Candidate gender: How much does it matter?
An experimental study investigating how the relationship between gender of the
candidate and the use of negative political campaigns is moderated by additional
information about the candidate.
Name: Lindsay Riley
Student Number: 11048581 Group: 1
Supervisor: J. van de Pol
Course: Graduation Project Political Communication and Journalism Assignment: Thesis
Date: January 16th 2017 Word count: 5224
Table of Contents
Abstract p. 2 Introduction p. 3 Conceptual Model p. 4 Method p. 7 Design p. 7Participants and Procedure p. 8
Stimuli p. 9
Operationalization p. 9
Dependent Measure p. 10
Manipulation Check and Control Variables p. 10
Analysis p. 11
Results p. 11
Randomization Check p. 11
Manipulation Check p. 11
Effect of the gender of the candidate on evaluation of the candidate p. 12 Moderation effect of additional information on evaluation of candidate p. 12
Conclusion and Discussion p. 13
Reference List p. 17 Appendix A: Stimuli p. 21 Stimulus Condition 1 p. 21 Stimulus Condition 2 p. 22 Stimulus Condition 3 p. 23 Stimulus Condition 4 p. 24
Candidate gender: How much does it matter?
Abstract
Studies have shown that negative advertisements can be an effective means of campaigning, because it attracts attention since people like to watch conflict. Despite the fact that it is an effective means, consultants still advise female candidates to be cautious when making use of a negative campaign. However, how much does gender really matter in certain situations? Many studies have focused on the relationship between gender and the use of negative political campaigns in terms of evaluation of the candidate. Nevertheless, they have not taken additional information, such as party, age and track record into account. Therefore, this current
experimental study aims at investigating how the relationship between gender of the candidate and the use of negative political campaigns is moderated by additional information about the candidate. In total 121 Dutch national registered voters were randomly assigned to one of the four conditions. Each condition contained an advertisement featuring a fictitious male or female candidate, with or without additional information. This current study did not find a significant moderated effect of additional information on the relationship between candidate gender and evaluation of the candidate. Furthermore, this study did not find a significant effect of candidate gender on evaluation of the candidate. However, a significant main effect of additional
information on evaluation of the candidate was found. Theoretical implications and implications for future political campaigns are also discussed in this current study.
Introduction
It is no secret that, while the number of female political leaders has more than doubled since 2005, a female political candidate is not the norm around the world. For instance, by June 2016 only two countries around the world had 50 percent or more women in parliament (UN Women, 2016). Why is it that while women comprise about half of the general population of the world (UN Women, 2016), they are still represented by a small percentage of women in politics? One impediment to women’s success in politics is known to be the role of gender stereotypes (Huddy & Terkildsen, 1993; Sanbonmatsu, 2002; Craig & Rippere, 2016; Krupnikov & Bauer, 2013).
Political campaigns are becoming more negative (Lau & Sigelman, 2000). Campaigns have especially been very negative during the last United States of America presidential election between Donald Trump and Hillary Clinton. In this current study, negative campaigns and advertisements are considered to be the same. Researchers state that negative advertisements can be an effective means of campaigning, because it attracts attention (Lee, 2014). People like to watch conflict (Lee, 2014). However, despite the fact that it is an effective means, consultants still advise female candidates to be cautious when making use of a negative campaign (Kahn & Kenney, 2000). The reason for this is because negative campaigns violate the expectation of appropriate behavior for a woman (Burgoon, Dillard, & Doran, 1983).
Many studies have focused on the relationship between gender and the use of negative political campaigns in terms of evaluation of the candidate. Many of these studies have concluded that negative campaigns hold greater risks for female than it does for male candidates (Craig & Rippere, 2016; Krupnikov & Bauer, 2013; Lee, 2014; Fridkin, Kenny, & Woodall, 2009). However, only a handful of studies have also examined how this risk interacts with the other relevant information, such as age, party, and track record of a candidate (Craig & Rippere, 2016). For instance, Craig and Rippere (2016) have conducted a study on the relationship between candidate gender, partisanship and evaluation of the candidate. In their research they concluded that partisanship plays a more important role than gender of the candidate when evaluating a candidate (Craig & Rippere, 2016). Nevertheless, they have not taken additional information, such as the candidate’s age and track record into account.
information about the candidate involved in a negative campaign moderate the effect of
candidate gender on evaluation of this candidate?” Given that the number of female politicians
have more than doubled in the last eleven years, it is important to examine whether negative campaigning is equally effective for equally competent female and male candidates. This research could benefit future political campaigns by knowing if it’s beneficial for female
candidates to go negative. Furthermore, it also helps political campaign developers in knowing if and when it is necessary to include additional information about a candidate in a negative campaign.
