Political differentiation and consolidation of choice in a U.S. media malaise environment: indirect effects of perceived alikeness on voter participation.

Political marketing campaigns expend enormous effort each campaign season to influence voter turnout. This cyclical democratic process and nonstop news cycle foster an environment of media malaise. Voter pessimism undercuts participation through increased perceived alikeness among ballot options. Differentiation and consolidation theory describe the voting decision process as reconciling rational and irrational information. Voters seek out differences to decide among presented options. More politically interested voters are more likely to vote. Counterintuitively, higher political organizational avocational interest is related to higher perceived alikeness. Across three studies, higher perceived alikeness of parties, candidates, and issues was related to a lower likelihood to vote (LTV). Conditional voting ineffectual beliefs exacerbated these indirect effects on LTV. In a saturated marketing atmosphere with massive spending during each election cycle, we discuss implications to influence LTV based on results.

Introduction

Political marketing is the use of marketing principles and strategies to influence the political process (Harris and McGrath 2012; Henneberg and O’Shaughnessy 2007). U.S. presidential and congressional elections fundraise and spend upwards of billions of dollars on political marketing campaigns (Federal Election Commission (FEC) 2021). The FEC (2021) reported over $4 billion in presidential campaigns and another $4 billion in congressional candidates’ receipts for the 2020 election cycle (from January 1 of the pre-election year through December 31 of the election year). These dollars influence voter perceptions, decisions made by politicians, and party politics (O’Cass 1996, 2009; Osuagwu 2008).

However, less studied are the effects of perceived alikeness of political parties, candidates, and issues on voter participation. For example, U.S. voters often discuss elections as a choice between two mediocre options (e.g., “Hold your nose and vote.”). We define perceived alikeness as the degree of similarity between two entities (0% alike to 100% alike). Despite political rallies and engrained civic responsibility (O’Brien and Macoun 2021), a vast portion of the voting-aged population does not vote (e.g., 33.2–40.5% between 2000 and 2020 for presidential elections) (U.S. Census Bureau 2021). The current project quantifies the perceived degree of similarity among political choices with electoral participation.

In fact, the 66.8% voter turnout of U.S. citizens of voting age in the 2020 presidential election was an all-time high. This turnout was partially attributed to the stark differences in presidential candidates and party platforms. Then Democratic presidential candidate, Joe Biden, called the election a “Battle for the soul of the nation” (Cummings 2020). Former President and Republican candidate Donald Trump accused his opponent of ending American liberties: “No guns, no religion, no energy, no oil. Remember that. Remember” (Martin 2020). Voters turned out for clear reasons (and many in opposition to the other political party) (Kernell 1977). This election had global consequences and voters risked casting in-person ballots despite the raging COVID-19 pandemic (Youde 2021).

However, perceived alikeness is the norm in twenty-first century U.S. politics. While campaigns emphasize their candidate is better for the working middle class citizens, coal industry, and military, voters often feel they are relatively the same. No doubt, agreeability is a likeable trait. It is difficult to appear likeable when someone expresses disagreement. Politicians voice agreements to build credibility and prepare counterarguments. However, with generally low voter attention spans (Heinrich et al. 2018) and skepticism that politicians will enact policies that help the working class (Fuhrman 1993; Skocpol and Amenta 1986), it is vital to articulate stark differences. With barriers to voting (such as long lines to cast a ballot), voters need strong motives (Pettigrew 2020). If candidates and parties are perceived about the same, what difference would it make who is in control of government? Hence, this study focuses on the lack of dissimilarity and the impact on likelihood to vote (LTV).

Literature review

Differentiation and consolidation theory

Political campaigns share commonalities with product choices (Baines et al. 2003). Voters follow a similar decision-making process for options at the ballot box as when choosing to buy products (Lau and Redlawsk 2006, p. 51). Purchasing preferences among a given set of choices favors a product with differentiating features (Chernev 2005). Differentiation and consolidation theory (DCT) describes the process of evaluating alternatives with information over time for a unique preferred option (Svenson 1992; Swider et al. 2015). Individuals seek differences so choices remain justifiable after selection. DCT decision making follows a three-stage process (Meyer 2018). First, in pre-decision, individuals recognize a problem and identify alternatives. Second, in differentiation, individuals establish criteria and assess available information. Third, in consolidation, individuals gain confidence about their decision and attempt to minimize negative outcomes. Differentiation provides unique offerings that individuals use to weigh the pros and cons of choices in the second stage. Individuals use this information to form a preference and settle on a decision.

Voters evaluate election decisions and choose the best perceived option (Geys 2006; Simon 1955). Such decisions include voting for certain candidates, registering to vote, or even whether to cast a ballot at all. Individual and societal benefits factor into the voting decision process. Political campaigns espouse support for values they believe voters will cast a ballot for if elected to control government (Downs 1957; Lau and Redlawsk 2006). Voters have limited immediate storage capacity to perform complex maximum utility calculations (Simon 1956). Social, emotional, and environmental structures obstruct a purely rational utilitarian decision (Simon 1957). Furthermore, political marketing campaigns manipulate perceptions with advertisements and other strategies (Baines et al. 2003; Baines and Egan 2001; Banker 1992; Butler and Harris 2009).

With politicians claiming similar virtues (e.g., patriotism, strong military, and improved economy), it becomes difficult to distinguish differences (Sigelman and Buell Jr 2004). Who will fight for my values (e.g., living wage/worker rights)? Which party is better at creating jobs? Which policy is better for our education system (e.g., common core/standardized testing)? Effective political marketing campaigns can influence voters’ perceptions for decades (e.g., Republicans lower taxes, Democrats support civil rights) (Jacobs and Helms 2001).

