Marquette University Marquette University

e-Publications@Marquette e-Publications@Marquette

Master's Theses (2009 -)

Dissertations, Theses, and Professional

Projects

Exploring Political Consumerism: Its Antecedents and Mediating Exploring Political Consumerism: Its Antecedents and Mediating

Role Between Family Communication and Political Activity Role Between Family Communication and Political Activity

Hannah Swarm

Marquette University

EXPLORING POLITICAL CONSUMERISM: ITS ANTECEDENTS AND

MEDIATING ROLE BETWEEN FAMILY COMMUNICATION AND

POLITICAL ACTIVITY

Milwaukee, Wisconsin

May 2022

ABSTRACT

EXPLORING POLITICAL CONSUMERISM: ITS ANTECEDENTS AND

MEDIATING ROLE BETWEEN FAMILY COMMUNICATION AND

POLITICAL ACTIVITY

Marquette University, 2022

socializing agents for children, especially when it comes to political development.

Conducting a national survey of 523 U.S. adults, the present study explored how one’s

family communication environment during childhood affects their likelihood of engaging

in political consumerism during adulthood, and in turn, how that influences their

involvement in offline and online political activity.

Through multiple regression analysis, the results showed that both the

conversation and conformity orientations are positively associated with boycotting and

buycotting. Path analysis further revealed the mediating role of political consumption on

the relationship between family communication and political activity. Overall, this study

ACKNOWLEDGMENTS

I would like to express my sincere gratitude to my thesis chair, Dr. Young Kim,

for his assistance throughout every stage of the thesis. Thank you for the continuous

support you have provided during my project, and during my time at Marquette. You are

extremely dedicated to helping your students succeed, and I am grateful to have worked

support throughout my graduate studies. Thank you for always cheering me on and

believing in me. I would also like to thank my graduate school cohort for being a great

support system. To the TAs, I am lucky to have worked with such an amazing group of

people, and I can’t wait to see all of the wonderful things you do after your time at

Marquette. To Professor Sturgal, my supervising faculty member, you are incredible.

Thank you for welcoming me to the COMM 1100 team and for your unwavering support

over the past two years. I have learned so much from you.

TABLE OF CONTENTS

ACKNOWLEDGMENTS………………………………………………………….

LIST OF TABLES………………………………………………………………… iv

LIST OF FIGURES………………………………………………………………...

CHAPTER

B. Types of Political Consumers: Boycotters and Buycotters……………. 10

C. Antecedents to Political Consumerism………………………………... 13

D. Family Communication Patterns as a Key Antecedent? ……………… 18

E. Family Communication Patterns and Political Consumerism………… 22

F. The Importance of Family Communication Patterns for Political

G. Political Consumerism: A Catalyst to Other Modes of Political

A. Participants……………………………………………………………. 39

B. Procedure……………………………………………………………….

C. Measurements…………………………………………………………. 44

iii. Political Activity…………………………………………………. 48

iv. Political Trust…………………………………………………….

v. Control Variables………………………………………………... 51

D. Analysis………………………………………………………………... 53

IV. RESULTS……………………………………………………………………... 54

B. Encouraging Offline and Online Political Activity through Family

C. Political Consumers are Politically Active Citizens…………………... 75

D. Political Consumption: A Mediator between Family Communication

E. Additional Antecedents to Political Consumerism………………….… 82

F. Implications…………………………………………………………… 85

G. Limitations and Future Directions……………………………………. 89

LIST OF TABLES

Table 1. Sample Descriptive Statistics (N = 523)…………………………………….

40

Table 2. Means and Standard Deviations of Family Communication Items (CVO)

and (CFO)…………………………………………………………………………….

44

Table 3. Political Consumerism (CONSM) Items…………………………………….

47

Table 6-2. OLS regression analyses for the relationship between family

communication and political activity………………………………………………….

59

Table 7. OLS regression analyses for the relationship between political consumerism

and political activity………………………………………………………………….

60

Table 8. OLS regression analyses for the relationship between all variables and

political activity……………………………………………………………………….

62

LIST OF FIGURES

Figure 1. Theoretical model of independent variables, mediators, and dependent

variables…………………………………………………………………………….

Figure 2. Bootstrapping (N = 5,000) results for path diagram of family

communication, political consumerism, and political activity……………………. 66

INTRODUCTION

According to the 2020 American National Election Studies (ANES) Time-Series

Study, the majority of respondents (57.7%) claimed to have either bought or declined to

purchase a certain product or service for political or social reasons. Of those who

deliberately purchased or avoided products from certain companies, 22.6% reported

contacting an elected state or local official (17.5%), signing a petition (28.4%), or posting

political comments online (38.7%) (ANES, 2021). This data suggests that as the avenues

of political participation undergo continuous expansion, Americans are more regularly

adopting political consumerism as a viable means of political action.