Conceptual Model
Before continuing to the hypotheses of this current study, it is of importance to first understand what exactly is negative campaigning. Within literature of negative campaigning, two types of definitions about negative campaigning can be distinguished, namely: an evaluative perspective and a directional perspective (Jamieson, 1992; Kamber, 1997; Mayer, 1996). In the evaluative perspective negative campaigning occurs when illegitimate criticism is given to the opponent (Jamieson, 1992; Kamber, 1997). In other words, the criticism is seen as unfair, manipulating, dishonest and irrelevant. In the directional perspective negative criticism is considered to be all forms of an attack on the opponent (Djupe & Peterson, 2002). Thus, for instance critiquing the opponent’s stand on political issues and/or critiquing the opponent as a person (Kaid &
Johnston, 1991). For this current study we will make use of the directional definition of negative campaigning, as the evaluative definition can lead to measurement bias (Walter & Vliegenthart, 2010).
The majority of people find political information difficult to process. Thus, most voters make use of ‘schemas’ (Neisser, 1976) to process information about political candidates in order to
develop evaluations (Lodge & Hamill, 1986; Rahn, 1993). Schemas are cognitive structures with which individuals receive and categorize information on a certain topic (Neisser, 1976). In other words, a schema is a mental structure that humans have of preconceived ideas. The structure of a schema helps us to make perceptions (Bem, 1983). According to Hastie (1984) all humans possess schemas in their long-term memory. Hastie (1984) states that once a schema is activated by a cue, it will encode the information it just received and set up expectations about how the world works. This expectation will give meaning onto the received information (Hastie, 1981). In the political arena voters often give meaning to certain messages by making use of schemas. These schemas assist voters in understanding a candidate’s position, fill in missing
information, make conclusion about what a candidate might be like as a person and generate evaluations of a candidate (Conover & Feldman, 1989).
Unlike schemas in general, gender schematic processing occurs when a meaning is given to certain information based on gender (Bem, 1983). It is when people evaluate others based on what is considered to be male and female, regardless of their individual differences (Bem, 1981). Gender schematic processing occurs particularly when little is known about a certain person (Craig & Rippere, 2016). As gender schematic processing is considered to be one of the most powerful schemas, it is expected to be present when voters watch a negative
advertisement about a candidate. In the study of Krupnikov and Bauer (2013) it is for instance evident that voters do rely on gender to evaluate a candidate. They conducted a study on the relationship between campaign negativity, gender and campaign context. They concluded that when female candidates attack their opponent, they suffer more in evaluation than when a male candidate attacks their opponent.
Thus, gender stereotyping is an example of schematic processing. Gender stereotyping occurs when over-generalizations about the characteristics of a group as a whole are made based on gender (Hitchon & Chang, 1995). On account of these stereotypes people view men as
independent, aggressive and forceful, while women are viewed as passive, sympathetic and less inclined to be aggressive (Diekman & Eagly, 2000). These views that individuals have of men and women also lead to expectations that people have about appropriate communication behavior for males and females (Miller & Burgoon, 1979; Burgoon et al., 1983; Lee, 2014; Sanbonmatsu, 2002; Chang & Hitchon, 2004; Gordon, Shafie, & Crigler, 2003). As mentioned before, this expectation will give meaning onto the received information (Hastie, 1981).
According to several studies (Huddy & Terkildsen, 1993; Gordon et al., 2003; Sanbonmatsu, 2002; Chang & Hitchon, 2004) people have an expectation that male candidates are better equipped to communicate about issues such as military and economy whereas female candidates are better equipped to communicate about issues such as education and health care. When the expectations that people have about appropriate communication behavior for males and females are violated, it can affect persuasion in a negative manner (Burgoon et al., 1983). In other words, people are less persuaded by a communication technique when their expectations about appropriate communication behaviors are not met. According to Taylor and
Negative advertising is seen as aggressive and assertive; women are not expected to employ these characteristics. Therefore, in regards to negative campaigning, female candidates may suffer more in evaluation, because negative campaigning violates the expectation of appropriate behavior for a woman. In the study of Krupnikov and Bauer (2013) for instance, it is evident that negative campaigns translates into poorer evaluations for a female candidate compared with a male candidate. Furthermore, in the study of Craig and Rippere (2016), it is evident that when a female candidate attacks her opponent, she suffers more in evaluation than when a male
candidate attacks his opponent. Based on these theories and studies, this study predicts that an identical negative campaign generates lower evaluations for a female candidate than it does for a male candidate.
H1: Female sponsor-candidates of a negative campaign receive a lower evaluation than male sponsor candidates of a negative campaign.