Differentiation is a basic tenet in marketing and advertising. Specifically, perceived and valued differentiation has substantive importance in decision making (Romaniuk et al. 2007). The success of many twenty-first century political parties and candidates across democratic nations was partially attributed to effective political branding (Needham and Smith 2015; Schneider 2004; Smith and Speed 2011). Branding uses a logo, name, term, sign, symbol, and/or design to identify a product/service (Armstrong et al. 2014, p. 221). Ongoing marketing activates these memory nodes to create brand relationships and perceived added value (Farquhar and Herr 1993; Keller 1993). Political branding is the interaction of branding tactics among leaders (e.g., name and banner design), their party (e.g., convention theatrics), and policies (e.g., The New Deal and The Patriot Act) to differentiate (Cwalina and Falkowski 2015; Smith and French 2009; Speed et al. 2015).

Differentiating simplifies the decision-making process by reducing effort and aiding in the consideration of alternatives (Swider et al. 2015). For instance, when choosing among workplace recruits, hirers focus on differences to determine preferred candidates. When political candidates highlight similarities, voters expend more effort to discern differences. When politicians do not distinguish themselves with unique offers, contemporary U.S. media have established politics as corrupt and unconstructive. This existing narrative can lump options together as no different. This arguably dissuades making any choice (Dhar 1997). Not participating in elections becomes justifiable because there is no apparent difference. Differentiation gives voters motive to participate (Marsh and Fawcett 2011).

Different product attributes can have low or high market impact on the selection process (Ramdas et al. 2010). For example, among cars, outer body panels that shape a car’s appearance are more likely than the characteristics of its axles to influence someone’s decision. Political campaigns try to mitigate losses by not having positions that voters disagree with, which often appear synonymous with a favorable policy. For example, politicians say they are for access to health care and support the national health insurance policy as if these are the same policy. However, those impacted by such policies are more likely know the difference between health insurance and universal healthcare policies salient to their needs. For instance, those with health insurance in the U.S. are more likely to seek care than those without (Hoffman and Paradise 2008). Involved voters have more practice discerning differences among political campaigns.

Preferences among presented political options

Political scientists have studied voter decision making through reviews of previous elections and analyzing preferences in surveys. For example, researchers posited that voters decide among political choices through valuation of fixed (e.g., political party identity) and fluid variables (e.g., candidate attributes) (Stegmaier et al. 2019). Across democratic nations, a combination of factors influences voters like the state of the national economy and assessment of the party in charge. Furthermore, political scientists modeled how closely a candidate shared voters’ position on a political issue to the likely performance of a candidate if elected (Rabinowitz et al. 1982). The closer a voter felt their values on an issue aligned with those of a politician, the more positively they evaluated them. Absent from these studies, however, are how the perceived alikeness of ballot options influences a voter’s decision to participate.

Furthermore, previous research has studied preferred candidates based on the presented information. Shafir (1993) created two fictitious candidates accompanied by candidate information. For example, a generic candidate with generally positive personal traits (e.g., two children enrolled in local elementary school and studied history in college) was preferred to a candidate with strong positive and negative past behaviors (e.g., honorable council vice president and refusal to disclose tax return) (Shafir 1993). Meanwhile, researchers in Finland evaluated endorsement from family/close friends, social media network, or a voting guide (Voting Advice Applications) on candidate favorability (Christensen et al. 2021). Family/close friends and voting guide endorsements increased candidate favorability. These studies have demonstrated voter decision making as multifaceted and often irrational (Brennan and Buchanan 1984).

Voter participation and U.S. media malaise

Voter interest in politics precedes voter participation. Youth in households with parents that discussed politics associated with greater political interest and participation (Dostie-Goulet 2009). With aging, political interest and political participation generally increases (Glenn and Grimes 1968; Goerres 2007). Voter participation remained relatively constant for voters forty years old and above (Glenn and Grimes 1968). Voter awareness of how politics affects someone’s personal interests increases participation. For example, awareness of casualties from war increased voter turnout across twenty-three democracies over a fifty-year study period (Koch and Nicholson 2016). Political interest involves an investment of time and energy to the political process. The conceptual model is depicted in Fig. 1.

Fig. 1

Conceptual model

H1

Political organizational avocational interest increases likelihood to vote.

While democratic participation is instilled in citizens from school, society, and family, increased political knowledge can distort perceptions of government (Matsusaka 1995). According to media malaise theory, media coverage tends to sensationalize and focus on negative events (e.g., conflict, political theater, and scandals) that foster public distrust in government (Avery 2009; Curran et al. 2014; Newton 1999). While government is widely considered important, political participation is influenced by the amount of political knowledge and type of media consumed (Kenski and Stroud 2006; Newton 1999). Sensationalized media depicting peril implies that dire conditions are insurmountable (Gerbner et al. 1986; Moeller and de Vreese 2013; Putnam 2000). This media malaise instills that attitude that government can offer little to no change (de Beus 2009).

Political information gained from media consumption can have both a mobilization and dissuading effects on the electoral participation (Curran et al. 2014; Newton 1999). Dramatized media coverage of conflict and strategic calculation over political issues contributes to distrust of government and a spiral of cynicism (Capella and Jamieson 1997). Meanwhile, informative public television [e.g., Canadian Broadcasting Corporation (CBC)] increases efficacy, cognition, and turnout compared to commercial television (Aarts and Semetko 2003; Curran et al. 2014). Therefore, the content and quality of political information coverage can influence public sentiment about politics. Viewership; however, is not exclusively commercial or public. Advertisements, debate news, social media posts, and political coverage during campaign seasons are ubiquitous and directed toward target audiences (Freedman et al. 2004; Wells et al. 2016). Furthermore, U.S. voters primarily view commercial (versus public) media, highlighting conflicts and political theater (Capella and Jamieson 1997; Curran et al. 2014). This behavior ripens the U.S. electorate to perceive elected officials as alike. Perceived alikeness is the degree a voter views two options as no different. For example, many voters believe there is relatively no difference between the Democratic and Republican parties in their ability to change government processes (Hibbing 2001). This attitude is often verbalized as ‘Washington D.C. is broken’ in terms of politicians’ inability to pass meaningful legislation that notably improves the lives of the average citizen. When voters are unable to distinguish options, they are less likely make a preferential decision.