Although traditional modes of political activity, like voting, remain important

al., 2005). According to Gotlieb and Cheema (2017), the growing prevalence of political

consumerism indicates a larger societal shift in younger citizens’ values, norms, and

practices. Those who engage in political consumerism deliberately choose certain

products and producers based on social, political, or ethical considerations (Shah et al.,

2007). This mode of political activity provides an alternative way of expressing

dissatisfaction, especially since companies vigilantly monitor consumer behavior and

Dalton (2008) points out, more traditional means of participation can be associated with

duty-based citizenship, which has a narrower view of participation and discourages

citizens from engaging in challenging activities that give them more of a direct say in

politics. Therefore, the decline in duty-based citizenship has made room for impactful

non-electoral modes of political action that afford citizens greater control over the focus

values and target various political actors aside from parties, candidates, and the

government (e.g., corporations, brands, international organizations) (Bennett, 2012;

Copeland, 2014a; Stolle et al., 2005, p. 251). As the nature of citizenship has evolved,

citizens have adopted more direct means of political action, such as political

consumerism, to influence policy makers. Rather than an unpromising decline in civic

that participation in boycotts has grown steadily over the decades. In 1999, boycotting

was more than four times as likely to occur compared to 1974 (Stolle et al., 2005). More

recent studies examining political consumption in the U.S. and Europe have found that

the percentage of citizens who engage in at least one form of political consumerism

ranges from 28% to nearly 50%, which confirms political consumption’s status as a

widely adopted method of political activity (Baek, 2010; Newman & Bartels, 2011;

Strømsnes, 2009). Political consumerism has gained public acceptance in other regions of

the world, such as Latin America, where between 15% and 29% of citizens elect political

consumption as a way of influencing the political system (Echegaray, 2015). Citizen

involvement in political consumerism varies globally, with it being particularly

pronounced in places like Scandinavia (Stolle et al., 2005). Nonetheless, political

consumerism has gained global recognition as a valuable means of political activity.

consumerism (Copeland, 2014a). Specifically, Micheletti (2003) attributes citizens’

increasing involvement in political consumerism to the rising dependence on the global

marketplace for the delivery of goods and services. Whether people recognize it or not,

shopping has become a survival mechanism given that nowadays few people produce

their own food and clothing. Citizens have also become increasingly concerned about the

more flexible, network-oriented, hands-on and that allows them to combine their daily

lives with political causes” (Micheletti, 2003, p. 24-24). Particularly in the Western

world, increased wealth has given some people the opportunity to consider additional

facets of a product aside from price and quality (e.g., environmental impact) (Micheletti,

2003). The growth of political consumerism has also been facilitated by the emergence

and pervasiveness of the Internet, e-commerce, and social media (Gil de Zúñiga et al.,

2014a; Kelm & Dohle, 2018).

Clearly, myriad forces have contributed to the growth of political consumerism as

a practical and appealing means of political action. Political consumption can be driven

by various reasons, whether issues of sustainability, worker’s rights, political party

affiliation, or social justice be at the forefront of an individual’s purchasing decisions.

However, for the purposes of this study, political consumerism is broadly defined as a

Types of Political Consumers: Boycotters and Buycotters

There are two main forms of political consumerism: boycotting and buycotting.

Both forms of consumerism are viewed as means of “lifestyle politics,” wherein everyday

life provides opportunities to make political statements (Copeland & Boulianne, 2020, p.

4). Individuals who boycott deliberately avoid purchasing products or brands because

associated with messages to “shop local” or with campaigns supporting companies that

provide a living wage, use sustainable resources, or adhere to fair trade standards (Kam &

Deichert, 2020). While boycotting aims to punish a company by damaging their

reputation, good will, or bottom lines, buycotting intends to reward a company for its

positive behavior (Copeland, 2014a). Previous scholars have found that different

motivational characteristics prompt people to boycott and buycott. Boycotting is

considered a form of collective action; networks of boycotters are often larger and highly

interconnected (Friedman, 1999; Jungblut & Johnen, 2021). Due to the collective nature

of boycotts and the need for organized efforts, offline and online communication

activities have been found to influence boycotts more than buycotts (Kelm & Dohle,

2018). Copeland (2014a) suggested that boycotting is more strongly associated with

dutiful citizenship norms, which emphasize obligation, loyalty to the state, and deference

Buycotting, on the other hand, shares more similarities with civic engagement

since it entails more informal and cooperative activity that takes place outside of

traditional political institutions. It involves the enactment of engaged citizenship norms,

which emphasize voluntary activity, forming independent opinions, and aiding others

(Copeland, 2014a). Since buycotting is more individualistic, self-motivated, and driven

Neilson (2010) to argue that more altruistic people are motivated to engage in the

complex process of buycotting because it is satisfying and fulfills their inherent desire to

benefit the well-being of others. This may also explain why buycotters tend to target

smaller or independent businesses, while boycotters often focus on targeting market

leaders and attracting media attention to promote broader industry change (Neilson,

2010).

Political consumers may engage in both boycotting and buycotting behaviors,

which scholars have referred to as ‘dualcotting.’ These individuals have been found to

possess distinct characteristics as well, which may explain their motivations for engaging

in both means of political consumerism. For example, Baek (2010) found that dualcotters

demonstrate higher levels of political knowledge and they follow politics more closely.

buycotters and suggested that dualcotters may be intensified versions of buycotters.

Likewise, Copeland (2014a) posited that buycotters, who already choose to reward

companies through purchasing goods, will likely consider boycotting goods or brands

with incongruent values. Due to this, dualcotters will likely resemble buycotters more

than boycotters in regard to citizenship norms.