As argued before, gender schematic processing occurs particularly when little is known about a certain person. One of the functions of schemas is to fill in the gap when there is information missing (Chang & Hitchon, 2004). This ‘fill in the gap’ process is referred to as default
processing (Taylor & Crocker, 1981). Default processing occurs when information is missing or when information is open to more interpretations and the receiver of the information is then most likely to make a conclusion based on what is already known in their schema.
Based on this logic, the author of this current study predicts that when more information about a candidate is available, the voter might rely less on the gender of the candidate. Craig and Rippere (2016) conducted a study on the relationship between candidate gender, partisanship and evaluation of the candidate. In their research they concluded that partisanship plays a more important role than gender of the candidate when evaluating a candidate. In addition, Chang and Hitchon (2004) conducted a study on the relationship between candidate gender, information about a candidate’s strength on issues and evaluation of the candidate. Their research indicates that female candidates were evaluated lower than male candidates;
however, this bias only seemed to operate in the absence of information about the candidate’s strengths on issues. Thus, when more information such as party, age of the candidate and track record is known about a candidate, voters might focus less on the gender of the candidate.
H2: When additional information (age, party membership and track record) about the candidate is given in the negative campaign, the effect of gender on evaluation will be less strong.
The conceptual model is visually displayed in Figure 1.
Figure 1. Conceptual model.
Method Design
The current study is part of a larger project designed to assess the impact of both negative and positive political advertisements on the evaluation of the candidate, efficacy and cynicism. To test the hypotheses we made use of an online experimental design, since an experiment
enables us to manipulate negativity and the amount of information given about a candidate. The design of this online experiment was a 2 (gender of candidate) x 2 (positive or negative
advertisement) x 2 (additional information or not) design. However, this current study will only focus on negative political advertisements and not positive advertisements. The results of the remaining conditions will be reported elsewhere. The design of this current study is displayed in Table 1.
Evaluation of the candidate Information available
about candidate’s age, party and track
record Male= 0 Female= 1 Gender of the candidate in negative ad
Table 1
Design of the experiment
Condition number Candidate gender Extra information or not
Condition 1 Male candidate No extra information Condition 2 Male candidate Extra information Condition 3 Female candidate No extra information Condition 4 Female candidate Extra information
Participants were randomly assigned to one of the four conditions. In each condition they were exposed to an advertisement for a fictitious male or female candidate. Afterwards, they were asked to fill in a questionnaire.
Participants and Procedure
The study relied on a sample of Dutch national registered voters that were recruited through a Qualtrics link on the social media channels of the researchers. In total, N= 121 Dutch national registered voters participated in this current study. The sample consisted of participants of the age 18 and older (M= 37, SD= 17,4). 63,6% of the participants were female and 36,4% of the participants were male. 79,3% of the participants consisted of individuals with a higher
education level and only 20,7% were individuals with a lower education level. Thus, there was an overrepresentation of females and individuals with a high education; this could be due to the use of a convenience sample. The data was collected from the 7th of November 2016 to the 28th of November 2016.
Before continuing to the experiment, the participants were asked to read the introduction and fill in the consent form. They were given limited information about the nature of the study in order to avoid changes in their behavior due to priming effects. The participants were told that they would have to read a political advertisement by a political candidate and then they would be asked a few questions. After filling in the questionnaire, the participants were informed about the fictitious candidates on the picture of the advertisement. The entire procedure lasted
Stimuli
Print advertisements within the newspaper, ‘De Volkskrant’, were created with the editing software program InDesign. The advertisement in all conditions includes a text attacking the party ‘Volkspartij voor Vrijheid en Democratie (VVD)’. The advertisement will be shown in a newspaper to make it look as realistic as possible. The chosen newspaper is ‘De Volkskrant’, as it is known as a quality newspaper. Quality newspapers are known for their seriousness
(Schaap & Pleijter, 2012). By making use of a quality newspaper, opinions about the newspaper’s seriousness would not affect the experiment. Pictures that represented the fictitious candidates of the province Friesland were included. In order to ensure that the participants do not have preexisting attitudes about the candidates, fictitious candidates were used. Since none of the researchers had online acquaintances in the province of Friesland, we decided to make use of the fictitious candidates within this province. Screenshots of the stimuli are presented in Appendix A. Number of words, picture size and layout were controlled across the conditions. The advertisements were manipulated to vary on the following variables: candidate gender and additional information.
Operationalization
In order to distinguish whether a male or a female candidate sponsored an advertisement, cues such as the name of the candidate, the picture of the candidate and gender-specific pronouns were given. In each condition, the picture was positioned at the top right corner of the
advertisement. The female candidate’s name was ‘Carina’ and the male candidate’s name was ‘Bas’. To ensure full control, both candidates’ last name was ‘van Dijk’ across all conditions.
