explores what strategies voters use to resist a dissonant political message and what persuasion technique is able to counter these defense mechanisms. More specifically, it sought to investigate if elaboration – that is thoughtful processing – and its antecedents are able to predict the type of resistance and if two-sided messages are effective in persuading voters of a counter-attitudinal position. To scrutinize these dynamics, Study 1 used a quasi-experimental design in which American respondents (N=527) read a counter-attitudinal statement about the implementation of gender quotas. The results suggest that elaboration significantly reduces resistance overall, which, in turn, increases the persuasion success. The results of Study 2, which employed a fully randomized experimental design to test the effectiveness of two-sidedness, on the other hand, indicate that two-sided refutational messages are not more persuasive than one-sided communication.
Keywords: Political Communication; Persuasion; Resistance; Elaboration Likelihood Model; Gender Quotas
Introduction
Persuaders – be it politicians or advertisers – often face the challenge of convincing their audience of a position opposite of their preexisting attitudes. Since individuals have a natural tendency to prefer attitudinal information over such that contradicts their views (Festinger, 1957), they may try to dismiss the dissonant message (see also Kunda, 1990; Lodge & Taber, 2000; Taber & Lodge, 2006). This desire to disconfirm counter-attitudinal arguments may lead to a resistant behavior that actively works against the acceptance of the persuasive appeal. Hence, persuaders often need to fight these resistant forces first before they can attempt to achieve effective persuasion (Knowles & Linn, 2004).
Although this inference seems relatively intuitive, resistance to persuasion has received surprisingly little attention from researchers. The strong emphasis on the ways and means of effective persuasion has long led the field to neglect the interrelated phenomenon of resistance (Zuwerink & Devine, 1996). This is not to say that resistance has been completely absent from literature. Rather, it has overwhelmingly been studied implicitly, as lack of persuasion (e.g., Jacks & O’Brien, 2004; Wegener, Petty, Smoak, & Fabrigar, 2004). Less often, it has been regarded as an explicit and active process that precedes and potentially determines the persuasion outcome (Tormala & Petty, 2004). According to this latter conceptualization, resistance is “not simply the inverse of persuasion” (Jacks & O’Brien, 2004, p. 236) but should be regarded as a separate force worthy of its own study. Literature suggests, for example, that resistance can take on distinct forms and can be manifested in cognitive, affective, or behavioral responses that range from counterarguing to negative affect (Jacks & O’Brien, 2004). This shift of focus on the processes through which resistance occurs has yielded an emerging field that studies resistance as a distinct but integral part of the persuasion process.
The insights that persuaders gain from going beyond the simple persuasion outcome should not be underestimated: “Understanding resistance can set the stage for more effective
attempts at persuasion” (Wegener et al., 2004, p. 26) because it allows persuaders to tailor their persuasion tactic to the resistance strategy of the receiver and counter it. An example can illustrate this mechanism: Individuals cannot be persuaded by the quality of an antismoking campaign’s arguments if they physically avoid official advertisements by, for example, changing the TV channel when the ads appear. By integrating the health message into a TV-series that is popular amongst the target group, for example, one can camouflage the
message’s persuasive intent and, possibly, undermine the individuals’ efforts to avoid and resist it. From the persuader’s perspective, most other persuasion tactics would have been utterly ineffective as the targets would have avoided the exposure to the health message in the first place.
The present thesis
Knowing the dynamics of resistance can therefore pave the way for more effective persuasion. To get a better understanding of why individuals choose certain resistance strategies over others, scholars have analyzed the diverse nature of these processes. They found that the defense mechanisms vary in the amount of thoughts they require and argued that the extent to which people are willing or able to engage in thoughtful processing (i.e., elaboration)1 influences what coping strategy they use to resist (e.g., Zuwerink & Devine,
1996; Tormala & Petty, 2004; Wegener et al., 2004). For example, it is likely that someone who is knowledgeable about the persuasive topic – and hence able to invest considerable thoughts into processing the message – is less reliant on superficial resistance strategies and opposes the message by refuting its content with counterarguments. Individuals who lack the knowledge to contest the opposing position may, on the other hand, resort to strategies that do not require the generation of substantive arguments, such as defending their opinion by
claiming that many people share their convictions (i.e., social validation; Jacks & Cameron, 2003).
Most of the studies that have investigated the role of thoughtful processing in
resistance explored elaboration in relation to one specific strategy (e.g., counterarguing; Petty & Cacioppo, 1979). Although these studies provide in-depth insights into the functioning of one particular strategy, the defense mechanisms rarely operate alone: In a natural setting, outside the laboratory, individuals are expected to use multiple resistance strategies
simultaneously, albeit to different extents. The isolated focus on one strategy might simplify the resistance forces and over – or underestimate the importance of the strategy under study. Other scholars focused on the cognitive and emotional responses more generally (Zuwerink & Devine, 1996) or investigated elaborative resistance by examining its effect on attitude
certainty (Tormala & Petty, 2004). Although these studies consider multiple types of
resistances, their broad conceptualization does not sufficiently account for the distinctiveness of the strategies, which might distort the conclusions that are drawn. These shortcomings call for a comprehensive investigation which considers several resistance strategies
simultaneously and that pays special attention to the uniqueness of each coping mechanism. Using the Elaboration Likelihood Model (ELM; Petty & Cacioppo, 1986), this thesis examines and compares the effect of elaboration on eight resistance strategies concurrently and tests how these influence the persuasion outcome.
Once persuaders understand the mechanisms of resistance, they can choose a persuasion tactic that neutralizes and reduces the anticipated resistance. One persuasion technique that has received considerable attention are two-sided messages, which include the pros and cons of the sponsor’s position. In regard to the neutralizing effect, this persuasion technique is particularly interesting because it not only reduces source derogation (Eisend, 2006) but also diminishes counterarguing (Fransen, Verlegh, Krmani & Smit, 2015). As such, it offers considerable advantages over one-sided messages or other persuasion techniques that
only tackle one specific type of resistance (e.g., increasing the attractiveness of the source). By additionally testing if two-sidedness reduces resistance and hence increases persuasion, this paper wants to take this investigation one step further and also address the important counterpart of resistance: effective persuasion.
The different foci of this thesis manifest themselves in two separate studies: Study 1 examines the role of elaboration in resistance and explores how the different defense
strategies affect the persuasion outcome, while Study 2 shifts the focus to the effectiveness of communication and examines the persuasiveness of two-sidedness. By investigating if
thoughtful processing can predict the type of resistance and if two-sided messages are able to pierce through that resistance, this thesis aims to provide an overview of the complex
dynamics of the persuasion process.
In brief, Study 1 follows a quasi-experimental protocol in which participants are exposed to a tailored counter-attitudinal message. Study 2 employs an experimental design whereby the respondents randomly receive either a one- or two-sided message. The
hypotheses are tested on an American sample of 527 participants who were recruited through the online platform “Amazon’s Mechanical Turk” (MTurk). The study was fielded in April 2020.
This paper measures peoples’ reactions to persuasive arguments on the topic of gender quotas. Despite the growing representation of women in political institutions around the globe, women are still vastly underrepresented in the U.S. (Maillé, 2015): To this day, less than 30% of the seats in the U.S. Congress are held by female representatives. Although gender quotas proved to be an effective tool to increase female representatives, they are often subject to political debate (Bacchi, 2013). The controversy often centers around the questions of individual and social justice (Teigen, 2000). Opponents argue that quotas stand in conflict with democratic values such as fairness, individualism, and competency (Dahlerup, 2003). Supporters justify the affirmative action on the same grounds but claim that it counteracts
preexisting discrimination that arises from the “tendencies to judge women as less competent than their qualifications actually warrant” (Teigen, 2000, p. 65). The controversy that
surrounds the affirmative action makes gender quotas a fruitful policy to test the dynamics of resistance and persuasion.