This may be explained by negativity bias, or the stronger propensity to perceive and

process negative stimuli compared to positive stimuli (Jungblut & Johnen, 2021).

Consistent with this idea, Kam and Deichert (2020) found that negative information

provokes boycotting far more powerfully than positive information induces buycotting.

The effectiveness of boycotts may also be attributed to them often containing overt

conflict between organized groups and corporations, which allows them to gain more

traction via the mass media (Copeland, 2014a). Despite these distinctions, it is critical to

remember that boycotting and buycotting involve more complex decisions. The decision

to deliberately avoid a product or brand is more feasible if there is an available alternative

of similar price and quality, and the decision to intentionally buy a certain product or

brand is much easier when there is a minimal gap between comparable products in terms

of price and quality (Kam & Deichert, 2020). Therefore, the product market plays a key

Previous research has sought to define the political consumer through examining

political behaviors and socio-demographic factors. Regarding political tendencies, prior

research has found that those who possess high levels of political interest (e.g.,

Strømsnes, 2009) and political knowledge (Baek, 2010; Endres & Panagopoulos, 2017)

are more likely to engage in political consumerism. The possession of postmaterialist

institutions (Stolle et al., 2005). This overall lack of trust stems from the fact that political

consumers may perceive political parties and government institutions as inept at

effectively addressing their political concerns, which motivates them to boycott and

buycott (Copeland, 2014b). Some scholars have found that boycotters exhibit lower

levels of political trust compared to buycotters (Copeland, 2014a; Neilson, 2010).

Explaining these results, Neilson (2010) posited that trust toward institutions inspires

positive behavior, and thus, encourages buycotting. Meanwhile, a lack of trust in

institutions elicits feelings of frustration and dissatisfaction linked to negative corporate

behavior, which encourages boycotting.

Generalized trust is another key variable to consider. For example, Neilson (2010)

found that more trusting individuals are more prone to buycott than boycott, perhaps

because buycotting is more covert and operates largely on the assumption that others are

regarding the current state of things in one’s life” (p. 814). Another positive predictor of

political consumerism, especially buycotting, is association involvement, or one’s

participation in various types of associations such as sports teams, voluntary

organizations, and social clubs (Neilson, 2010). In general, people involved in

organizations are more likely to be asked to participate in politics and follow through

Ideological intensity is also an important factor, with those who identify as very

liberal or very conservative being more likely to participate in boycotts and buycotts than

those who identify as moderate (Copeland, 2014b; Endres & Panagopoulos, 2017).

Comparing the two ideological camps, Hooghe and Goubin (2022) found that political

consumerism was practiced more by those with a left-wing political orientation.

Moreover, scholars have concluded that Democrats are more active political consumers

compared to Republicans (Endres & Panagopoulos, 2017; Strømsnes, 2009). However,

somewhat inconsistent with these findings, Baek (2010) concluded that buycotters were

more likely to identify as Republican and conservative while boycotters were more likely

to identify as Democrat and liberal. Additionally, Endres and Panagopoulos (2017) found

that partisans with more stable policy positions were more active in recent boycotts and

buycotts compared to partisans with less consistent positions. Although partisan

“partisan cues” (Copeland, 2014b, p. 276).

Considering the impact of socio-demographic factors, prior research has indicated

that women (e.g., Hooghe & Goubin, 2022; Stolle et al., 2005), younger individuals (e.g.,

Newman & Bartels, 2011), those who are well-educated (e.g., Hooghe & Goubin, 2022;

Strømsnes, 2009), and those with a high income (Micheletti & Stolle, 2005) are more

of political consumerism among women may further be explained by their higher

involvement in marketplace activities, as they often assume the traditional role of

shopping for the family (Micheletti, 2003; Neilson, 2010). Nevertheless, findings related

to socio-demographic factors have been inconsistent, especially with regard to gender and

income. For example, Copeland (2014a) found no clear gender or income differences

when dividing political consumers into boycotters and buycotters; yet, significant

differences emerged for dualcotters, with them being more likely to be female and earn

higher levels of income. Contrary to these findings, Neilson (2010) found significant

gender differences between boycotters and buycotters, revealing that women are 53%

more likely to buycott than men, while gender does not influence the probability of

boycotting.

Although political consumers possess some common characteristics, they are not

controversial issues for corporations, as history has proven, which makes it imperative for

companies to carefully consider how and why it manufactures products (Micheletti,

2003). Poor labor practices and values that are incompatible with consumers’ values can

elicit negative reactions that damage a company’s reputation and good will. Large-scale

boycotts and buycotts can impact corporate policy, company personnel, and public

policies.

Even though a consumer decision may seem unrelated to politics, political

identity can have subtle and nuanced effects on consumer behaviors (Jung & Mittal,

2019). Jung and Mittal (2019) describe political identity as a broad concept comprised of

“a wide array of attitudes, motives, logics, and interpretations of how the world works,

how the world should work, and how societies may approach various issues and

problems” (p. 69). Therefore, a variety of political issues can impact consumption

choices. For example, according to Gromet et al. (2013), environmental protection

through government-mandated sustainability efforts is a key issue on the liberal agenda,

which in turn, makes conservatives less likely to purchase an energy-efficient product

linked to environmental benefits. Although this is just one example, empirical research

has demonstrated that consumers evaluate products and brands more favorably if they are

issues, and communication with consumers (Jung & Mittal, 2019).