Negative advertisements in the condition in which additional information was not given
addressed the wrongdoings and criticized the weaknesses of the attacked opponent. Negative advertisements in the condition in which additional information was given, were created exactly the same as the ones in which no additional information was given, except for the fact that they also included the age, the track record and the party of the candidate. In these conditions the attacker belonged to the party ‘Democrats 66 (D66)’. The party D66 was chosen, because it is known to have right as well left positions, and could therefore arguably be the most center-oriented party in the Netherlands.
Dependent Measure
The measures of the dependent variable ‘evaluation of the candidate’ in this current study are based on the measures used in Baumgartner and Morris’ (2006) study. This variable was an additive index of several items. The participants were asked to rate how well some personality traits described each candidate. These traits were: (a) “really cares about people like me”, (b) “knowledgeable”, (c) “provides strong leadership”, (d) “inspiring”, (e) “honest”, (f) “competent as a politician”, (g) “decisive”. All traits were assessed using a 7-point Likert Scale (1= strongly disagree, 2= disagree, 3= disagree somewhat, 4= undecided, 5= agree somewhat, 6= agree, 7= strongly agree). The higher index scores represent more positive evaluations of the candidates and the lower index scores represent more negative evaluations of the candidates. A factor analysis with the Varimax rotation produced one component with an eigenvalue more than 1,0. The factor loadings for each item are displayed in Table 2. These items explained 70,9% of the variance. Thus, one scale was created in order to assess evaluation of the candidate. The Cronbach’s reliability alpha for the scale is deemed satisfactory at 0,93.
Table 2
Factor Loadings with Varimax rotation for 7 items of evaluation of the candidate
Evaluation of the candidate
Provides strong leadership 0,883
Decisive 0,862
Competent as a politician 0,857
Honest 0,852
Inspiring 0,849
Really cares about people like me 0,809
Knowledgeable 0,779
Manipulation Check and Control Variables
Gender, age and education of the participant were also taken into account as control variables within the analyses. Furthermore, to ensure that the findings of this study result from the focal factors instead of unexpected factors, two additional manipulation check questions were asked:
“What is the gender of the politician featured on the picture of the article?” and “To which party does this politician belong to?”. In order to measure the participants’ preexisting attitudes about the parties D66 en VVD, a political preference question was asked before the stimuli was shown. The complete questionnaire can be found in Appendix B.
Analysis
In order to analyze both hypotheses a factorial ANOVA analysis will be conducted. The independent categorical variable in hypothesis 1 is gender. The independent categorical variables in hypothesis 2 are: gender of the candidate and additional information or not. The dependent continuous variable in both hypotheses is ‘evaluation of the candidate’.
Results Randomization Check
In order to ensure that the randomization was done correctly, a Oneway ANOVA analysis has been conducted. The randomization process appears to have been successful. No statistically significant differences were observed among the members of the four conditions regarding age:
F(3, 117) = 1,55, p = 0,207; gender: F(3, 117) = 0,63, p = 0,597; and education: F(3, 117) =
0,16, p = 0,922.
Manipulation Check
To examine the effectiveness of the manipulation of the gender of the politician, the participants were asked to indicate whether they have seen a male or a female politician on the print
advertisement. 86,1% of those in the conditions featuring a male politician and 92,9% of those in the conditions featuring a female politician, recalled the correct gender. Therefore, the manipulation on gender was successful.
To examine the manipulation of additional information, the participants were asked to which party the politicians belonged to. 72,5% of those in the conditions in which additional information was not given answered the question correctly. 81,4% of those in the conditions featuring
additional information also answered the question correctly. Therefore, the manipulation of additional information was successful.
Table 3
Candidate gender x Additional information Factorial Analysis of Variance for Evaluation of the candidate
Sum of Squares
df Mean Square F sign
Candidate gender 0,08 1 0,08 0,07 0,788 Additional information 14,02 1 14,02 12,17 0,001 Candidate gender * Additional information 0,52 1 0,52 0,46 0,502 Error 128,97 112 1,15
Effect of the gender of the candidate on evaluation of the candidate
Hypothesis 1 predicts that female sponsor-candidates of a negative campaign receive a lower evaluation than male sponsor-candidates of a negative campaign. In other words, participants will rate a female sponsor-candidate lower than a male sponsor-candidate based on their gender. To test hypothesis 1 a factorial ANOVA analysis was conducted. Within the sample, those who were in the condition featuring a female candidate with no additional information evaluated the politician slightly higher (M = 3,93, SD = 1,07) than those who were in the condition featuring a male with no additional information (M = 3,85, SD = 1,21). Despite the difference, we did not find a significant effect of gender of the candidate on evaluation of the candidate. Table 3 indicates an insignificant main effect of gender of the candidate on evaluation of the candidate. Therefore, hypothesis 1 was not supported.