This paper’s contribution is two-fold: First, by examining the effect of elaboration on no less than eight resistance techniques, it addresses an important gap in literature and is a first attempt to provide a more complete understanding of the resistance dynamics. Second, by testing established theories from related disciplines on a political issue, it uses an
interdisciplinary approach to contribute to the literature of political communication – a field that, apart from a few exceptions (e.g., Ahluwalia, 2000; Daignault, Soroka, & Giasson, 2013; P. Meirick, 2002; P. C. Meirick & Nisbett, 2011), has paid relatively little attention to the phenomenon of resistance.
Theoretical Framework Study 1: Resistance
To be able to tailor the persuasion attack to the resistance of the target, one must first know the multitude of resistance strategies that exist. In 2015, Fransen et al. synthesized the literature from various disciplines, created an overview of the different techniques, and categorized them into avoidance, contesting, and empowering strategies (Table 1): People using avoidance strategies resist dissonant information by devoting less or no attention to the unwanted content (i.e., selective exposure; Freedman & Sears 1965). Individuals who resist through contesting strategies refute certain aspects of the message such as its source (i.e., source derogation), its content (i.e., counterarguing), or the persuasive tactic with which it was transmitted (i.e., derogating the persuasive tactic) (Fransen et al., 2015). Individuals may also shift their focus on their own beliefs and engage in empowering strategies which include generating thoughts that support their preexisting opinions (i.e., attitude bolstering; Lydon, Zanna, & Ross, 1988), looking for others in their social environment who share their attitude
(i.e., social validation), or reminding themselves that their opinions cannot be changed (i.e., self-assertion) (Jacks and Cameron, 2003). An additional defense mechanism that was not listed by Fransen et al. (2015) is negative affect, which involves getting angry about the message (Zuwerink Jacks & Devine, 2000)2.
As discussed, these strategies vary in the amount of elaboration they involve (Wegener et al., 2004). Some resistances focus on undermining the validity of the message and thus require considerable scrutiny, and the creation of relevant thoughts and arguments (Jacks & Cameron, 2003). These defense mechanisms can be categorized as cognitively effortful resistance strategies and primarily include counterarguing and attitude bolstering (Jacks & Cameron, 2003). Non-effortful resistance strategies, so this paper argues, manifest themselves in affective response (e.g., negative affect) or in strategies that rely on heuristic cues that do not build on substantive arguments (e.g., social validation). Following this logic, the
remaining six strategies would fall under this latter categoryi.
Table 1.
Resistance strategies
Notes. Description of the strategies adapted from Jacks & Cameron (2003, p. 151). For derogation of
persuasive tactic, see Fransen et al. (2015).
2 This list does not claim to be conclusive and more resistance strategies have been addressed by Fransen et al.
(2015). Because testing additional defense mechanisms would have exceeded the scope of this thesis, physical and mechanical avoidance as a form of selective exposure were not considered.
Cognitive Effort Strategy Example
Effortful
Counterarguing Respond by arguing with the person who is challenging my opinion.
Attitude bolstering Respond by talking about the facts that support what I believe.
Non-effortful
Selective exposure Respond by tuning-out the arguments that contradict my opinion. Source derogation Respond by thinking negative things about the person who is
challenging my opinion.
Derogating persuasive tactic Respond by being suspicious to the sponsor’s manipulative intent. Social Validation Respond by thinking about the fact that lots of people share my
convictions.
Self-assertion Respond by thinking about how there is nothing the other person can say that will change my mind.
Elaboration-Resistance Hypothesis
How much thought targets of persuasion invest in processing the persuasive message should be of utmost importance to persuaders: Research on the role of elaboration has shown that persuasion through thoughtful processes is more persistent than through mechanisms that require less cognitive activity (Petty, Haugtvedt, & Smith, 1995). Similarly, resistance
through elaboration is more resilient than through non-elaborative processes (Petty et al., 1995).
The role of elaboration in the persuasion process has firstly been addressed by the dual-process theory of attitude change, according to which persuasion occurs through the operation of either the “central” or the “peripheral” route (Lau, 2019). While the central route requires effortful and systematic information processing, the latter involves relatively little cognitive activity and primarily relies on heuristic cues (Zuwerink & Devine, 1996). Whether targets of persuasion choose the peripheral or the central route to process the persuasive message, thus, depends on the amount of elaboration they engage in (O’Keefe, 2008).
The ELM suggests that people who have the necessary cognitive resources (e.g., knowledge) or motivation (e.g., personal relevance) use the central route to scrutinize the relevant information to arrive at a reasoned evaluation (Petty & Wegener, 1998). While objective considerations may guide this process, it is often driven by the peoples’ prior attitudes and proceeds in a biased manner (Petty & Cacioppo, 1986). When people either lack the cognitive resources or motivation to engage in elaborative processing, they are likely to use the peripheral route and rely on heuristics instead (Petty & Wegener, 1998). Although this terminology stems from the ELM, other models, including the Heuristic-Systematic Model (HSM; Chaiken, 1980), have identified comparable mechanisms and disclose similar empirical results (Eagly & Chaiken, 1993).
Applying this logic to the coping strategies, this paper expects that individuals who engage in more elaboration employ cognitively effortful resistance strategies to oppose the
dissonant information. People who engage in less thoughtful processing, on the other hand, should be more likely to resort to non-effortful resistance strategies. Hence, the first
hypotheses can be formulated:
H1: Elaboration a) increases the use of effortful resistance strategies and b) decreases the use of non-effortful resistance strategies.
As indicated, elaborative resistance is generally more resilient than non-elaborative resistance (Petty et al., 1995). According to Petty et al. (1995), engaging in issue-relevant thinking increases the consistency of existing, pro-attitudinal knowledge structures. Through repeated accessing, these knowledge structures become more accessible and, therefore, more useful to defend an attitude from an attack. As the individuals believe that they have invested considerable effort in arriving at their conclusion, they are expected to become more
confident and develop stronger motives to resist changes in that attitude (also see Abelson, Aronson, McGuire, Newcomb, Rosenberg, & Tannenbaum, 1968). As such, cognitive effort is able to influence and potentially diminish the persuasion success. Following this
argumentation, this paper expects that people who resist through elaborative processes and effortful resistance strategies are less susceptible to persuasion overall:
H2: Effortful resistance strategies have a greater negative effect on the persuasion outcome than non-effortful resistance strategies.
Antecedents of Elaboration
In order to understand under what circumstances individuals engage in thoughtful resistance, it is crucial to examine the determinants of elaboration. As discussed, whether people engage in elaboration depends on several factors that revolve around their elaboration motivation or elaboration ability (O’Keefe, 2008). These factors can be triggered by the persuasion setting but often depend on the target’s characteristics and personal circumstances (Petty & Wegener, 1998). As Study 1 focuses on the receivers of the persuasive message and not the content or context of the persuasive attack, the following section will focus on
personal attributes only and scrutinize three factors that have received considerable attention in research. These include issue importance, issue knowledge, and the need for cognition.