In general, it is crucial for companies to examine their customer base as they

navigate their political brand communication, especially given that boycotters’ punitive-

oriented actions can be particularly damaging to a brand’s success (Jungblut & Johnen,

2021). A broad, heterogeneous group of consumers possessing a variety of political views

Neilson’s (2010) findings, differences among boycotters’ and buycotters’ thought

processes can influence corporate messaging strategies. Given that boycotters tend to

have less trust in institutions, companies targeted by boycotters could recruit trustworthy

third parties to endorse their products. Conversely, since buycotters experience higher

trust in institutions, companies may preemptively craft messages that promote their social

responsibility efforts (Neilson, 2010).

interest in politics and produce a desire to engage in various forms of political

participation.

Consistent with these findings, Wicks and Warren (2014) found that parents’

political consumption can lead to greater political consumption among their teens. Shared

political consumption between parents and their teens can also enhance teen political

results regarding the effect of parental political discussions on youth political

consumption. These insignificant findings may be explained by the fact that dispositions

like political interest are still developing through childhood and adolescence (Neundorf et

al., 2013), meaning that the desire to boycott or buycott may not be fully present until

adulthood. Another reason for these findings may be that adolescents do not have the

motivated and maintained by repeated attempts from others to influence individuals’

consumption. Although these efforts take place in both online and offline communication

contexts, Kelm and Dohle (2018) found that face-to-face communication remains more

influential on political consumerism activities, perhaps because these interactions carry

greater social pressure or there is greater trust in the information shared during an in-

person discussion. Adding to this, Kam and Deichert (2020) asserted that social groups

can transform an individual’s expression of personal beliefs into an expression of

collective identity, as well as amplify one’s perception that their action, along with the

actions of others, will make an impact. In line with this, Lindén (2005) found that friends

and family members can impact one’s food consumption choices, such as the decision to

become a vegetarian, through inspiring or pressuring them to alter their private food

strategies (Neilson & Paxton, 2010).

certain businesses, which may simulate intentional purchasing tendencies among

children. This may be the case given that buycotting often involves informal learning,

meaning that people learn about products or brands that are consistent, or inconsistent,

with their values through their friends, family members, and other social networks

(Copeland, 2014a). Additionally, the collective nature of boycotting may motivate family

subsequently, their internal and external communication (Koerner & Fitzpatrick, 2002b,

p. 36). Political consumers view politics as extending beyond the political system to

include private attitudes, choices, and behaviors, and therefore, the family may be a

prime location for defining those values and attitudes that would prompt political

consumption.

Given that the family is a primary agent of political socialization, growing up in

an open communication environment may expose children to more political discussions

and encourage the development of policy positions, which in turn, increases one’s

likelihood of engaging in political consumerism during adulthood. Although previous

research addresses family-based discussions to some extent, it is important to conduct a

more comprehensive examination regarding the characteristics of one’s family

communication and how that relates to specific modes of political consumption. The

effects of family communication patterns may vary for each form of political

consumption. Boycotts challenge companies by punishing them for objectionable labor

policies or practices, making them more conflict-oriented (Copeland, 2014a; Friedman,

1999). By nature, boycotting is more likely to involve explicit conflict between organized

groups and corporations and gain greater attention from the mass media (Copeland,

children’s political knowledge and interest. Therefore, given the prominence of collective

consumer boycotts, children who grew up discussing politics may be more motivated to

take part in this organized form of political activity during adulthood, as they view it as a

viable means of influencing public policy. Buycotting, on the other hand, is less

pronounced and may be more of a private endeavor; one may engage in buycotting based

on their independent opinions formed through acquiring information and insights from a

variety of sources. Therefore, this study proposes the following hypotheses:

H1a: Family political communication during childhood will be positively

associated with boycotting during adulthood.

H1b: Family political communication during childhood will be negatively

associated with buycotting during adulthood.

influence of government and politics in everyday life, the overlap between private and

public spheres, and increased access to political information. Scholars have recognized

the expanding definition of political participation, which has led them to study distinct

forms of political engagement, such as online and offline participation (e.g., Jung et al.,

2011) or institutionalized and non-institutionalized participation (e.g., Hooghe & Marien,

required to perform various political activities (Jung et al., 2011). This distinction is also

important because online political participation can aid adolescents’ entry into the

political sphere and allow young adults to exercise greater influence in politics (Kim et

al., 2017).

It is important to consider the relationship between family communication and

adolescent political development, as the family is a key agent of political socialization.

Regarding political viewpoints, Ledbetter (2015) found that high conversation families

(i.e., pluralistic and consensual) are more successful at transmitting political beliefs

between parents and children. Research has also shown that family political discussions,

which may be more frequent or robust among families high on the conversation

orientation, can elevate a children’s likelihood of participating in a variety of political

activities. For example, Graham et al. (2020) found that laissez-faire families (i.e., low

(1986) suggested that growing up in a high conformity family hinders adolescents’

political development, while high conversation families are more likely to discuss politics

and cultivate greater political interest among adolescents. Liebes and Ribak (1992)

observed a similar trend within high conversation families, concluding that pluralistic

families (i.e., high conversation, low conformity) promote greater levels of political

among families high on one orientation and low on another orientation (i.e., pluralistic

and protective). The findings were surprising, given the assumption that those raised in a

harmonious and conflict-averse environment would tend to gravitate toward similar,

homogeneous networks (Hively & Eveland, 2009).