One would imagine the more advertising voters see; the more different campaigns appear. In contrast, the more aware and interested a voter is in politics, the more alike parties, candidates, and issues will appear because campaigns are talking about the same issues with nuanced differences. For example, common core and standardized testing are both within the context of education. These policies both attempt to improve the system. While stark differences exist, voters can perceive them as alike despite fundamentally different policy approaches because of regular negative media coverage. Hence, greater perceived alikeness is hypothesized to decrease civic participation. We contend that in a media malaise environment; more political interest increases exposure to this cynical political coverage and options appear more alike.

H2

Political organizational avocational interest increases perceived alikeness.

In a political environment of media malaise in the U.S., political campaigns have limited media airtime and potential voter attention spans to differentiate themselves. For example, individuals may read a campaign slogan in an ad, watch a video clip of a speech, listen to a commentator, and thereafter have their preferences solidify. Many voters prefer options that share their values and will positively improve their living conditions (Wildavsky 1987). Wildavsky (1987) called this preferred voting option “efficacious in supporting” someone’s beliefs. In fact, options with slight differences compared to an authentic original are more likely to elicit post-selection regret and decision reversals (Svenson 1992). For instance, an elected politician who accomplishes incremental change could leave voters disaffected and believing the previous officeholder did more for them. Many voters are likely to view them as having little difference in a media malaise environment that influences public opinion. In the absence of perceived unique offerings, we propose that voters are less likely to vote because they lack sufficient reasoning to justify their decision.

H3

Perceived political alikeness decreases likelihood to vote.

Cynicism in politics can be measured in the degree someone believes their vote will not have an impact on government (Craig and Maggiotto 1982). Individual action (e.g., voting) and a response (e.g., beneficial policy) generate a sense that government is functioning on someone’s behalf (Coleman and Davis 1976). Among young adults, lack of government responsiveness is associated with disengagement (Farthing 2010). Voters are motivated to participate when they feel a civic duty, identification with a public problem, and that voting matters. We define the belief that voting is ineffectual as a voter’s perception that their vote will make no difference in an election.

Media malaise has contributed to voter pessimism and less support for democratic institutions (Newton 2012, p. 4). Public trust in U.S. government declined from 1964 (76%) to 1996 (32%) (Earl Bennett et al. 1999). Researchers attributed the war in Vietnam, Watergate, societal disorder, economic woes, and dissatisfaction with government policies as some of the reasons for this decline (Craig 1993; Nye et al. 1997, p. 264). Notably, it is media outlets that deliver this information to citizens. The choice of media to focus on negative news, influences viewers to think of government inaction and maleficence. This decision promotes the belief that voting is ineffective because regardless of which elected official is in office, there will be little difference. Neither option will produce obvious gains.

H4

Belief that voting is ineffectual will moderate the relationship between political parties and perceived alikeness. Specifically, higher belief that voting is ineffectual will exacerbate perceived alikeness.

Overview of studies

The purpose of conducting the three studies herein was to validate through predictive modeling a measure of perceived alikeness in three political contexts. We modeled the effects of political avocational interest and political alikeness on LTV. The measure of perceived alikeness presented two political choices with a rating option from 0 alike to 100% alike. Study 1 evaluated the perceived alikeness of political parties. Study 2 evaluated the perceived alikeness of generic political candidates. Study 3 evaluated the perceived alikeness of political issues. Identical predictive analysis was conducted across the three studies except for replacing the context of political alikeness. Analyzing three datasets assessed for replicability among different samples. The large samples with different participants followed standard predictive modeling practices (Preacher and Kelley 2011; Williams and MacKinnon 2008). The three studies tested each of the hypotheses across political contexts.

Method

Study 1

A national sample of 412 registered voters participated in the online questionnaire study during the summer of 2020. Nine participants did not complete the questionnaire in its entirety and eight failed attention checks (N = 395). After agreeing to consent, participants completed dependent and independent variables. There was random presentation of individual difference measures (i.e., conservatism, education, independence) and random presentation of items within measures. The participates then completed demographic questions. History of voting (recent voter), party affiliation, and religiosity were collected last to avoid influencing prior responses (Unsworth and Fielding 2014).

Females represented 56% of the sample. The average age was 43 years old. Party affiliation was 41% Democrat, 32% Republican, and 27% independent/other. Eighty-nine percent of participants identified as heterosexual and 75% as Caucasian. The average household income was between $50,000 and $59,000. Participants read/watched approximately 6.22 hours of news per week. Sixty percent of participants were classified as recent voters (i.e., voted in the last three elections eligible by voting age).

Qualitative responses indicated that participants read/watched news from major new channels (e.g., Fox News, CNN, MSNBC, NPR, PBS, ABC, CBS, BBC, Al Jazeera, etc.) and online sources (e.g., Facebook, The Guardian, Yahoo.com, AOL, Apple News, New York Times, USA TODAY, etc.).

Measures

Political Organizational Avocational Interest (PO). Voter PO was measured using the seven-items on the original continuous scale (e.g., “Signed a petition,” “Attended a rally or demonstration,” etc.) (alpha = 0.70) (Goldberg 2010). All variables were mean centered in the analysis (Hayes and Rockwood 2020).

Mediation variable

Political alikeness (parties). At the core of branding is differentiation. This study utilized a 0% to 100% alikeness continuous sliding scale (“Please rank the degree the two entities are the same”). 0% represented completely different and 100% represented exactly the same. The major U.S. political parties were randomly alternated in presentation from left and right from one participant to the next (e.g., Democrat: Republican; Republican: Democrat).

Moderation variable

Belief Voting Ineffective. Belief that voting is ineffectual in elections evinced high reliability (alpha = 0.839) with three items on a continuous scale from 1—Strongly disagree to 7—Strongly agree: “Regardless of who is elected into government, it makes no difference,” “I have no influence on the outcome of government elections,” and “My vote will not change the results of an election.”

Dependent variable

Likelihood to vote (LTV). The outcome variable, LTV, was measured using a continuous scale item (1–Strongly disagree to 7–Strongly agree), “I vote in government elections.”