According to Petty & Wegener (1998), issue importance is one of the most important variables to influence a person’s motivation to think about the message. When people expect the issue to have significant consequences for their lives (Apsler & Sears, 1968, p. 162), they are likely to devote more issue-relevant thoughts to evaluate the object and use the central route to assess the true merits of the argument (Petty et al., 1983, p. 137). Especially when the message challenges preexisting opinions, it may be perceived as a threat or incongruity that needs to be resolved (Cacioppo & Petty, 1979). The increased necessity to form a reasoned opinion, therefore, encourages a person to engage in more thoughtful message processing (Petty et al. 1983; Thomsen, Borgida, & Lavine, 1995; Johnson & Eagly, 1989). Following this argumentation, this paper hypothesizes that:
H3a: Issue importance increases the elaborative processing of a counter-attitudinal message.
Another strong predictor of information processing is the amount of issue-relevant knowledge. Evidence from a wide range of studies has demonstrated that knowledgeable people scrutinize the persuasive message more thoroughly than individuals who lack these resources (see Wood, Rhodes, & Biek, 1995, for a review). When this knowledge is absent, individuals lack the ability to critically assess the persuasive arguments and resort to more superficial cues to analyze the validity of the message (Wood & Kallgren, 1988). The issue-relevant knowledge has to be accessible, however, in order to be effective: Only individuals who have access to the relevant information may utilize it to generate strong
counterarguments against the position (Wood, 1982). Hence, the hypothesis posits that: H3b: Issue knowledge increases the elaborative processing of a counter-attitudinal message.
Last, scholarly literature has scrutinized the role of need for cognition (NFC), which captures the peoples’ motivation to think. Individuals are said to have high NFC when they are motivated to reason through evaluation and enjoy engaging in cognitive endeavors (Cacioppo & Petty, 1982). See, Petty & Evans (2009) found, for example, that people who possess high NFC experience higher levels of arousal when confronted with a complex task than a simple task. When exposed to a persuasive message, recipients with high NFC tend to focus on the content rather than on extraneous factors (Cacioppo & Petty, 1984) while individuals low in NFC resort to superficial cues in the evaluation of the message (Petty, Brinol & Loersch, 2009). This trait is, therefore, likely to increase a person’s motivation to engage in an elaborative process. The hypothesis that follows argues that:
H3c: Need for cognition increases the elaborative processing of a counter-attitudinal message.
Study 2: Persuasion
As discussed in Fransen et al. (2015), specific persuasion tactics neutralize certain resistances more effectively than others. A vast amount of literature has examined potential factors that might diminish resistance and, thereby, increase persuasion. One message variable that has been scrutinized from literature is the two-sidedness of messages. Other than one-sided communication, two-one-sided messages include pros and cons of the sponsor’s stance. Although incorporating critical elements may seem counterintuitive at first, studies have highlighted their effectiveness and underlined their ability to counter resistance. Presenting the recipients with negative features tends to decrease their likelihood of generating
counterarguments themselves, especially when the negative aspect is refuted and undermined in the same message (Fransen et al., 2015). As early as 1964, McGuire noted that mild attacking arguments reduced counterarguments and, hence, increased persuasion. Two-sidedness has further proven to increase the credibility of the source because it leads the receivers to conclude that the sponsor is telling the truth (Eisend, 2006, p.188). According to
the meta-analysis of O’Keefe (1999), this is especially true when applied to non-advertising issues such as political affairs. While many scholars acknowledge the importance of
moderating variables, there seems scholarly consensus that two-sided, refutational
communication is more persuasive than one-sided messages (Eisend, 2006), and even more so in cases of counter-attitudinal appeals (Lau, 2019). Based on the existing literature, this paper tests the last hypothesis:
H4: Two-sided messages are more persuasive than one-sided messages. The conceptual model can be visualized as follows:
Figure 1.
Conceptual Model
Methods
The two studies followed a quasi-experimental and experimental design, respectively. Each study will be elaborated separately in the following sections. The designs are visualized in Figure 2.
Participants
The participants were recruited from “Amazon Mechanical Turk”. Although the online platform enables researchers to recruit a diverse subject pool due to its low cost and ease of use, it is not without flaws. Besides the self-selection concerns,MTurk users are considerably younger, more liberal and tend to pay more attention to tasks than the general public
(Berinsky, Huber, & Lenz, 2012). Although the demographical and political differences
Issue Importance Issue Knowledge Need for Cognition
Cognitively Effortful Resistance Strategies
Elaboration Opinion Change 1
H3 Two-sidedness Opinion Change 2
+
H1
+ − +
H2 H4
Figure 2. Study 1 & Study
Notes. Dashed lines indicate random assignment of the participants.
Q1: Initial Opinion
“After reading this short piece of information, to what extent would you be in favor or against
a voluntary party quota?”
Rather against
Rather in favor
Treatment 1
Pro quota, one-sided
Treatment 2
Pro quota, two-sided
Persuasive Message
Contra quota, one-sided
Treatment 1
Anti quota, one-sided
Treatment 2
Anti quota, two-sided
Q3: Opinion Change 2
“After reading the arguments of a (second)
senator…”
Q2: Opinion Change 1
“After reading these arguments…”
Persuasive Message
Pro quota, one-sided
Q3: Opinion Change 2
“After reading the arguments of a (second)
senator…”
Q2: Opinion Change 1
“After reading these arguments…”
between MTurk and population-based samples have sparked considerable debate, growing literature seems to agree that MTurk is suitable for experimental research. Studies have demonstrated that the convenience sample produces comparable effects to those found in population-based samples (e.g., Berinsky et al., 2012; Krupnikov & Levine, 2014), thus allowing researchers to make generalizable experimental inferences from MTurk data.
Participants who showed indications of straight-liningii (N = 6) or failed the attention
checkiii (N = 76) were excluded from the analysis. The final sample consists of 527
participants, of whom 59.2% are male. 75.1% of the respondents are White, 13.1% Black, and 9.7% Hispanic. The mean age of the sample is 37.84 and the respondents are generally well educated (bachelor’s degree or higher = 69%). 47.8% of the subjects identified as Democrats (Republicans = 31.9%; Independents = 17.8%), and 46.3% placed themselves on the left side of the political spectrum (M = 5.05/10, SD = 2.9). The participants showed a general interest in politics (M = 3.8/5, SD = .99) and were somewhat informed about the political
representation of men and women in the U.S. (M = 1.93/4, SD = 1.09). The participants were rewarded with 1 U.S. Dollar for completing the survey.
Study 1: Resistance
Procedure and Material
After providing informed consent, participants revealed some demographic
information about themselves, including their age, racial background, and education. The next block examined their political engagement and knowledge as well as their partisanship and political ideology.
In order to study people’s resistance, Study 1 employed a quasi-experimental design in which participants were exposed to a tailored counter-attitudinal message. More specifically, participants were informed that the U.S. Congress is currently debating a voluntary party quota that demands the parties to “put forth female candidates in at least 30 percent of all contested congressional races” (extracted from the questionnaire; see Appendix C)iv. Upon
reading a short piece of information that described the bill’s central elements, participants
were asked to what extent they would be in favor of or against a quota. The answers were
recorded on a 11-point scale (0 = completely against, 10 = completely in favor) and served as
baseline for the allocation of the conditions: Participants who indicated a dissent with the
gender quota (initial opinion = 0-4) were asked to read a statement of a co-sponsor of the bill
that contained arguments in support of the quota. Participants who expressed their agreement
with the quota (initial opinion = 6-10) were exposed to a statement of a senator that allegedly
opposed the bill. Respondents who were neither in favor nor against the bill (initial opinion =
5) skipped Study 1 and were directed to Study 2v. Both statements addressed the same issues
but their arguments were reversed (see Appendix E, Table E1).