Parents can also facilitate the interaction between children and media messages by

affecting their political news exposure, influencing their consumption habits, and

engaging them in news-related political discussions (Austin & Pinkleton, 2001). For

example, parents can encourage youth news consumption habits and positively affect

their level of civic engagement through modeling news behaviors and engaging in family

discussions about news content (York & Scholl, 2015). Yet, it is essential to recognize

that adolescents are not “passively socialized” (Valenzuela et al., 2019, p. 1097).

Adolescents are active members of the socialization process, considering different

baseline levels of youth news consumption and political discussion with peers were

positively associated with future political discussion with parents. Therefore, it is

important to acknowledge that among the varying family communication environments,

families may engage in political discussions that are spurred primarily by parents,

primarily by children, or a mix of both. Nonetheless, as previous literature suggests, the

cannot be overlooked. As Valenzuela et al. (2012) asserted, these forms of political

participation “share a foundation but are empirically different” (p. 17). Although offline

political activity is still a key element of civic engagement, political expression via digital

channels may appeal to younger generations in particular (Kim et al., 2017), as well as

individuals of all ages due to its ability to decrease the cost (i.e., time and physical

inconvenience) needed to take political action (Jung et al., 2011). By requiring less time

and effort to participate, online activities offer citizens more control over their

engagement in politics (Jung et al., 2011). Another important difference is that online

political engagement and activity persists during and between elections, whereas

traditional methods of political participation are “punctuated and concentrated around

elections” (Feezell, 2016, p. 496). Yet, prior research has regarded participation as a

predominately offline occurrence, failing to consider how the Internet has created new

in participation (Vissers & Stolle, 2014).

The Internet and social media can enhance civic engagement, as it provides an

arena for people to express themselves and develop their own identities, as well as

connect with many social groups and individuals at once. Compared to in-person

conversations, online discussion networks can facilitate higher levels of political

political activities. As people consume content and interact with others online, they may

develop their “political self” and engage in political expression, which can cultivate

participation habits that lead to further political action (Gil de Zúñiga et al., 2014b, p.

627; Jung et al., 2011). In line with this research, Kim and Chen (2016) found that the use

of blogs and social network sites were positively related to online political participation.

Valenzuela et al. (2012) also found that discussing politics with individuals outside of the

more tight-knit group of friends and family results in a higher degree of online political

engagement. Ultimately, research points to the idea that external influences outside of the

family, especially political information and discussion networks accessed online, can

prompt more online political activity.

Meanwhile, past research has indicated a positive relationship between family

communication and more traditional means of political participation. For example, Liebes

consensual families. In their study, political participation was measured by asking

respondents about their involvement in more conventional forms of political activity,

such as voting, working for a political party, and displaying a political campaign button,

sticker, or sign. In general, offline political participation entails more tradition-based

political action, which may be promoted through discussions among family members,

H2a: Family political communication during childhood will be positively

associated with offline political activity during adulthood.

H2b: Family political communication during childhood will be negatively

associated with online political activity during adulthood.

Although it may seem that political consumers are detached from political

processes, research has shown that politically active individuals are more likely to engage

in political consumerism compared to those who are less politically active (Echegaray,

2015; Stolle et al., 2005; Strømsnes, 2009). Those who engage in political consumerism

possess the same types of characteristics that have traditionally been attributed to

politically active individuals, which include being well-educated, having a left political

orientation, and displaying high levels of political interest (Endres & Panagopoulos,

approach to politics, political consumerism can be regarded as an addition to

conventional forms of political participation (Echegaray, 2015; Strømsnes, 2009). It is an

individualistic form of political participation involving high initiative (Newman &

Bartels, 2011; Strømsnes, 2009).

According to Ward and de Vreese (2011), consumer and citizen arenas have

purchasing power to enact social change, has a positive effect on both online and offline

civic participation, but no significant effect on offline political participation. The positive

association between online and more civic forms of participation may be explained by the

shift away from traditional means to citizenship, such as voting, working for a political

party, or sending letters to government officials (Ward & de Vreese, 2011). These results

are consistent with findings from Stolle et al. (2005), who found that political

consumerism is practiced together with non-traditional and individualistic forms of

political action, but it is not practiced consistently with conventional political acts. These

findings suggest that political consumers are more engaged in forms of political activity

that share common traits with political consumption.

Contrary to these findings, Gotlieb and Wells (2012) found that citizens who

made intentional purchasing decisions based on social or ethical considerations were

these results, Gotlieb and Wells (2012) posited that political consumerism can serve as a

gateway into larger collective political activity for younger individuals who have less

experience in civic engagement. Another possible explanation for why political

consumption is often tied to other forms of political activity comes from Stolle et al.

(2005), who asserted that citizens who actively practice political consumerism believe

effectively advocate for desired change.