Control variables

The study controlled for gender, age, household income, ethnicity (Caucasian), political party identification (Republican/Democrat/Independent/other), news (hours read/watch news per week), sexual orientation (heterosexual/other), and religiosity (two items, continuous scale 1—Strongly disagree to 7—Strongly agree) (e.g., “I currently consider myself to be a member of a religious or spiritual organization”). Political party identification and sexual orientation were dummy coded. Independent/other and heterosexual participants were classified as the reference groups. The inclusion of control variables accounted for potential influences on the model.

Recent voting. Participants who indicated that they had voted in the last three general elections (i.e., 2018, 2016, and 2014; because of the timing of data collection) were classified as recent voters. Based on age, young participants who had voted in each general election they were eligible for, were classified as recent voters. For example, a 21-year-old participant who voted during the last general election was considered a recent voter.

Conservatism was measured using the 10-item IPIP representation of Goldberg’s (1992) MPQ traditionalism (e.g., “Tend to vote for conservative political candidates”) (alpha = 0.83) (Goldberg 1992; Goldberg et al. 2006; Johnson 2014). Education was measured using the 7-item IPIP representation of Goldberg’s (1992) markers for education (e.g., “Have a rich vocabulary”) (alpha = 0.79). Independence was measured using the 10-item IPIP representation of Goldberg’s (1992) markers for independence (e.g., “Don't care what others think”) (alpha = 0.69). There was random presentation of measures to participants. Within measures, there was also random presentation of questions. Reverse coding was performed on negative items following procedures of the original measures.

Results

Averages, standard deviations, and correlations are depicted in Table 1. Moderated mediation analysis utilized IBM SPSS Statistics 23.0 PROCESS (Version 3.5, Model 7) (Hayes 2012). First the researchers conducted predictive analysis on the main effect (H1) of political avocational interest (PO) on LTV. Second, we assessed indirect effects of PO on perceived alikeness (H2) and perceived alikeness on LTV (H3). Third, we assessed belief voting is ineffectual on PO on perceived alikeness (H4). The results are presented in this order.

Table 1

Study 1: Descriptive statistics and cross-level correlations

Variables	M	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17
Constructs
1. LTV	5.73	1.55	–
2. Alikeness (parties)	31.97	31.85	− 0.19**	–
3. Voting ineffectual	3.64	1.61	− 0.30**	0.35**	–
4. PO	4.11	1.45	0.25**	0.28**	− 0.02
Covariates						–
5. Conservatism	4.36	1.01	0.04	− 0.03	− 0.03	0.01	–
6. Education	5.13	0.95	0.34**	− 0.25**	− 0.24**	0.02	− 0.03	–
7. Independence	4.02	0.83	0.01	− 0.07	0.04	− 0.19**	− 0.16**	0.16**	–
8. Gender (female)	1.56	0.50	0.05	− 0.28**	− 0.24**	− 0.12*	− 0.01	0.08	− 0.04	–
9. Age	42.61	13.79	0.27**	− 0.27**	− 0.15**	− 0.07	0.09	0.16**	0.14**	0.11*	–
10. Household income	6.40	2.94	0.09	− 0.09	− 0.10*	0.05	0.03	0.17**	− 0.05	− 0.01	− 0.05	–
11. Ethnicity (Caucasian)	0.75	0.43	0.17**	− 0.17**	− 0.12*	− 0.06	− 0.08	0.16**	0.11*	0.06	0.27**	0.01	–
12 Recent voter	0.60	0.49	0.44**	− 0.05	− 0.14**	0.19**	0.04	0.22**	0.04	− 0.05	0.30**	0.11*	0.16**	–
13. Democrat	0.41	0.49	0.10	0.02.02	− 0.13**	0.18**	− 0.43**	0.09	− 0.04	0.08	− 0.− 0.12*	− 0.01	0.00	0.07	–
14. Republican	0.32	0.47	0.04	0.00	0.02	− 0.01	0.42**	− 0.05	− 0.09	− 0.01	0.15**	0.07	0.09	0.08	− 0.58**	–
15. News week (h)	6.22	2.34	0.32**	− 0.25**	− 0.19**	0.12*	0.01	0.35**	0.10	0.05	0.31**	0.09	0.07	0.25**	0.13*	− 0.07	–
16. Heterosexual	0.89	0.32	− 0.09	− 0.08	− 0.09	− 0.11*	0.21**	− 0.03	0.03	0.06	0.10*	0.05	0.04	− 0.06	− 0.08	0.11*	− 0.09	–
17. Religiosity	4.37	1.88	0.10	0.19**	0.01	0.25**	0.53**	0.09	− 0.16**	− 0.03	0.02	0.02	− 0.12*	0.08	− 0.14**	0.20**	0.03	0.07	–

N = 395, *p < 0.05; **p < 0.01 level (2 tailed)

The predicted direct effects of PO and perceived alikeness on LTV were supported (see Table 3). Holding all control variables constant, higher PO was related to higher LTV, t(384) = 3.858, SE = 0.056, p < 0.005 (H1). The average percentage of perceived alikeness of political parties was overall dissimilar (Mparties = 31.97%, SD = 31.85) (see Table 2). Higher PO related to higher perceived alikeness, t(378) = 4.205, SE = 1.128, p < 0.0001 (H2) (see Fig. 2 and Table 3). Higher perceived alikeness associated with lower LTV, t(379) = − 2.235, SE = 0.003, p < 0.05 (H3). Thus, the effect of PO on LTV was partially mediated by perceived alikeness. Alikeness had a total negative mediating indirect effect on PO on LTV, supporting the hypothesis that perceived alikeness of parties decreases participation in voting.