After the exposure to the first statement, people answered multiple questions that
tapped into their cognitive and emotional reactions to the message. Before moving on to
Study 2, subjects indicated to what extent they were in favor of or against the party quota after
reading the Senator’s arguments.
Measures
Independent Measures.
In line with the expectations, elaboration is assumed to guide the choice of resistance strategies. Elaboration was assessed with two questions that asked the participants to what extent they felt personally involved and how deeply they thought about the statement (adapted from Tormala & Petty, 2004). Responses were provided on a 9-point scale anchored at not at all and very much (M = 5.98/9, SD = 1.91; Cronbach’s α = .64).
The resistance strategies were assessed via several questions. All 7-point Likert scales were reversed and averaged to form the final scale. The means and standard deviations refer to the final scale.
Selective exposure was measured by presenting the participants with six headlines (Appendix D, Table D1-D6). Three of the headlines were in favor of the gender quota (e.g.,
“Why American politics needs gender quota”) and three were against a quota (e.g., “Women don’t need state-mandated quota to succeed”). Hence, they either supported or contradicted the respondents’ stance. The participants were instructed to select a minimum of one and a maximum of three articles they would like to read. Selective exposure was calculated by examining the choice of the participants with 0 = no selective exposure (i.e., balanced choice between congruent and incongruent articles), 1 = some selective exposure (e.g., more
congruent than incongruent articles) and 2 = full selective exposure (e.g., exclusively congruent articles) (M = 1.48/2, SD = 0.7).
To measure counterarguing, this study used two items that asked the respondents to indicate if they actively counterargued the message and tried to find flaws in the arguments (adapted from Moyer-Gusé & Nabi, 2010). The scale ranged from 1 = strongly agree to 7 = strongly disagree (M = 4.62/7, SD = 1.51; Cronbach’s α = .82).
Negative affect was measured by asking the subjects if the statement enraged, irritated, and annoyed them or if it made them angry (van Reijmersdal et al., 2016; Cronbach’s α = .85). Responses were captured on a 7-point Likert scale (M = 3.81/7, SD = 1.67; Cronbach’s α = .90).
To assess if participants resisted by derogating the source, they were asked to rate the Senator’s trustworthiness, credibility, and level of expertise. Responses were captured on a 7-point scale that ranged from 1 = strongly agree to 7 = strongly disagree (M = 3.7/7, SD = 1.4; Cronbach’s α = .92).
The index for attitude bolstering was adapted from the Bolster-Counterargue Scale (Briñol, Rucker, Tormala, & Petty, 2003) and included two statements (e.g., “I made a mental list of the reasons in support of my perspective”; Cronbach’s α = .72). Respondents were asked to indicate to what extent they agreed with the statements on a 7-point Likert scale (M = 4.64/7, SD = 1.37). In the same battery of questions, it was also examined if the participants engaged in social validation (“While reading the Senator’s statement, I thought about people
that share the same opinion which made me more confident in my opinion”; M = 1.45/7, SD = 1.76) and self-assertion (e.g., “While reading the Senator’s statement, I thought that no
argument would change my opinion on the subject”; Cronbach’s α = .78; M = 4.08/7, SD = 1.54). Both measures were inspired by Jacks & Cameron (2003).
Last, this paper explored if the respondents resisted by derogating the persuasive tactic and denouncing the statement as manipulative. The items were adapted from Cotte, Coulter, & Moore (2005; Cronbach’s α = .89) and adjusted to the specific case (e.g., “The Senator tried to manipulate the audience in ways I do not like”; Cronbach’s α = .77). Responses were captured on a 7-point Likert scale (M = 3.78/7, SD = 1.26).
According to the hypotheses, elaboration increases with issue knowledge, issue importance and NFC. The three variables were measured as follows: First, issue knowledge was measured with four questions that asked about the political representation of genders in various political institutions in the U.S. (e.g., “Do you know approximately how many % of the 100 seats in the Senate are currently held by women?”). The final scale ranged from 0-4 (M = 1.94/4, SD = 1.09). Second, issue importance was measured with three items that asked the respondents how important equal political representation is to them (adapted from
Zuwerink Jacks & Devine, 2000; Cronbach’s α = .80). Answers were recorded on a 7-point Likert scale that ranged from 1= strongly agree to 7 = strongly disagree (M = 4.78/7, SD = 1.44; Cronbach’s α = .78). Third, this paper measured NFC with a six-item scale (Lins de Holanda Coelho, H. P. Hanel, & J. Wolf, 2018; Cronbach’s α = .87). The participants were asked to indicate to what extent they agree with the six statements (e.g., “I would prefer complex to simple problems”; Cronbach’s α =.84) ranging from 1 = strongly agree to 7 = strongly disagree (M = 4.46/7, SD = 1.4).
Opinion Change.
Participants were asked to what extent they were in favor of or against the bill before and after the exposure to the persuasive statement (0 = completely against, 10 = completely in
favor). The discrepancy between the measurements indicated the attitude change. Persuasion
was assumed to have occurred if participants moved in the direction of the presented
argumentsvi. The inspection of the variable reveals that persuasion was relatively effective:
37.4% of the respondents changed their attitude in the direction of the message (M = 0.9/10,
SD = 1.64). Out of the participants who initially disagreed with the bill, 32.12% changed their opinion. 39.5% of the respondents who initially agreed with the quota changed their attitude.
The main variables are visualized in Figure 3-6.
Figure 3.
Frequency Distribution Elaboration
Figure 4.
Figure 5.
Frequency Distribution Issue Importance, Issue Knowledge, Need for Cognition
Figure 6.
Frequency Distribution Persuasion Outcome Study 1
Covariates.
This study included an adapted version of the ‘neosexism’ scale by Tougas, Brown, Beaton, & Joly (1995), which assessed gender prejudice and resistance against affirmative action. The index included three items that asked the respondents’ opinion on the situation of women in the U.S. labor market (e.g., “Discrimination against women in the labor force is no longer a problem in the United States”; Cronbach’s α = .90). The answer scale ranged from 1 = strongly agree to 7 = strongly disagree (M = 4.49/7, SD = 1.76). As study 1 was quasi-experimental, it needed to control for a variety of variables. The covariates included the
participants’ age, gender, sexual orientation, ideological position, education, political knowledge, political interest and attitude strengthvii as well as the direction of the message
(pro or contra gender quota). The test for multicollinearity did not reveal any problematic correlations.
Study 2: Persuasion
Procedure and Material
Following the completion of Study 1, subjects were exposed to a second statement that
either argued in favor of or against the adoption of the bill. This time, however, half of the
participants read a one-sided and half a two-sided message (see Appendix E, Table E2-E3).
Same as in Study 1, all arguments addressed the same elements but argued in opposite
directions. After reading the second statement, all participants were again asked to what
extent they agreed with the party quota. Before completing the survey, the participants were
debriefed and informed that the stimuli material was fictitious and created for this thesis only.
Measures
Opinion Change.
Same as in Study 1, participants were asked to what extent they were in favor of or
against the bill before and after the exposure to the second statement (0 = completely against, 10 = completely in favor). Attitude change was operationalized as the discrepancy between
these measurements (Figure 7; M = 0.68/10, SD = 1.32). 34.41% of the participants who
received the one-sided message, and 33.73% of the respondents who read the two-sided
message were persuaded by the arguments.
Covariates.
Several tests indicate that the randomization was successful. Issue knowledge was the
only variable that was not equally distributed across the two conditions and hence, included as
Figure 7.