Moderation effect of additional information on evaluation of candidate
Hypothesis 2 predicts a less strong effect of candidate gender on evaluation when this relationship is moderated by additional information. To test hypothesis 2 a factorial ANOVA analysis was conducted. Table 3 indicates an insignificant interaction effect between candidate gender and additional information. Therefore, hypothesis 2 was not supported.
Furthermore, Table 3 also indicates a moderate significant main effect of additional information on evaluation of the candidate. Those who were in the conditions in which no additional
information was given, evaluated the candidate slightly lower (M = 3,89, SD = 1,13) than those who were in the conditions with additional information (M = 4,60, SD = 0,96). The effect size is, η2 = 0,10.
Conclusion and Discussion
The purpose of this current study was to investigate how the relationship between gender of the candidate and the use of negative political campaigns is moderated by additional information about the candidate. This was done with the following research question: “How does additional
information about the candidate involved in a negative campaign moderate the effect of candidate gender on evaluation of this candidate?” Contrary to previous studies (Craig &
Rippere, 2016; Chang & Hitchon, 2004) focusing on candidate gender and evaluation of the candidate, this current study did not find a significant effect of candidate gender on evaluation of the candidate. In addition, this current study did not find a significant moderated effect of
additional information on the relationship between candidate gender and evaluation of the candidate. Nevertheless, a significant main effect of additional information on evaluation of the candidate was found. In other words, regardless of the gender of the candidate, additional information in a negative campaign leads to a higher evaluation of the candidate.
Several studies (Miller & Burgoon, 1979; Burgoon et al., 1983; Lee, 2014; Sanbonmatsu, 2002; Chang & Hitchon, 2004; Gordon, Shafie, & Crigler, 2003) suggest that views that individuals have of men and women lead to expectations of appropriate communication behavior for males and females. When these expectations are violated it can affect persuasion (Burgoon et al., 1983). That is to say, people are less persuaded by a communication technique when their expectations about appropriate communication behaviors are not met. According to Taylor and Falcone (1982), less persuasion leads to a lower evaluation of a political candidate. Therefore, it was expected that when female candidates make use of negative campaigns, they receive a lower evaluation than when male candidates make use of negative campaigns. This current study, however, did not find any evidence to support this logic.
Previous theories and studies about gender schematic processing and political gender stereotyping (Taylor & Crocker, 1981; Craig & Rippere, 2016; Chang & Hitchon, 2004) have suggested that more information about a candidate leads to a better evaluation for female
candidate was available. However, this current study did not find any evidence to support this. There was no significant difference between the conditions with and without additional
information in the effect of gender on evaluation of the candidate.
A reason for the insignificant results could be because of the size of the sample and the sampling method. This current study consisted of a small sample size. A small sample size could have led to insufficient statistical power to detect differences between the conditions. Future studies should increase the size of the sample in order to ensure significant results. Moreover, this current study made use of a convenience sample, since it relied on a sample of Dutch national registered voters that were recruited through a Qualtrics link on the social media channels of the researchers. Thus, a selection bias could have taken place. Therefore, this study has a low external validity. According to Petty and Cacioppo (1986) people who are personally interested in a certain topic tend to process information in a critical manner. Since the researchers of this current study are all interested in politics, it could be that acquaintances of the researchers hold similar political interest. This could have led to an overrepresentation of voters with high political interest. If this is indeed the case, it could be that their evaluations depend more on the political issues instead of the gender and additional information of the candidate. However, this is simply a speculation and is beyond the scope of this study. Future research would have to look further into this.
An additional possible reason for the insignificant results is the overrepresentation of female participants in the sample. Sanbonmatsu (2002) conducted a study on the relationship between gender stereotyping and vote choice. As appears from the results of her study, men tend to have a preference for male candidates and women towards female candidates (Sanbonmatsu, 2002). This is what Sanbonmatsu (2002) refers to as the ‘baseline gender preference’. In this current study female participants could have been less critical about the female candidate and therefore making additional information about the candidate also less important for them. This difference between female and male participants in this study could have disabled the possibility to make proper comparison between the conditions. Future research would have to look further into this.