Table 3

Study 1: test of moderated mediation of political organizational avocational interest and likelihood to vote

Perceived alikeness (political parties as a mediator)
Antecedent	Outcome
	Alikeness (political parties)				Likelihood to vote (LTV)
	Coeff	SE	t	p	Coeff	SE	t	p
Political organizational avocational interest (PO)	4.742	1.128	4.205	< 0.0001	0.217	0.056	3.858	< 0.001
Alikeness (political parties)	–	–	–	–	− 0.007	0.003	− 2.235	< 0.05
Voting ineffectual	4.515	0.975	4.629	< 0.0001	–	–	–	–
PO × voting ineffectual	2.268	0.577	3.929	< 0.001	–	–	–	–
Gender (female)	− 8.669	2.851	− 3.041	< 0.05	0.044	0.146	0.299	0.765
Age	− 0.255	0.114	− 2.235	< 0.05	0.011	0.006	1.952	0.052
Household income	− 0.804	0.479	− 1.680	0.094	0.004	0.027	0.156	0.877
Ethnicity (Caucasian)	− 3.836	3.385	− 1.133	0.258	0.195	0.174	1.12	0.262
Recent voter	1.446	3.062	0.472	0.637	0.893	0.163	5.49	< 0.0001
Democrat	− 1.878	3.537	− 0.531	0.596	0.178	0.191	0.935	0.351
Republican	1.144	3.676	0.311	0.756	0.124	0.184	0.673	0.501
News per week (h)	− 1.940	0.708	− 2.742	< 0.05	0.053	0.038	1.416	0.158
Sexual orientation (heterosexual)	0.193	3.639	0.053	0.958	− 0.340	0.170	− 2.008	< 0.05
Religiosity	3.226	1.039	3.106	< 0.05	0.018	0.045	0.411	0.682
Conservatism	− 4.715	1.965	− 2.400	< 0.05	0.054	0.081	0.668	0.505
Education	− 3.720	1.657	− 2.245	< 0.05	0.288	0.100	2.878	< 0.05
Independence	1.846	1.829	1.010	0.314	− 0.025	0.084	− 0.298	0.766
Model summary	R2 = 0.383				R2 = 0.341
	F(16, 378) = 19.053, p < 0.0001				F(15, 379) = 13.087, p < 0.0001

Variables were mean centered. Democrat and Republican variables were dummy coded (reference group: independent/other). Sexual orientation (heterosexual) variable was dummy coded (reference group: all other identities)

Table 2

Means and standard deviations

	Study 1 (parties)		Study 2 (candidates)		Study 3 (issues)
	N = 395		N = 375		N = 400
	Mean	SD	Mean	SD	Mean	SD
PO	4.11	1.45	3.83	1.37	4.12	1.38
Belief voting ineffectual	3.64	1.61	3.50	1.53	3.61	1.48
Perceived alikeness (%)	31.97	31.85	45.89	24.45	51.83	18.68
LTV	5.73	1.55	5.74	1.57	5.62	1.54

Fig. 2

Modeled political organizational avocational interest effects on likelihood to vote for the three studies

Estimated PROCESS moderated mediation analysis calculated 95% confidence intervals using bias-corrected and accelerated bootstrap with 10,000 resamples (Hayes 2017). The omnibus test of moderated mediation [effect = − 0.015, SE = 0.008 (LLCI − 0.033 ULCI − 0.002)] indicated a significant moderating conditional effect of belief that voting is ineffectual with the indirect path of PO to perceived alikeness (H4). Conditional indirect effects are presented in Fig. 3 and Table 4. As predicted, high belief that voting is ineffectual moderated the effect of PO on perceived alikeness (interaction effect: − 0.056, LLCI − 0.106 to ULCI − 0.010). Belief that someone’s vote does not influence government exacerbated perceived alikeness and thereby decreased LTV. Meanwhile, low belief that voting is ineffectual did not moderate the effect of PO on perceived alikeness (interaction effect: − 0.007, LLCI − 0.030 to ULCI 0.015). Belief that someone’s vote does affect government did not significantly influence perceived alikeness. Contextually, the results support the conditional indirect effects of the perceived alikeness of political parties.

Fig. 3

Moderation effects of belief voting ineffectual and political organizational avocational interest on perceived alikeness and likelihood to vote for the three studies

Table 4

Conditional indirect effects with moderator (belief voting ineffectual)

		Left-leaningLow (− 1 SD)	AverageMean	Right-leaningHigh (+ 1 SD)
Study 1	Political organizational → perceived alikeness (parties) → likelihood to vote	− 0.007 (LLCI − 0.030 ULCI .015)	− 0.032 (LLCI − 0.061 ULCI − 0.006)	− 0.056 (LLCI − 0.106 ULCI − 0.010)
Study 2	Political organizational → perceived alikeness (candidates) → likelihood to vote	− 0.018 (LLCI − 0.046 ULCI .002)	− 0.036 (LLCI − 0.069 ULCI − 0.010)	− 0.054 (LLCI − 0.101 ULCI − 0.016)
Study 3	Political organizational → perceived alikeness (issues) → likelihood to vote	− 0.004 (LLCI − 0.028 ULCI .017)	− 0.025 (LLCI − 0.054 ULCI − 0.006)	− 0.047 (LLCI − 0.091 ULCI − 0.014)

Bootstrap 10,000 resampled confidence intervals

SPSS AMOS 21 maximum likelihood estimates demonstrated adequate model fit, χ2(2, N = 395) = 21.606, p < 0.001, TLI = 0.914, CFI = 0.983 (Jackson et al. 2009; Markland 2007).

Discussion

As predicted, PO was related to increased LTV. The perceived alikeness of parties had a total negative indirect effect on LTV in the model. High levels of belief that voting is ineffectual exacerbated the indirect effect. Each of the hypotheses were supported. However, while conditional indirect effects results were found among the perceived alikeness of parties, could these results be replicated and found in a different political marketing context? Study 2 attempted answer this question by examining the perceived alikeness of statements made by fictitious candidates.

Study 2

The purpose of study 2 was to replicate findings with a different sampled population and in a different political marketing context. Study 2 replicated the procedures from study 1. A national sample of 396 registered voters participated in the online questionnaire study. Twelve participants did not complete the questionnaire in its entirety and nine failed attention checks (N = 375). Females represented 57% of the sample. The average age was 43 years old. Party affiliation was 38% Democrat, 30% Republican, and 32% independent/other. Eighty-nine percent of participants identified as heterosexual and 75% as Caucasian. The average household income was between $50,000 and $59,000. Participants read/watched approximately 6.13 hours of news per week. Fifty-seven percent of participants were classified as recent voters (i.e., voted in the last three elections for which they were eligible by voting age). All variables were measured using the same scales and measures as in study 1, except for the mediation variable (perceived alikeness of candidates). Study 2 included political statements made by fictious candidates to measure perceived alikeness in a different context.