Political consumers show great political interest, and they are willing to take part

in a time-consuming process that involves learning about company practices and

deliberately seeking out alternative products or brands that align with their values

(Copeland, 2014b). Given the individualized and self-motived nature of political

consumerism, individuals who use their purchasing power to send political messages may

be inclined to engage in other forms of political participation. As Copeland (2014a)

Table 1. Sample Descriptive Statistics (N = 523)

Table 1 (continued).

M-Turk), Qualtrics has been found to be the most demographically and politically

representative in the U.S. (Boas et al., 2018). Furthermore, Qualtrics’ large participant

family communication, political consumerism, political activity, and political trust. Other

questions regarding demographic information (e.g., education, gender, and income),

partisan identity, and political ideology were included in the questionnaire. The

participants received $4.51 as compensation for participating in the study.

Two pre-tests of the survey were conducted with 100 participants (N = 100) to

ensure that the survey was working properly and contained clear instructions and

questions. One pre-test was administered via Amazon’s Mechanical Turk (M-Turk), an

online platform that collects data quickly and provides researchers with respondents who

are demographically more diverse than traditional student samples. Several studies have

shown that M-Turk samples are reasonably congruent with other samples and responses

ease, the scale ranged from 1 (“Hard to answer”) to 5 (“Easy to answer”). Finally,

pre-test confirmed that the questions were clear (M = 4.44, SD = 0.79), understandable

(M = 4.32, SD = 0.90), easy to answer (M = 4.44, SD = 0.71), and relevant (M = 4.42,

SD = 0.73). The same questions were asked in the survey administered via Qualtrics. The

results of the Qualtrics pre-test showed that the questions were clear (M = 4.10, SD =

1.15), understandable (M = 3.68, SD = 1.35), easy to answer (M = 3.65, SD = 1.24), and

relevant (M = 3.71, SD = 1.28).

The MTurk pre-test was also used to determine the median amount of time it took

to complete the survey, which was 10 minutes. Based on the median duration of the

survey in the pre-test, responses that were completed in under five minutes and greater

than 30 minutes were automatically eliminated by the Qualtrics system in the main test.

At the beginning of the survey, participants were asked if they intended to

provide their best answer for each question; if they did not agree to provide their best

answers, Qualtrics screened out the participant, preventing them from taking the rest of

the survey. To further ensure data quality, an attention-check question was placed in the

middle of the survey (i.e., “This is to check your attention for this survey. Please check

‘strongly disagree’ (1)”). All participants provided the correct answer for the attention

To measure family communication patterns within the home, this study employed

the Revised Family Communication Pattern (RFCP) scale (Graham et al., 2020; Koerner

the family shape family relationship schemas, which in turn, determines how one

perceives their familial relationships and how one behaves in the family context. The

had the last word”; Cronbach’s α = .902, M = 3.34, SD = 0.93) (See Table 2).

Respondents were instructed to indicate their level of agreement with the statements on a

5-point scale ranging from 1 (“Strongly disagree”) to 5 (“Strongly agree”).

Table 2. Means and Standard Deviations of Family Communication Items (CVO) and

(CFO)

(SD)

Conversation Orientation (CVO)

CVO1: In our family we often talked about topics like politics and religion

2.67

Table 2 (continued).

CVO2: My parents often said something like "Every member of the family

should have some say in family decisions."

2.57

CVO3: My parents often asked my opinion when the family is talking

about something.

2.65

CVO4: My parents encouraged me to challenge their ideas and beliefs.

2.54

CVO5: My parents often said something like "You should always look at

both sides of an issue."

3.07

CVO9: My parents and I often had long, relaxed conversations about

nothing in particular.

2.89

CVO10: I really enjoyed talking with my parents, even when we disagreed.

3.11

CVO11: My parents liked to hear my opinions, even when they didn't

agree with me.

2.95

CVO12: My parents encouraged me to express my feelings.

3.01

CVO15: In our family we often talked about our plans and hopes for the

future.

3.07

Conformity Orientation (CFO)

CFO1: My parents often said something like “You’ll know better when

you grow up.

3.53

CFO2: My parents often said something like "My ideas are right and you

should not question them."

2.80

Table 2 (continued).

CFO3: My parents often said something like "A child should not argue

with adults."

3.28

CFO4: My parents often said something like "You should give in on

arguments rather than risk making people mad."

2.51

CFO5: My parents often said something like "There are some things that

just shouldn't be talked about."

3.20

CFO6: When anything really important was involved, my parents expected

3.71

CFO9: My parents sometimes became irritated with my views if they were

different from theirs.

3.11

CFO10: If my parents didn't approve of it, they didn't want to know about

it.

3.05

CFO11: When I was at home, I was expected to obey my parents' rules.

4.05

the product is made, or because you dislike the conduct of the company that produces it”;

Cronbach’s α = .795, M = 1.88, SD =1.56) and buycotting (e.g., “Bought a certain

product or service because you like the social or political values of the company that

produces or provides it”; Cronbach’s α = .841, M = 1.35, SD = 1.30) (See Table 3). The

measure of political consumerism was a dichotomous variable coded “1” if the

respondent had participated in at least one form of political consumerism (i.e., boycotting

or buycotting), and “0” if they had never participated in either form (Newman & Bartels,

2011). Rather than measure political consumerism as a single variable, it was divided into

boycotting and buycotting. All the items for boycotting were combined, and the

frequency of engaging in boycotting was indicated from “0” (never participated in any

all forms of buycotting).