The insignificant results could have also been caused because of the overrepresentation of individuals with a higher education level in the sample. The overrepresentation of higher educated individuals disabled the possibility to make a proper comparison between the
conditions. Furthermore, the overrepresentation of higher educated individuals also influences the generalizability of this study. Future studies should ensure that the ratio of education level is equal.
Although no significant results were found for the two hypotheses, this current study did find a significant main effect of additional information on evaluation of the candidate. This suggests that, when voters receive more information about a certain candidate in a negative campaign, they tend to evaluate the candidate higher than when they do not receive more information about a candidate, regardless of the gender of the candidate. This result is consistent with previous research on the effect of civility of a negative campaign on evaluation of the candidate and trust (Fridkin & Kenney, 2011; Fridkin & Kenney, 2008; Mutz & Reeves, 2005). Uncivil negative campaigns have a strident tone while civil campaigns have a more respectful tone (Fridkin & Kenney, 2008). It is evident that uncivil negative campaigns continuously translate into more negative evaluations of candidates (Fridkin & Kenney, 2011; Mutz & Reeves, 2005). Voters state that these uncivil negative campaigns are unfair and make a candidate seem less decent (Fridkin & Kenney, 2011). By including additional information about the candidate in this current study, it is possible that it may make the candidate seem more fair and decent. Thus, making the negative campaign more civil. This could explain why it is that when voters receive additional information about a candidate in a negative campaign, they evaluate this candidate more positively.
In this current study I have deliberately chosen to exclude a pretest from the current study. By having a pretest participants could have noticed that their evaluations of the candidate are being measured, therefore influencing the answers given in the posttest. This could have had a
negative influence on the effect of the stimulus. Thus, by only making use of a pretest a testing effect was prevented.
However, the results of this study are of course not without limitations. This current study did not take the type of negativity into account. In other words, for example whether the negativity within the campaign was issue driven, civil or personality driven. According to Fridkin et al. (2009) different types of negativity produces different gender effects. By not taking this into account, I could have not examined how negativity affected evaluations of the candidate.
Furthermore, this current study deliberately excludes any consideration of whether the negative campaign consists of a female topic or a male topic. According to Gordon et al. (2003) and Lee (2014) this is a factor that may influence the way voters look at negative advertisements. Thus, for instance if a candidate is a female she is expected to pay attention to female-linked issues, such as education and healthcare. Male candidates, on the other hand, are expected to pay attention to male-linked issues, such as military and economy. Future research should include the type of topic of the negative campaign when analyzing the effect of candidate gender on evaluation of the candidate.
Due to the scope of this study, we were not able to examine the role of repetition. The participants of this current study viewed one single attack by a candidate. This does not
represent a complete version of realistic political communication. Fernandes (2013) states that too much repetition in a short period of time can create a backlash towards the
sponsor-candidate of the negative campaign. King and McConnell (2003) add to this by stating that repetition of a negative campaign backlashes against a female candidate and not particularly against a male candidate. Thus, future studies should examine whether this is the case when additional information is given.
Despite its limitations this current study does have some implications for future political
campaign developers. First, regardless of the gender of the candidate, it is of great importance to always include additional information of the candidate when making use of negative
campaigns. This could be for instance information about the candidate’s experience in the political arena. In addition, this current study also suggests that experimental results cannot easily be used to represent the real life. In real life voters are likely to have preexisting evaluations of a candidate, which may influence their evaluations after seeing a negative campaign of the candidate. Thus, a suggestion for future research is to look for ways to include real politicians in an experiment in order to make it more applicable for real life situations. However, they would have to ensure that the popularity of the candidates is randomly assigned to each condition.
Although this current study did not find evidence to support the hypothesis, that additional information in a negative campaign decreases the effect of gender on evaluation, female as well as male politicians can still see this result in a positive light. As they say: “No news is good news!”.
Reference List
Baumgartner, J., & Morris, J. (2006). The daily show effect: Candidate evaluations, efficacy, and American youth. American Politics Research, 34(3), 341-367. doi:10.1177/1532673x05280074
Bem, S. L. (1983). Gender schema theory and its implications for child development: Raising gender-aschematic children in a gender-schematic society. Signs: Journal of Women in Culture
and Society, 8(4), 598-616. doi:10.1086/493998
Burgoon, M., Dillard, J. P., & Doran, N. E. (1983). Friendly or unfriendly persuasion. The effects of violations of expectations by males and females. Human Communication Research, 10(2), 283-294. doi:10.1111/j.1468-2958.1983.tb00018.x
Chang, C., & Hitchon, J. C. B. (2004). When does gender count? Further insights into gender schematic processing of female candidates’ political advertisements. Sex Roles, 51(3/4), 197-208. doi:10.1023/b:sers.0000037763.47986.c2
Conover, P. J., & Feldman, S. (1989). Candidate perception in an ambiguous world:
Campaigns, cues, and inference processes. American Journal of Political Science, 33(4), 917-940. doi:10.2307/2111115
Craig, S. C., & Rippere, P. S. (2016). He said, she said: The impact of candidate gender in negative campaigns. Politics & Gender, 12(2), 391-414. doi:10.1017/s1743923x15000550
Diekman, A. B., & Eagly, A. H. (2000). Stereotypes as dynamic constructs: Women and men of the past, present, and future. Personality and Social Psychology Bulletin, 26(10), 1171-1188. doi:10.1177/0146167200262001
Fernandes, J. (2013). Effects of negative political advertising and message repetition on candidate evaluation. Mass Communication and Society, 16(2), 268-91.