Political alikeness (candidates). Political alikeness of candidates’ positions was measured on a 0% to 100% continuous sliding scale (“Please rank the degree the two candidates' positions are the same”). Candidates’ statements were randomly presented first or second. Then candidates were randomly alternated in presentation left and right from one participant to the next to rate on the slide scale (e.g., Candidate A: Candidate B; Candidate B: Candidate A).

Statement set one (term limits) (103 characters each):Statement set two (universal preschool) (97 characters each):

Candidate A states there should be term limits on the length of time elected officials serve in office.

Candidate B believes term limits are in place and voters should decide if they reelect representatives.

Candidate X explains universal preschool should be publicly funded and available to all children.

Candidate Y asserts early childcare is available to all families with children and at a low cost.

Each pair of candidate statements was uniform in character numbers to control for potential differences in length. The pair of statements were also on the same topics (i.e., term limits and universal preschool) (Cronbach alpha = 0.579).

The results of study 2 replicated the moderated meditation analysis results found in study 1 in the context of candidates (see Fig. 2). The main effect was supported. Holding all control variables constant, regression analysis supported PO had a direct positive effect on LTV, t(359) = 3.724, SE = 0.060, p < 0.0001 (H1).

The average percentage of perceived alikeness of political candidates was slightly more dissimilar overall (Mcandidates = 45.89%, SD = 24.45) (see Table 2). Higher PO related to higher perceived alikeness, t(358) = 4.217, SE = 1.016, p < 0.0001 (H2). Meanwhile, higher perceived alikeness associated with lower LTV, t(359) = − 2.871, SE = 0.003, p < 0.05 (H3). Thus, the effect of PO on LTV was partially mediated by perceived alikeness. Alikeness had a total negative mediating indirect effect on PO with LTV. This provides evidence that the perceived sameness of candidates is associated with a lack of voter participation.

Estimated PROCESS moderated mediation analysis calculated 95% confidence intervals using bias-corrected and accelerated bootstrap with 10,000 resamples (Hayes 2017). The omnibus test of moderated mediation (effect = − 0.012, SE = 0.006 (LLCI − 0.025 ULCI − 0.002) indicated a significant moderating conditional effect of belief that voting is ineffectual with the indirect path of PO to perceived alikeness. Conditional indirect effects are presented in Table 4 and depicted in Fig. 3. As predicted, high belief that voting is ineffectual moderated the effect of PO on perceived alikeness (interaction effect: − 0.054, LLCI − 0.101 to ULCI − 0.016). Belief that someone’s vote does not influence government exacerbated perceived alikeness and thereby decreased LTV (H4).

Meanwhile, low belief that voting is ineffectual did not moderate the effect of PO on perceived alikeness (interaction effect: − 0.018, LLCI − 0.046 to ULCI 0.002). Belief that someone’s vote does affect government did not significantly influence perceived alikeness. The results support the finding of conditional indirect effects on the perceived alikeness of political candidates, extending the political marketing context from study 1.

SPSS AMOS 21 maximum likelihood estimates demonstrated adequate model fit, χ2(2, N = 375) = 17.996, p < 0.001, TLI = 0.923, CFI = 0.985 (Jackson et al. 2009; Markland 2007).

Study 2 replicated the results found in study 1 in the context of the perceived alikeness of political candidates and with a different sampled population. The perceived alikeness of candidates had an indirect negative effect on LTV that was exacerbated by high levels of the believe voting is ineffectual. The results support the importance of differentiation in political marketing campaigns to give voters substantial reason(s) to cast a ballot. However, are these findings transferable to political issues that are more amorphous and cross over party lines? Study 3 attempted to answer this question by examining the perceived alikeness of political issues.

Study 3

The purpose of study 3 was to increase generalizability by replicating the findings of the previous studies with a third sampled population and in the context of political issues. Study 3 replicated the procedures from studies 1 and 2. A national sample of 416 registered voters participated in the online questionnaire study. Ten participants did not complete the questionnaire in its entirety and six failed attention checks (N = 400). Females represented 64% of the sample. The average age was of 41 years old. Party affiliation was 39% Democrat, 31% Republican, and 30% independent/other. Eighty-three percent of participants identified as heterosexual and 77% as Caucasian. The average household income was between $50,000 and $59,000. Participants read/watched approximately 6.14 hours of news per week. All variables were measured using the same scales and measures as in studies one and two, except for the mediation variable (perceived alikeness of issues).

Political alikeness (issues). Political alikeness of issues was measured on a 0% to 100% continuous sliding scale (“Please rank the degree the two items are the same”). Seven paired issues representative of domestic policy (healthcare: health insurance), economic (income inequality: minimum wage; trade policy: tariffs), immigration (sanctuary cities: immigration assimilation), education (standardized testing: common core; student debt: college reform), and national security (foreign aid: federal assistance programs) formulated the inventory (Cronbach alpha = 0.793). Presentation of issues was randomly alternated from left and right from one participant to the next.

The results replicated significant findings from study 1 and 2 in the context of political issues (see Fig. 2). The main direct effect was supported. Holding all control variables constant, higher PO related to higher LTV, t(379) = 3.291, SE = 0.051, p < 0.05 (H1).

The average percentage of perceived alikeness of political issues was slightly more similar overall (Missues = 51.83%, SD = 18.68) (see Table 2). Higher PO related to higher perceived alikeness, t(383) = 3.275, SE = 0.712, p < 0.001 (H2). Meanwhile, higher perceived alikeness associated with lower LTV, t(384) = − 3.007, SE = 0.004, p < 0.05 (H3). Thus, the effect of PO on LTV was partially mediated by perceived alikeness. Alikeness had a total negative mediating indirect effect on the effect of PO on LTV, supporting the hypothesis that the perceived sameness of issues contributed to decreased voter participation.