Table 3. Political Consumerism (CONSM) Items

%

company that produces it.

CONSM3: Participated in a boycott (certain products or stores). Y – 37.7%

CONSM4: Avoided buying something in order to register a protest or

send a message.

CONSM5: Engaged in "buycotting," that is buying a certain product

or service because you like the social or political values of the

Table 3 (continued).

CONSM7: Bought a certain product or service because you like the

social or political values of the company that produces or provides it.

Adopting Jung et al.’s (2011) scale, political activity was divided into two

categories for measurement: offline and online political activity. Respondents were

instructed to recall their political activity during the 2020 election year. Offline political

participation was measured with 12 items, including whether respondents “voted in the

2020 election,” “spoke with public officials in person,” or “worked for a political party or

candidate” (Cronbach’s α = .854, M = 3.45, SD = 3.01). Online political activity was

measured by six items, which asked participants if, during the 2020 election cycle, they

updated to reflect today’s digital environment. For example, respondents answered

indicated from “0” (never participated in any form of online political activity) to “6”

(participated in all forms of online political activity).

were all recoded as “0.” An “other parties” variable was also created by recoding

independent and other party as “1” and Democrat and Republican as “0.” Together, all

independent (conformity and conversation dimensions of family communication) and

control variables in the model accounted for a significant portion of the variance in

boycotting behaviors, R

= 0.13, F(9, 513) = 8.26, p < 0.001 and buycotting behaviors, R

associated buycotting during adulthood. As expected, family political discussions during

childhood appeared as a strong predictor of boycotting. However, family communication

also positively predicted buycotting, which was not anticipated. The results indicated that

one unit change in the conversation orientation results in an increase of 0.20 in

boycotting behaviors (b = 0.20, t = 3.35) and an increase of 0.25 in buycotting behaviors

buycotting behaviors (b = 0.28, t = 4.66), controlling for effects of other independent

variables in the model (See Table 6-1).

The two orientations of family communication and the control variables in the

model also accounted for a significant portion of the variance in offline political activity,

= 12.35, p < 0.001. H2a proposed that family political communication during childhood

will be positively associated with offline political activity during adulthood, and H2b

posited that family political communication during childhood will be negatively

associated with online political activity during adulthood. H1a was supported, as family

political discussions during childhood appeared as a strong predictor of offline political

activity. However, contrary to expectations in H2b, family political communication

appeared to positively predict online political activity, as well. The results indicated that

results further showed that one unit change in the conformity orientation results in an

increase of 0.67 in offline political activity (b = 0.67, t = 4.95) and an increase of 0.40 in

online political activity (b = 0.40, t = 4.82), controlling for effects of other independent

variables in the model (See Table 6-2).

Furthermore, both forms of political consumerism and control variables in the

political activity, but negatively associated with online political activity. Meanwhile, H3b

proposed that buycotting will be positively associated with online political activity, but

negatively associated with offline political activity. However, the analysis revealed that

both forms of political consumption were strong predictors of both offline and online

political activity. The results showed that one unit change in boycotting behaviors results

in an increase of 0.52 in offline political activity (b = 0.52, t = 5.21) and an increase of

0.51 in online political activity (b = 0.51, t = 4.52), controlling for effects of other

independent variables in the model. The results further showed that one unit change in

buycotting behaviors results in an increase of 0.32 in offline political activity (b = 0.32, t

= 5.25) and an increase of 0.43 in online political activity (b = 0.43, t = 5.76), controlling

for effects of other independent variables in the model (See Table 7).

Regarding RQ2, those with lower levels of political trust were more likely to

–2.48). Income was a positive predictor for offline political activity (b = 0.13, t = 2.18),

while gender was negatively associated with offline political activity (b = –0.55, t = –

2.34). Lastly, education was positively associated with online political activity (b = 0.10,

t = 1.98), controlling for the effects of other independent variables in the model (See

Tables 6-1 and 6-2).

Table 6-1. OLS regression analyses for the relationship between family communication

and political consumerism.

Note. ***p < .001, **p < .01, *p < .05. Independent variables were not in violation of

Political Trust 0.18 0.97 –0.08 –0.65

N 513 513

0.25 0.18

F 16.13*** 12.35***

Note. ***p < .001, **p < .01, *p < .05. Results for offline and online political activity

were based on White’s heteroskedastic robust standard errors because the Breusch-

Pagan/Cook-Weisberg test revealed that there was heteroskedasticity (OFA: χ

2

(1) =

25.54, p < 0.05, ONA: χ

Table 7. OLS regression analyses for the relationship between political consumerism and

political activity.

F 24.82*** 25.64***

Note. ***p < .001, **p < .01, *p < .05. Results for offline and online political activity

were based on White’s heteroskedastic robust standard errors because the Breusch-

Pagan/Cook-Weisberg test revealed that there was heteroskedasticity (OFA: χ

2

(1) =

74.08, p < 0.001, ONA: χ

2

(1) = 76.16, p < 0.001). Independent variables were not in

violation of multicollinearity (i.e., VIF of each variable < 10 and Tolerance of each

variable > 0.10).