Frequency Distribution Persuasion Outcome Study 2
Results Study 1: Resistance
This paper used Hayes' (2012) SPSS macro PROCESS (Model 6) to test the proposed serial mediation model. The results are summarized in Figure 8 (also see Appendix A, Table A1-A2). The first hypotheses (H1a and H1b) predicted that elaboration increases the use of effortful resistance strategies and decreases non-effortful strategies. While there is no significant effect of elaboration on attitude bolstering (b = 0.02, t = 0.48, p = .634), the analysis reveals that elaboration leads to less counterarguing, thereby pointing in the opposite direction of the one hypothesized (b = -.11, t = -2.93, p = .004). Hence, H1a has to be
rejected. Turning to the non-effortful resistance strategies, the findings reveal the following: An increase in elaboration leads to a significant decrease in negative affect (b = 0.19, t = -4.08, p = .000), source derogation (b = -0.29, t = -9.07, p =.000), self-assertion (b = -0.14, t = -3.8, p = .000), and the derogation of the persuasive tactic (b = -0.18, t = -5.54, p = .000). No effects can be found for selective exposure (b = -0.01, t = -.62, p = .536) or social validation (b = 0.02, t = 3.85, p = .701). H1b, which claimed that elaboration decreases non-effortful resistance strategies, can partially be accepted.
Figure 8.
Summarized Results Serial Mediation Models
Notes. Unstandardized beta coefficients.
***p < 0.001, **p < 0.01, *p < 0.05 Issue Importance
Issue Knowledge
Need for Cognition
Elaboration Negative Affect Selective Exposure Attitude Bolstering Counterarguing Self-Assertion Source Derogation Persuasive Tactic Persuasion .288*** -.006 -.119 -.112** .017 -.012 -.189*** -.144*** -.290*** -.183*** Social Validation .017 .010 -.272*** -.256*** -.176** -.197*** -.020 -.070 -.325***
22 H2 hypothesized that resistance through counterarguing and attitude bolstering reveal a greater negative effect on persuasion than the six non-effortful resistance strategies. The analysis reveals that persuasion decreases with more counterarguing (b = -.33, t = -6.34, p = .000) but is not affected by attitude bolstering (b = -.07, t = -1.21, p = .225). Looking at the effect of non-effortful resistance strategies, one finds that negative affect (b = -0.2, t = -4.08, p = .000), source derogation (b = -0.26, t = -4.08, p = .000), self-assertion (b = -0.18, t = -3.31, p = .001) and the derogation of the persuasive tactic (b = -0.27, t = -4.48, p = .000)
significantly decrease persuasion. Again, no comparable effects can be found for selective exposure (b = -0.02, t = -0.19, p = .854) or social validation (b = 0.01, t = 2.13, p = .832). A look at the standardized coefficients reveals that although counterarguing displays the largest effect size with b = -.30, it is closely followed by the derogation of the persuasive tactic with b = -0.21, which this paper categorized as a non-effortful strategy. To examine whether there is a statistically significant difference between effortful vs. non-effortful strategies, this study additionally conducted an analysis of variance (ANOVA)viii. The analysis reveals that there is
no significant difference between those participants who similarly engaged in both types of resistance, or primarily used effortful or non-effortful types of resistances, respectively (F(2,486) = 2.66, p = .071). Since the test violated the assumption of the homogeneity of variance due to the unequal sizes of the groups, an additional chi-square difference test was conducted. Once again, the result is insignificant. These findings indicate that there is no significant difference between the two types of resistances and that H2 should be rejected.
Finally, H3a–H3c predicted a positive effect of issue importance, issue knowledge and NFC on elaboration. Although the model proves significant with an adjusted R2 = .25 and a
significance of p = .000, only issue importance shows a significant effect in the expected direction. The findings indicate that participants who felt personally affected by the issue of equal political representation engaged in more elaboration (b = 0.29, t = 4.22, p = .000). Other than hypothesized, the analysis fails to uncover significant effects of NFC (b = -0.01, t
= -0.10, p = .921) and issue knowledge (b = -.12, t = -1.5, p = .135). Following these findings, we can accept H3a but reject H3b and H3c.
A look at the direct and indirect effects reveals that the proposed mediation was insignificant in most cases (see Appendix A, Table A3-A5).
Robustness Checks
To test the arising pattern from the main analysis, this study ran an additional serial mediation model using a comprehensive measure of resistance, which did not distinguish between the specific strategiesix. The findings suggest that elaboration significantly decreases
respondents’ motivation to resist the dissonant message and that more resistance leads to less persuasion. These results are in line with the main effects described in the preceding section (see Appendix B, Table B1).
Study 2: Persuasion
This study further sought to investigate if the two-sidedness of a message increases its persuasiveness. To test H4, this study ran a linear regression with sidedness as independent and persuasion as dependent variable (Table 2). The adjusted R2 of Model 1 of .002 reveals
that the model has low predictive power, indicating that the sidedness of messages only explains 0.2 % of the variance in the persuasion outcome (F(2, 493) = 1.51, p = .221). The insignificant p-value of .454 indicates that two-sided messages are not more persuasive than one-sided messages (b = -0.09, t = -0.75, 95%CI [-0.32, 0.14]). In the light of these findings, H4 must be rejected.
To test for possible moderating effects, the linear regression was replicated with various different moderators. The results do not significantly differ when taking the persuasion outcome of Study 1 into account (Model 2). Neither do we find distinct effects when adding the direction of the message (Model 3)x or the strength of the initial opinion as
24 Table 2.
Persuasion Outcome 2, moderated by Persuasion Outcome 1, Message Direction, Opinion Strength and Initial Opinion Model 1 B (SE) Model 2 B (SE) Model 3 B (SE) Model 4 B (SE) Model 5 B (SE) Constant .973*** (.206) .928*** (.229) .508 (.556) .761 (.407) -.173 (.450) Two-sidedness -.089 (.119) -.058 (.137) .058 (.350) .128 (.250) .460 (.271) Issue Knowledge -.081 (.054) -.063 (.055) -.074 (.055) -.057 (.055)
Persuasion Outcome Study 1 .018 (.111) Persuasion Outcome Study
1*Two-sidedness -.043 (.076) Message Direction .269 (.298) Message Direction*Two-sidedness -.090 (.192) Opinion Strength*Two-sidedness -.082 (.055) Opinion Strength*Two-sidedness .074 (.122) Initial Opinion .176** (.063) Initial Opinion*Two-sidedness -.088* (.039) Adjusted R2 .002 –.001 .003 .002 .017*
Notes. Unstandardized beta coefficients. For message direction 0 = neutral, 1 = pro-attitudinal, 2 = counter-attitudinal.
Including the initial opinion into the regression, however, we find that persuasion increased the more participants were in favor of the gender quota (b = .18, t = 2.82, p = .005, 95%CI [0.05, 0.3]) (Model 5). The interaction term further indicates that one-sided messages were more effective for these respondents (b = 0.09, t = 2.26, p = .024, 95%CI [0.17, -0.01]).
As discussed, two-sided messages work particularly well to reduce the generation of counterarguments (Fransen et al., 2015) and diminish source derogation (Eisend, 2006). Because this paper did not analyze the respondents’ resistance to the second message, it is not possible to investigate the direct effect of sidedness on resistance. Analyzing the reverse order – hence, if two-sided messages were more effective for participants who employed these strategies – might nevertheless shed some light on the neutralizing effect of this persuasion tactic. To examine this mechanism, this study ran an additional moderation analysis which included the resistance strategies and their interaction terms (adjusted R2 = .04, F(18, 459) =
2.07, p = .006) (Table 3).