Fridkin, K. L., & Kenney, P. J. (2011). Variability in citizens’ reactions to different types of negative campaigns. American Journal of Political Science, 55(2), 307-325. doi:10.1111/j.1540-5907.2010.00494.x
Fridkin, K. L., Kenney, P. J., & Woodall, G. S. (2009). Bad for men, better for women: The impact of stereotypes during negative campaigns. Political Behavior, 31(1), 53-77.
doi:10.1007/s11109-008-9065-x
Gordon, A., Shafie, D. M., & Crigler, A. N. (2003). Is negative advertising effective for female candidates? An experiment in voters’ uses of gender stereotypes. The International Journal of
Press/Politics, 8(3), 35-53. doi:10.1177/1081180x03008003003
Hastie, R. (1984). Causes and effects of causal attribution. Journal of Personality and Social
Psychology, 46(1), 44-56. doi:10.1037//0022-3514.46.1.44
Hitchon, J. C., & Chang, C. (1995). Effects of gender schematic processing on the reception of political commercials for men and women candidates. Communication Research, 22(4), 430-458. doi:10.1177/009365095022004003
Huddy, L., & Terkildsen, N. (1993). Gender stereotypes and the perception of male and female candidates. American Journal of Political Science, 37(1), 119-147. doi:10.2307/2111526
Kahn, K., & Kenney, P. (2000). Negative campaigning and knowledge of senate elections. In J. Thurber, C. Nelson, & D. Dulio (Eds.), Crowded Airwaves (pp. 45-61). Washington D.C.: Brookings Institution.
Kaid, L. L., & Johnston, A. (1991). Negative versus positive television advertising in U.S. presidential campaigns, 1960-1988. Journal of Communication, 41(3), 53-64.
doi:10.1111/j.1460-2466.1991.tb02323.x
King, J. D., & McConnell, J.B. (2003). The effect of negative campaign advertising on vote choice: The mediating influence of gender. Social Science Quarterly (Blackwell Publishing
Krupnikov, Y., & Bauer, N. M. (2013). The relationship between campaign negativity, gender and campaign context. Political Behavior, 36(1), 167-188. doi:10.1007/s11109-013-9221-9
Lau, R. R., & Sigelman, L. (2000). Effectiveness of negative political advertising. In J. Thurber, C. Nelson, & D. Dulio (Eds.), Crowded Airwaves (pp. 172-175). Washington D.C.: Brookings Institution.
Lee, Y. (2014). Gender stereotypes as a double-edged sword in political advertising: Persuasion effects of campaign theme and advertising style. International Journal of Advertising, 33(2), 203-234. doi:10.2501/IJA-33-2-203-234
Lodge, M., & Hamill, R. (1986). A partisan schema for political information processing. The
American Political Science Review, 80(2), 505-519. doi:10.2307/1958271
Miller, M. D., & Burgoon, M. (1979). The relationship between violations of expectations and the induction of resistance to persuasion. Human Communication Research, 5(4). 301-313.
doi:10.1111/j.1468-2958.1979.tb00642.x
Mutz, D. C., & Reeves, B. (2005). The new videomalaise: Effects of televised Incivility on
political trust. American Political Science Review, 99(1), 1-15. doi:10.1017/s0003055405051452
Neisser, V. (1976). Cognition and reality: Principles and implications of cognitive psychology. San Francisco: Freeman.
Petty, R. E., & Cacioppo, J. T. (1984). Communication and persuasion: Central and peripheral
routes to attitude change. New York: Springer.
Rahn, W. M. (1993). The role of partisan stereotypes in information processing about political candidates. American Journal of Political Science, 37(2), 1915-1926. doi:10.2307/2111381
Schaap G. J., & Pleijter, A. R. J. (2012). Het sensatiegehalte van voorpaginafoto’s: Een inhoudsanalyse van populaire en kwaliteitskranten in Nederland. Tijdschrift voor
Communicatiewetenschap, 40(1), 71-86.