Estimated PROCESS moderated mediation analysis calculated 95% confidence intervals using bias corrected and accelerated bootstrap with 10,000 resamples (Hayes 2017). The omnibus test of moderated mediation [effect = − 0.015, SE = 0.007 (LLCI − 0.030 ULCI − 0.004)] indicated a significant moderating conditional effect of belief that voting is ineffectual on the indirect path of PO to perceived alikeness. Conditional indirect effects are presented in Table 4 and depicted in Fig. 3. As predicted, high belief that voting is ineffectual moderated the effect of PO on perceived alikeness (interaction effect: − 0.047, LLCI − 0.091 to ULCI − 0.014). Belief that someone’s vote does not influence the government exacerbated perceived alikeness and thereby decreased LTV (H4).

Meanwhile, low belief that voting is ineffectual did not moderate the effect of PO on perceived alikeness (interaction effect: − 0.004, LLCI − 0.028 to ULCI 0.017). Belief that someone’s vote does effect government did not significantly influence perceived alikeness. The results support the hypothesis of conditional indirect effects on the perceived alikeness of political issues, extending political the marketing context from studies 1 and 2.

SPSS AMOS 21 maximum likelihood estimates demonstrated adequate model fit, χ2(3, N = 400) = 50.293, p < 0.001, TLI = 0.867, CFI = 0.960 (Jackson et al. 2009; Markland 2007).

Study 3 replicated the results from studies 1 and 2 in the context of political issues. The results provided evidence that perceived alikeness exists among abstract political issues. The findings reiterate the importance of differentiating political marketing campaigns. Within the context of differentiation and consolidation theory in a media malaise environment, differentiation precedes LTV. Perceived alikeness had a negative indirect effect on LTV.

General discussion

Within the context of differentiation and consolidation theory in a U.S. media malaise environment, differentiation helps voters through the decision-making process. With differentiation, individuals generate preferences that help them gain confidence in their decision (Meyer 2018). Without differentiation, existing cynicism in politics negatively influences voters’ perceptions (O’Keefe 1980). Across three political contexts, the same direction of significant results was replicated. PO increased LTV. Mediation findings of perceived alikeness reoccurred with different values of the independent variable, in different contexts, and with different groups of individuals following recommended procedural standards (Preacher et al. 2007). Perceived alikeness decreased LTV. This indirect effect was exacerbated by a high belief that voting is ineffectual. However, those with a low belief that voting is ineffectual did not exhibit differences in perceived alikeness and thereby their LTV. Indeed, voters who believe their votes can change government will likely vote regardless of perceived alikeness.

Differentiation and consolidation theory postulates individuals seek differences among choices to decide (Svenson 1992). Svenson (1992) explains a sufficient degree of differentiation becomes the reasons individuals use to justify behaviors. Individuals calculate post-decision outcomes when processing information, like friends asking why they voted for a particular candidate. The political alikeness scale quantifies the perceived uniformity of choices (i.e., greater perceived alikeness, the less different choices appear). The scale provides a suitable tool to study differentiation and consolidation among political choices. With annual elections, campaigns can measure real candidates for practical applications. For instance, political campaigns can present two candidates running for office, measure perceived alikeness, and adjust strategies to energize voters with substantive reasons. By quantifying perceived differentiation, it is possible for researchers to compare choices and a potential threshold to generate a favorable vote.

Moreover, differentiation and consolidation theory conceptualized how individuals can process and justify decisions. The alikeness scale provides researchers a method to investigate the types of individuals that perceive options as alike. For example, some individuals use campaign slogans like “Build the wall” (Trump 2016 campaign) as sufficient reason to consolidate or justify their choice (Ryman et al. 2017). Meanwhile, some individuals painstakingly remain undecided until they enter the voting booth (Arcuri et al. 2008). Which types of people notice subtle differences? Which types of people act on subtle differences? Svenson (1992) discussed after decisions are made, individuals are less likely to reverse their decision because attractiveness increases. This suggests first-time voters, not set in their voting patterns, could be swayed easier because a campaign slogan appears substantive. Comprehending individual differences can improve target marketing and empower campaigns with sharpened messaging.

Implications

Findings from this project reinforce the importance of political differentiation because of the saturation of campaign marketing and limited attention span of voters (Aarts and Semetko 2003). Many voters tune out opposing sides or do not pay attention to politics until it is near the election to inform their decision to cast a ballot. If a voter views a handful of ads, reads a couple of news articles, and watches part of a debate, they have little information with which to form a comprehensive understanding of issues. Succinct and lucid arguments are necessary to reach voters in the oversaturated digital media age (Gainous and Wagner 2013). Given the complexities of government politics, differentiation helps voters make decisions to discern information about options. The results in this paper show that brand differentiation could bypass media malaise. Differentiation can emphasize beneficial values to voters and give voters a reason to cast a ballot.

Furthermore, greater political interest relating to higher perceived alikeness indicated that involved voters are more skeptical of ballot options. With cyclical political campaigns, poor policy implementations, and negative marketing spin, it becomes more difficult for voters to decide (Aarts and Semetko 2003; Denton Jr et al. 2019; Towner and Dulio 2012). The distance from votes cast, policies enacted, and benefits to citizens requires effective political marketing (Lock and Harris 1996). The media’s cynicism, general disillusionment in government, and opposing marketing campaigns, are underlying campaign problems. For example, ‘swift boating’ was used to disparage 2004 presidential candidate John Kerry’s Vietnam War (three Purple Hearts) military service background (Daniel 2010). Unanswered attack ads from The Swift Boat Veterans for Truth group nullified Kerry’s positive military experience. Anticipation of negative advertisement with a rapid response plan is an integral part of contemporary campaign strategies (Reyes 2006). Effective campaigns need to defend their position, offer substantive solutions, and define opposition to win votes.