The negative interaction effects indicate that people who engaged in social validation (b = -0.20, t = -2.21, p = .028, 95%CI [-0.37, -0.02]) and derogation of the persuasive tactic (b = -0.30, t = -2.30, p = .022, 95%CI [-0.56, -0.04]) were more convinced by the one-sided message. Only people who engaged in attitude bolstering found two-sided messages more persuasive (b = 0.30, t = 2.47, p = .014, 95%CI [0.06, 0.54]). No effects can be found for counterarguing (b = 0.90, t =84, p = 0.402, 95%CI [1.22, 0.30]) or source derogation (b = -.06, t = .51, p = .611, 95%CI [-0.16, 0.27]). As such, these results do not support the
assumption that two-sidedness is particularly effective to reduce the latter two types of resistance.
Table 3.
Persuasion Outcome 2, moderated by Resistance Strategies
Persuasion Outcome 2 B (SE) Constant 2.047* (.973) Two-sidedness -.466 (.618) Issue Knowledge -.062 (.057) Counterarguing -.228 (.165) Attitude Bolstering -.451* (.199) Negative Affect -.316* (.154) Self-Assertion .245 (.164) Source Derogation -.206 (.172) Persuasive Tactic .562** (.211) Social Validation .323* (.143) Selective Exposure -.381 (.264) Counterarguing*Two-sidedness .090 (.108) Attitude Bolstering*Two-sidedness .301* (.122) Negative Affect*Two-sidedness .175 (.102) Self-Assertion*Two-sidedness -.145 (.100) Source Derogation*Two-sidedness .055 (.107) Persuasive Tactic*Two-sidedness -.301* (.131) Social Validation*Two-sidedness -.195 *(.088) Selective Exposure*Two-sidedness .189 (.168) Adjusted R2 .039**
Notes. Unstandardized beta coefficients.
***p < 0.001, **p < 0.01, *p < 0.05
Discussion and Conclusion
Resistance has long been reduced to being the simple antithesis of persuasion. Only in the past few decades, researchers have recognized the potential the explicit study of resistance unlocks: Understanding the processes through which individuals resist enables persuaders to neutralize the resistance forces that work against the acceptance of their message. By
analyzing the dynamics of resistance and persuasion simultaneously, this paper aimed to contribute to our understanding of the complex mechanisms that underly the persuasion process. As such, it explored to what extent thoughtful processing predicted peoples’ choice of resistance strategies and tested if two-sided messages as one specific persuasion tactic are able to counter resistance and ultimately, achieve persuasion.
To investigate this question, this paper conducted two distinct studies that followed a quasi-experimental and randomized experimental design, respectively. Study 1 exposed the participants to a tailored counter-attitudinal message about gender quotas and examined their cognitive and affective responses to the dissonant information. The results of the analysis can be summarized as follows: First, in contrast to the hypotheses, this paper failed to demonstrate that individuals who generally enjoyed thinking or were knowledgeable about the issue were more likely to engage in an effortful thought process. Only those that felt strongly about the equal political representation of genders reported higher levels of elaboration. Second, against this paper’s expectation, the analysis revealed that people who engaged in elaborative
processing did not resist through more cognitively effortful resistance strategies. Rather, the findings suggest that elaboration leads to less resistance overall, regardless of the amount of thoughts the defense strategies required. Third, except for selective exposure and attitude bolstering, all resistance strategies significantly reduced persuasion with no meaningful difference between effortful and non-effortful strategies.
A closer look at the direct and indirect effects (Appendix A) reveals that while there is no direct effect of elaboration on persuasion (Table A1-A2), this relationship is significantly and positively mediated by resistance (Table A5)3. This finding, therefore, offers convincing
evidence that the persuasion process is indeed mediated by the peoples’ cognitive and
3 This is true for counterarguing, negative affect, self-assertion, source derogation and the derogation of the
affective responses that manifest themselves in resistance. The fact that people who processed the dissonant message more deeply were generally less resistant and, therefore, more
susceptible to persuasion may go against the common intuition. It could, however, be explained by the quality of the presented arguments: In their work, Petty & Cacioppo (1986) state that “when people are highly motivated and able to process message arguments, strong arguments are more effective than weak ones (…)” (p. 205) (see also Petty, Brinol, Loersch & McCaslin, 2009). This is because people high in elaboration will be more attentive to the flaws in weak and the virtues in strong communication (Petty & Cacioppo, 1979, p. 1920). Strong arguments, in turn, result in more favorable responses (Areni & Lutz, 1988). Applying this logic and assuming that the presented arguments were fairly strong, it could be that participants who engaged in elaborative processing generated fewer antagonistic feelings and were, therefore, less resistant and more persuadable. As this paper did not measure or
manipulate the quality of the arguments, this assumption is a mere speculation and calls for further investigation. The relatively high persuasion rate of 32% to almost 40% indicates, however, that the arguments were moderately persuasive and that the discussed mechanism could possibly be at work.
Study 2 explored the persuasiveness of two-sided messages, arguing that they are specifically useful in reducing resistance. Overall, the results reject the assumption that two-sided refutational communication is more persuasive than one-two-sided communication. The possible reasons for the inconsistent result are manifold: Scholars have claimed that the order, nature, and amount of the negative information are critical to the effectiveness of two-sided messages. According to Jackson & Allen (1987), the support-then-refute order is more effective and also more powerful than the interweaving discussion of pro and con arguments and the refute-then-support order, respectively. This notion is supported by Alba, Hutchison & Lynch (1991), who argue that placing the positive information first serves as an anchor for the subsequent evaluation of the message. The interwoven structure that this thesis employed
could have, therefore, hindered the persuasiveness of the two-sided message. Scholars have also claimed that the amount and perceived importance of the negative arguments is critical to the message’s persuasiveness (see Crowley & Hoyer, 1994; Eisend, 2006 for an overview). As both of these aspects could have possibly affected the effectiveness of this particular treatment material, it is recommended that future studies pay close attention to these moderating variables when analyzing two-sided messages. An additional analysis further revealed that the two-sided messages did not have a particularly positive effect on participants who derogated the source or counterargued the first persuasive message. These findings call the mechanisms that supposedly drive the technique’s effectiveness into question. The reverse order that was used to analyze this relationship suffers from obvious drawbacks and limits the validity of the results. Future studies are, therefore, encouraged to scrutinize this relationship in more detail.
This thesis has some notable limitations: First, the sample was skewed such that more people were initially accepting of the gender quota (66.7%) compared to those who were against the gender quota (25.6%). While this does not pose a major threat to the validity of the fully randomized Study 2, the uneven distribution of participants could have possibly biased the outcome of Study 1. Although this study controlled for a multitude of demographical and attitudinal variables, it is impossible to rule out the possibility that these groups possess considerably different attributes that might have affected the outcome. Second, the
manipulation check – which assessed to what extent people perceived the dissonant message as conflicting with their initial attitude – failed. This fact is rather surprising, given that respondents read a message that argued in the opposite direction of what they first indicated. Also, the inspection of the resistance variables proves that the participants were, in most cases, relatively resistant to the message. As such, it is not expected to have had major implications for the main variable of interest. A pre-test of the stimuli material could have nevertheless circumvented this issue.