Taylor, S. E., & Crocker, J. (1981). Schematic bases of social information processing. In E. T. Higgins, C. A. Harman, & M. P. Zanna (Eds), Social cognition: The Ontario symposium (pp. 89-134). Hillsdale, NJ: Erlbaum.
Taylor, S. E., & Falcone, H. T. (1982). Cognitive bases of Stereotyping: The relationship between Categorization and prejudice. Personality and Social Psychology Bulletin, 8(3), 426-432. doi:10.1177/0146167282083006
UN Women. (2016). Facts and figures: leadership and political participation. Retrieved on September 25, 2016 from
http://www.unwomen.org/en/what-we-do/leadership-and-political-participation/facts-and-figures
Walter, A. S., & Vliegenthart, R. (2011). Negative campaigning across different communication channels: different ball games?. International Journal of Press Politics, 15(4), 441-461.
Appendix A: Stimuli
Stimulus Condition 1
Stimulus Condition 2
Stimulus Condition 3
Stimulus Condition 4
Appendix B: Questionnaire
Politieke Communicatie 2016
Allereerst willen wij u bedanken voor uw deelname aan ons onderzoek. Voordat u begint met het invullen van de vragenlijst willen wij u vragen onderstaande tekst te lezen. Het gaat om een korte introductie en uitleg over het onderzoek en het akkoord gaan met dit onderzoek.
In dit onderzoek wordt eerst om uw demografische gegevens (geslacht, leeftijd, et cetera) gevraagd. Daarna krijgt u een krantenbericht te zien dat u goed moet lezen. Hierna zullen enkele vragen en stellingen aan u worden voorgelegd die u mag beantwoorden.
Ik verklaar hierbij op voor mij duidelijke wijze te zijn ingelicht over de aard en methode van het onderzoek. Ik stem geheel vrijwillig in met deelname aan dit onderzoek. Ik behoud daarbij het recht deze instemming weer in te trekken zonder dat ik daarvoor een reden hoef op te geven. Ik besef dat ik op elk moment mag stoppen met het onderzoek. Als mijn onderzoeksresultaten worden gebruikt in wetenschappelijke publicaties, of op een andere manier openbaar worden gemaakt, dan zal dit volledig geanonimiseerd gebeuren. Mijn persoonsgegevens worden niet door derden ingezien zonder mijn uitdrukkelijke toestemming.
Voor eventuele klachten over dit onderzoek kan ik me wenden tot het lid van de Commissie Ethiek namens ASCoR, per adres: ASCoR secretariaat, Commissie Ethiek, Universiteit van Amsterdam, Nieuwe Achtergracht 166, 1018 WV Amsterdam; 020‐ 525 3680; ascor‐ secr‐ fmg@uva.nl.
Heeft u de bovenstaande tekst goed begrepen en gaat u akkoord met verdere deelname aan het onderzoek?
o Ja o Nee
1. De volgende vraag gaat over u politieke voorkeur. Zou u in de toekomst misschien op deze partij stemmen? Ja Nee PvdA o o SP o o VVD o o CDA o o D66 o o PVV o o
2. De volgende stellingen gaan over uw mening over de politicus op de foto in het
krantenartikel. Geef hieronder aan in hoeverre u het eens bent met de volgende stellingen: De politicus op de foto in het krantenartikel...
Helemaal mee oneens Mee oneens Een beetje mee oneens Neutraal Een beetje mee eens Mee eens Helemaal mee eens is slim o o o o o o o geeft om mensen zoals ik o o o o o o o is een sterke leider o o o o o o o is inspirerend o o o o o o o Is oprecht o o o o o o o is competent o o o o o o o Is besluitvaardig o o o o o o o 3. Wat is uw geslacht? o Man o Vrouw 4. Wat is uw leeftijd? ____________
5. Wat is uw hoogst genoten opleiding? Dat wil zeggen wat is de hoogste opleiding die u volgt of heeft gevolgd.
o Geen onderwijs/ basisonderwijs
o LBO/ VBO/ VMBO (kader- en beroepsgerichte leerweg)
o Eerste 3 jaar HAVO en VWO/ MAVO/ VMBO (theoretisch en gemengde leerweg) o MBO
o HAVO en VWO bovenbouw/ WO en HBO propedeuse o HBO of WO-bachelor of kandidaats
o WO- doctoraal of master o Geen opgave
6. Wat is het geslacht van de politicus op de foto in het krantenartikel dat u net gelezen heeft? o Man
7. Bij welke partij zat deze politicus? o D66 o SP o CDA o PvdA o VVD