Limitations

A potential issue for the generalization of the results is LTV versus actual votes. LTV does not always translate to actual votes. Often polls will include a plus/minus ( ±) margin of error to account for surveying error, sampling bias, and other spurious potential issues at that time of the poll. Furthermore, poll respondents and likely voters in a sampled population are not equivalent. A likely voter may be more inclined to vote for any given number of reasons, than a poll respondent who answered a phone call. Hence, this project controlled for history of voting with recency of voting in the last three general elections in moderation mediation PROCESS analysis. Recency voting across the three studies demonstrated a strong direct positive relationship with LTV (study 1: t(384) = 5.49, SE = 0.163, p < 0.0001; study 2: t(359) = 6.649, SE = 0.170, < 0.0001; study 3: t(379) = 7.448, SE = 0.156, p < 0.0001). These results supported intent to vote as a precursor of voting.

Furthermore, the U.S. is an imperfect democratic nation that differs from other political systems. For example, the U.S. government is primarily controlled by two major political parties (i.e., Republicans and Democrats). Meanwhile, other democracies elect multiple parties into office with legislative power. For example, in the U.K. and Germany, a coalition government is usually formed by multiple parties to form a governing majority. With more viable choices, parties must distinguish themselves to stand out to voters. This effort can result in prioritizing key issues and legislative victories for constituents, unlike gridlock in the U.S. Furthermore, while a coalition government has the appearance of a minority and majority voting bloc, it is perceived to be less fixed than the U.S. two-party system. For example, a single political party may not receive complete blame for government failures during a legislative session. Therefore, cross-cultural analysis of perceived alikeness among choices could provide insight for other nations and their political participation.

Future research

Modeled perceived alikeness on LTV results highlighted the importance of campaigns to emphasize substantive differences. Despite different philosophies and policy positions, many voters tend to see the major U.S. political parties as very similar (Hibbing and Theiss-Morse 2002, p. 33). An underlying possible reason is the general perception of government inaction. Hibbing and Theiss-Morse (2002) found U.S. political parties perceived similar in their inability to change government processes to pass meaningful policies. Recent successfully elected U.S. presidents campaigned on changing the system (e.g., former President Obama “Change We Can Believe In” and former President Trump “Drain the swamp”) (Hudson 2018; Milkis et al. 2012). In study 1, higher belief that voting is ineffectual related to higher perceived alikeness of political parties, t(378) = 4.629, SE = 0.975, p < 0.0001 (Table 3). Expansion of the alikeness scale from two candidates to include specific political issues could provide greater insight. For example, participants can evaluate two candidates on perceived ability to change how government operates. This adds a second dimension of measurable differentiation. This can quantify different reasons individuals use to consolidate or justify their decision.

Moreover, the complex model of results can be expanded to include other additional variables and relationships. For example, perceived impact on the political system is another precursor to voting. Blondel et al. (1997) found lower proportionality of voters to elected representatives increased voter participation. The U.S. is a representative democracy with an electoral college system that modifies proportionality based on geography (Grofman and Feld 2005; Pattie and Johnston 2014). Official selection of government representatives is based on votes cast but ultimately decided by an indirect system (Doorenspleet and Pellikaan 2013). For example, during the presidential elections from 2000 to 2016, the popular vote did not decide two of the last five presidential race winners (Federal Election Commission (FEC) 2018). Each presidential race is decided by the electoral college system that favors rural states with smaller overall populations, so they have substantial representation in government (Pattie and Johnston 2014). For example, a vote from California has disproportionally less value in the electoral college system than a vote cast in Iowa on the presidential race. This disproportionality can dissuade voters because voting can appear to have less influence on electoral outcomes. Future studies can analyze perceived alikeness relative to geographical locations on LTV. It becomes more vital to emphasize differentiation because belief that voting is ineffectual could be greater in larger states such as California. This disparity can impact ballot issues and candidates affect by media malaise.

In the U.S., financial contributions impact a campaign’s ability to reach voters with marketing materials. Compared to presidential races, down-ballot candidates and issues often receive substantially less campaign contributions (Federal Election Commission (FEC) 2017). More resourceful marketing campaigns have highlighted popular voter demands (Butler and Collins 1994). Given the foreseeable future of unlimited campaign contributions, future studies can investigate clear and concise differentiation campaigns compared to those that focus on political rhetoric. For example, grassroots campaigns have focused on specific policy changes such as increasing the federal minimum wage in contrast to soaring rhetoric highlighting a candidate’s small-town beginnings. Future research can compare generic candidates and speeches for perceived alikeness based on campaign strategies.

Politics is increasingly online and on social media. The 0% to 100% alikeness scale provides an additional tool for researchers to study online political engagement. In practice, campaigns could present different ads and evaluate whether voters gained distinctive reasons to vote accordingly. Politicians can adjust strategies based on geographical region, voter party registration in the district, and feedback from voters to influence outcomes. Campaigns can use this as a method to evaluate whether becoming more alike or dissimilar can alter preferences in the race. Do leading candidates benefit by highlighting differences? Do trailing candidates benefit by highlighting similarities to siphon votes? Social media ads can target specific types of voters. For example, one marketing strategy can focus on turning out passionate supporters while another can seek to persuade undecided voters (Barbu 2014; Chester and Montgomery 2017; Fulgoni et al. 2016). By understanding behaviors on social media, we can better understand voters’ process of differentiation and consolidation on political options. We propose mapping the journey voters typically take on social media to reach a decision.

The turbulent U.S. 2020 election during the COVID-19 pandemic with socially distant rallies and canceled debates shook historical precedence (Youde 2021). This election also had the highest voter turnout in the U.S. history (U.S. Census Bureau 2021). Post-COVID-19 pandemic campaigns will need to find new ways to inspire voters to go to the polls. With unprecedented global changes and economic strife, campaigns have new issues to differentiate themselves in an elevated fear-based news environment (Sulkowski 2020; Verma and Gustafsson 2020). For example, with the cynicism of corporate rule commonplace, public safety and workers’ rights can become energizing priorities. Citizens benefit from direct assistance and social programs to recover from the pandemic (Garner et al. 2020). Effective political marketing can distinguish campaigns and navigate these challenges.