The main goal of this paper was to provide an overview of the resistance and persuasion dynamics. To do so, it sought to consider all elements of the persuasion process and examine no less than eight resistance strategies simultaneously. This broad approach prevented a more in-depth analysis of the variables and their relationships, leaving some questions unanswered, however. For example, this thesis did not assess the quality and content of peoples’ resistance. This is particularly important for categorizing cognitively effortful vs. non-effortful resistance strategies: As discussed in the footnotes (i), whether a resistance strategy can be considered effortful depends on the amount and quality of thought people invest. As such, this paper primarily built on a theoretical argument and did not statistically test the underlying presumption. Future research should address these points and analyze the resistance strategies more thoroughly. This would enable more in-depth insights into these psychological processes, especially in regard to their relationship with elaboration.
Moreover, it seems that the implementation of gender quotas was not as controversial as expected. Therefore, it would be desirable to test the model on a more contested topic (e.g., gun control). Testing the assumptions on a subject that receives considerable attention could also equip the receivers with stronger arguments and attitudes, which could shed a different light on the processes of resistance.
In conclusion, this paper provides strong support for the notion that resistance matters and that it plays a crucial role in the persuasion process. This paper contributed to the existing literature by proposing an innovative design that tested relationships that, so far, have
overwhelmingly been studied distinctively. By analyzing these mechanisms in a systematic manner, this paper was a first attempt to harmonize the literature of political communication that seems to lack a coherent, homogenous, and exhaustive approach to the study of
resistance. As such, it aimed to set the stage for future research that investigates how voters resist persuasive messages in the political arena.
References
Abelson, R. P., Aronson, E., McGuire, W. J., Newcomb, T. M., Rosenberg, M. J., & Tannenbaum, P. (1968). Theories of cognitive consistency: A sourcebook. Chicago: Rand McNally.
Ahluwalia, R. (2000). Examination of Psychological Processes. Journal of Consumer Research, 27(9), 217–232
Alba, J. W., Hutchinson, J. W., & Lynch, J. G. (1991). Memory and decision making. In T. S. Robertson & H.H. Kassarjian (Eds.). Handbook of consumer behavior (pp. 1−49). Englewood Cliffs, NJ: Prentice Hall.
Apsler, R., & Sears, D. O. (1968). Warning, personal involvement, and attitude change. Journal of Personality and Social Psychology, 9, 162-166.
Areni, C. S., & Lutz, R. J. (1988). The role of argument quality in the elaboration likelihood model. Advances in Consumer Research, 15, 197-203.
Bachhi, Carol. (2006). Arguing for and against quotas: theoretical issues. In D. Dahlerup (Ed.), Women, quotas and politics (pp. 32–51). London: Routledge.
Berinsky, A. J., Huber, G. A., & Lenz, G. S. (2012). Evaluating Online Labor Markets for Experimental Research: Amazon.com’s Mechanical Turk. Political Analysis, 20(3), 351–368. https://doi.org/10.1093/pan/mpr057
Boden, A. O. C. (2020, Februar 10). Mandatory „woman quotas“ undermine Justice Ginsburg’s fight for equality. Retrieved from https://thehill.com/opinion/civil- rights/481430-mandatory-woman-quotas-undermine-justice-ginsburgs-fight-for-equality
Briñol, P., Rucker, D. D., Tormala, Z. L., & Petty, R. E. (2003). Individual differences in resistance to persuasion: The role of beliefs and meta-beliefs. In E. S. Knowles & J. A. Linn (Eds.) Resistance and Persuasion (pp.83–104). Mahwah, NJ: Erlbaum
social psychology, 42(1), 116.
Cacioppo, J. T., Petty, R. E., & Morris, K. J. (1983). Effects of need for cognition on message Cacioppo, J.T. and Petty, R.E. (1984) The elaboration likelihood model of persuasion.
Advances in Consumer Research, 11, 673–675.
Chaiken, S. (1980). Heuristic versus systematic information processing in the use of source versus message cues in persuasion. Journal of Personality and Social Psychology, 39, 752–766.
Chavez, K. (2020, Januar 24). Women don’t need state-mandated quotas to succeed. Retrieved from https://www.ocregister.com/2020/01/24/women-dont-need-state-mandated-quotas-to-succeed/
Cotte, J., Coulter, R. A., & Moore, M. (2005). Enhancing or disrupting guilt: the role of ad credibility and perceived manipulative intent. Journal of Business Research, 58(3), 361–368. https://doi.org/10.1016/s0148-2963(03)00102-4
Crowley, A. E. & Hoyer, W. D. (1994). An Integrative Framework for Understanding Two-Sided Persuasion. Journal of Consumer Research, 20(4), 561.
https://doi.org/10.1086/209370
Dahlerup, D. (2003). Comparative Studies of Electoral Gender Quotas. International IDEA Workshop on the Implementation of Quotas: Latin American Experiences, February 23-24, Lima, Peru:
http://aceproject.org/ero-en/topics/parties-and-candidates/Dahlerup.pdf
Dahlerup, D., & Freidenvall, L. (2010). Judging gender quotas: Predictions and results. Policy and Politics, 38(3), 407–425. https://doi.org/10.1332/030557310X521080
Daignault, P., Soroka, S., & Giasson, T. (2013). The Perception of Political Advertising During an Election Campaign: A Measure of Cognitive and Emotional
Effects. Canadian Journal of Communication, 38(2), 167–186. https://doi.org/10.22230/cjc.2013v38n2a2566
Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Fort Worth, TX: Harcourt, Brace, Jovanovich.
Eisend, M. (2006). Two-sided advertising: A meta-analysis. International Journal of Research in Marketing, 23(2), 187–198.
https://doi.org/10.1016/j.ijresmar.2005.11.001
evaluation, recall, and persuasion. Journal of Personality and Social Psychology, 45, 805– 818.
Festinger, L. (1957). A theory of cognitive dissonance. Evanston, IL: Row, Peterson. Fransen, M. L., Verlegh, P. W. J., Kirmani, A., & Smit, E. G. (2015b). A typology of consumer strategies for resisting advertising, and a review of mechanisms for countering them. International Journal of Advertising, 34(1), 6–16.
https://doi.org/10.1080/02650487.2014.995284
Freedman, J. L., & Sears, D. O. (1965). Warning, distraction, and resistance to influence. Journal of Personality and Social Psychology, 1(3), 262-266.
Grüll, P. (2020, Januar 29). IT specialist: Gender quotas is ‘how female role models are created’ Retrieved from https://www.euractiv.com/section/digital/interview/it-specialist-gender-quotas-is-how-female-role-models-are-created/
Hauser, D. J. & Schwarz, N. (2015). It’s a Trap! Instructional Manipulation Checks Prompt Systematic Thinking on “Tricky” Tasks. SAGE Open, 5(2), 215824401558461. https://doi.org/10.1177/2158244015584617
Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable moderation, mediation, and conditional process modeling. Manuscript Submitted for Publication.
Jacks, J. Z., & O’Brien, M. E. (2004). Decreasing resistance by affirming the self. In E. S. Knowles & J. A. Linn (Eds.) Resistance and Persuasion (pp.235–257). Mahwah, NJ: Erlbaum
Jackson, S., & Allen, M. (1987). Meta-analysis of the effectiveness of one-sided and two-sided argumentation. Annuel meeting of the International Communication Association, May, Montreal.
Johnson, B. T., & Eagly, A. H. (1989). Effects of involvement on persuasion: A meta-analysis. Psychological bulletin, 106(2), 290.
Kamalnath, A. (2020, Februar 26). Diversity quotas will only lead to token appointments, doing more harm than good. Abgerufen von https://theconversation.com/diversity-quotas-will-only-lead-to-token-appointments-doing-more-harm-than-good-132244 Knowles, E. S., & Linn, J. A. (2004). Resistance and persuasion. Mahwah, NJ: Erlbaum Krupnikov, Y., & Levine, A. S. (2014). Cross-Sample Comparisons and External
Validity. Journal of Experimental Political Science, 1(1), 59–80. https://doi.org/10.1017/xps.2014.7
Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin, 108(3), 480– 498. https://doi.org/10.1037/0033-2909.108.3.480
Lau, R. R. (2019). Classic Models of Persuasion. The Oxford Handbook of Electoral Persuasion, 27–50. https://doi.org/10.1093/oxfordhb/9780190860806.013.1 Lins de Holanda Coelho, G., H. P. Hanel, P., & J. Wolf, L. (2018). The Very Efficient
Assessment of Need for Cognition: Developing a Six-Item Version. Assessment, 107319111879320. https://doi.org/10.1177/1073191118793208
Lodge, M. & C. S. Taber. (2000). Three Steps Toward a Theory of Motivated Political Reasoning. In A. Lupia, M. McCubbins, & S. Popkin (Eds.). Elements of Reason: Cognition, Choice, and the Bounds of Rationality. London: Cambridge University Press.
Lydon, J., Zanna, M. P., & Ross, M. (1988). Bolstering attitudes by autobiographical recall: Attitude persistence and selective memory. Personality and Social Psychology Bulletin, 14(1), 78-86.
Maillé, C. (2015). Feminist Interventions in Political Representation in the United States and Canada: Training Programs and Legal Quotas. European journal of American
studies, 10(1), 1–20. https://doi.org/10.4000/ejas.10502
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.
Meirick, P. (2002). Cognitive responses to negative and comparative political advertising. Journal of Advertising, 31(1), 49-62.
Meirick, P. C., & Nisbett, G. S. (2011). I Approve This Message: Effects of Sponsorship, Ad Tone, and Reactance in 2008 Presidential Advertising. Mass Communication and Society, 14(5), 666–689. https://doi.org/10.1080/15205436.2010.530381
Moyer-Gusé, E., & Nabi, R. L. (2010). Explaining the effects of narrative in an entertainment television program: Overcoming resistance to persuasion. Human Communication Research, 36(1), 26–52. https://doi.org/10.1111/j.1468-2958.2009.01367.x
O’Keefe, D. J. (1999). How to Handle Opposing Arguments in Persuasive Messages: A Meta-Analytic Review of the Effects of One-Sided and Two-Sided Messages. Annals of the International Communication Association, 22(1), 209–249.
https://doi.org/10.1080/23808985.1999.11678963
O’Keefe, D. J. (2008). Elaboration likelihood model. International Encyclopedia of Communication, 4, 1475–1480.
Petty, R. E., & Cacioppo, J. T. (1979). Issue involvement can increase or decrease persuasion by enhancing message-relevant cognitive responses. Journal of Personality and Social Psychology, 37(10), 1915–1926. https://doi.org/10.1037/0022-3514.37.10.1915
Petty, R. E., & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of Persuasion. Advances in Experimental Social Psychology, 123–205. https://doi.org/10.1016/s0065-2601(08)60214-2
Petty, R. E., & Wegener, D. T. (1998). Attitude change: Multiple roles for persuasion variables. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (pp. 323–390). New York: McGraw-Hill.
Petty, R. E., Briñol, P., Loersch, C., & McCaslin, M. J. (2009). The need for cognition. Handbook of individual differences in social behavior, 318-329.
Petty, R. E., Haugtvedt, C. P., & Smith, S. M. (1995). Elaboration as a determinant of attitude strength. In R. E. Petty & J. A. Krosnick (Eds.). Attitude strength: Antecedents and consequences (pp. 93–130). Mahwah, NJ: Erlbaum
Piatti-Crocker, A. O. C. (2019, Oktober 9). Why American politics needs gender quota. Retrieved from https://thehill.com/opinion/campaign/465074-why-american-politics-needs-gender-quota
See, Y. H. M., Petty, R. E. & Evans, L. M. (2009). The impact of perceived message
complexity and need for cognition on information processing and attitudes. Journal of Research in Personality, 43(5), 880–889. https://doi.org/10.1016/j.jrp.2009.04.006 Somani, A. A. (2013). The Use of Gender Quotas in America: Are voluntary party quotas the
way to go? William and Mary Law Review, 54(1), 1451–1488. Retrieved from https://scholarship.law.wm.edu/wmlr/vol54/iss4/7/
Taber, C. S. & Lodge, M. (2006). Motivated Skepticism in the Evaluation of Political Beliefs. American Journal of Political Science, 50(3), 755–769.
https://doi.org/10.1111/j.1540-5907.2006.00214.x
Teigen, M. (2000). The affirmative action controversy. NORA - Nordic Journal of Feminist and Gender Research, 8(2), 63–77. https://doi.org/10.1080/08038740050167515 Thomsen, C. J., Borgida, E., & Lavine, H. (1995). The causes and consequences of personal
Tormala, Z. L., & Petty, R. E. (2004). Resistance to Persuasion and Attitude Certainty: The Moderating Role of Elaboration. Personality and Social Psychology Bulletin, 30(11), 1446–1457. https://doi.org/10.1177/0146167204264251
Tougas, F., Brown, R., Beaton, A. M., & Joly, S. (1995). Neosexism: Plus Ça Change, Plus C’est Pareil. Personality and Social Psychology Bulletin, 21(8), 842–849.
https://doi.org/10.1177/0146167295218007
van Reijmersdal, E. A., Fransen, M. L., van Noort, G., Opree, S. J., Vandeberg, L., Reusch, S., … Boerman, S. C. (2016). Effects of Disclosing Sponsored Content in
Blogs. American Behavioral Scientist, 60(12), 1458–1474. https://doi.org/10.1177/0002764216660141
Wegener, D. T., Petty, R. E., Smoak, N. D., & Fabrigar, L. R. (2004). Multiple routes to resisting attitude change. In E. S. Knowles & J. A. Linn (Eds.) Resistance and Persuasion (pp.13–38). Mahwah, NJ: Erlbaum
Wood, W. (1982). Retrieval of attitude-relevant information from memory: Effects on susceptibility to persuasion and on intrinsic motivation. Journal of Personality and Social Psychology, 42, 798–910.
Wood, W. & Kallgren, C. A. (1988). Communicator Attributes and Persuasion. Personality and Social Psychology Bulletin, 14(1), 172–182.
https://doi.org/10.1177/0146167288141017
Wood, W., Rhodes, N., & Biek, M. (1995). Working knowledge and attitude strength: An information-processing analysis. Attitude strength: Antecedents and consequences, 4, 189-202.
Zhou, L. (2019, August 14). Congress has a massive gender gap. Here’s how to close it. Retrieved from https://www.vox.com/the-highlight/2019/8/7/20746147/congress-women-2019-gender-parity
Zuwerink Jacks, J., & Cameron, K. A. (2003). Strategies for Resisting Persuasion. Basic and Applied Social Psychology, 25(2), 145–161.
https://doi.org/10.1207/s15324834basp2502_5
Zuwerink Jacks, J., & Devine, P. G. (2000). Attitude Importance, Forewarning of Message Content, and Resistance to Persuasion. Basic and Applied Social Psychology, 22(1), 19–29. https://doi.org/10.1207/s15324834basp2201_3
Zuwerink, J. R., & Devine, P. G. (1996). Attitude Importance and Resistance to Persuasion: It’s Not Just the Thought That Counts. Journal of Personality and Social Psychology, 70(5), 931–944. https://doi.org/10.1037/0022-3514.70.5.931