Path Analysis: Testing the Mediating Role of Political Consumerism

simple bivariate correlations to estimate the strength of the causal paths between

variables in the model (Hair et al., 2010; Kline, 2011). To determine which control

factors to include in the path diagram, an additional two multiple regression analyses

were run that included all variables and analyzed offline political activity and online

political activity as the dependent variables. The tests of VIF and tolerance indicated no

violation of multicollinearity in all independent variables. However, the Breusch-

Pagan/Cook-Weisberg test revealed that there was heteroskedasticity as the fitted values

of offline political activity, χ

2

(1) = 67.62, p < 0.001, and online political activity, χ

2

(1) =

61.85, p < 0.001, were smaller than 0.05. Due to this, White’s heteroskedastic robust

24.95, p < 0.001 and online political activity, R

The control factors of education, income, political ideology, and partisan identity (i.e.,

Republican or Democrat) did not have a significant association with offline political

activity. However, the control factors of gender, other political parties (i.e., not

Republican or Democrat), and political trust were significantly associated with offline

2.96). Lastly, one unit change in political trust results in an increase of 0.47 in offline

political activity (b = 0.47, t = 2.58), controlling for the effects of other independent

variables in the model. Regarding online political activity, none of the control variables

were significant. Therefore, gender, other parties, and political trust were the control

variables included in the path diagram, with pathways between them and offline political

activity (See Table 8).

Table 8. OLS regression analyses for the relationship between all variables and political

activity.

Buycotting 0.36** 3.23 0.36*** 4.81

Education 0.06 0.77 0.06 1.46

Income 0.10 1.68 0.03 0.89

Political Ideology –0.23 –1.72 –0.07 –0.89

Female –0.46* –2.10 –0.19 –1.31

Republican –0.12 –0.34 –0.20 –0.98

Other Parties –0.80** –2.96 –0.28 –1.70

Political Trust 0.47* 2.58 0.12 0.98

Constant –1.76 –1.73 –1.23* –2.10

N 513 513

Pagan/Cook-Weisberg test revealed that there was heteroskedasticity (OFA: χ

2

(1) =

67.62, p < 0.001, ONA: χ

2

(1) = 61.85, p < 0.001). Independent variables were not in

violation of multicollinearity (i.e., VIF of each variable < 10 and Tolerance of each

variable > 0.10).

To validate the mediation effects of political consumerism, this study employed a

bootstrapping procedure (N = 5,000), which is a computer-based technique of resampling

that allows for the estimation of standard errors and confidence intervals (Kline, 2011).

Preacher and Hayes (2008) recommend bootstrapping to examine mediation effects, as it

“provides the most powerful and reasonable method of obtaining confidence limits for

(2010) was followed to determine whether the model was considered to have acceptable

fit. Specifically, the desired values for the different fit indices were based on the sample

size (N ≥ 250) and the number of indicators (m ≥ 30). The model achieved good fit if the

χ

2

statistic was significant, as well as CFI ≥ 0.90, SRMR ≤ 0.08 with CFI ≥ 0.92, and

RMSEA ≤ 0.07 with CFI ≥ 0.90. To evaluate the mediation effects of political

criteria set forth by Hair et al. (2010), the final path analysis achieved an acceptable

model fit, χ

SRMR = 0.034.

Path analysis confirmed the results of multiple regression analyses for hypothesis

testing. Specifically, the path analysis revealed the positive relationship between the two

associated with both boycotting: β = 0.13, p < 0.01, 95% CI [.041, .215] and buycotting:

β = 0.21, p < 0.001, 95% CI [.130, .293]. Similarly, the conformity orientation was

statistically and positively associated with boycotting: β = 0.20, p < 0.001, 95% CI [.116,

.286] and buycotting: β = 0.23, p < 0.001, 95% CI [.145, .311]. The path analysis further

confirmed the positive relationship between the two orientations of family

communication and the two modes of political activity (H2a, H2b). The conversation

orientation was statistically and positively associated with both offline political activity: β

= 0.24, p < 0.01, 95% CI [.157, .304] and online political activity: β = 0.18, p < 0.01,

95% CI [.100, .250]. Similarly, the conformity orientation was statistically and positively

associated with both offline political activity: β = 0.13, p < 0.01, 95% CI [.054, .206] and

online political activity: β = 0.12, p < 0.01, 95% CI [.043, .184]. Finally, the path analysis

further confirmed the positive relationship between the two forms of political

were statistically and positively associated with offline political activity. Additionally,

boycotting: β = 0.26, p < 0.001, 95% CI [.166, .357] and buycotting: β = 0.25, p < 0.001,

95% CI [.152, .343] were statistically and positively associated with online political

activity (See Table 9).

Based on the regression results, three control variables – gender, other parties, and

negatively associated with offline political activity. Meanwhile, political trust was found

as a significant and positive factor associated with offline political activity: β = 0.07, p <

0.05, 95% CI [.013, .121] (See Table 9).

The present study also posed a research question (RQ1) about the mediating role

of political consumerism on the relationship between family communication and political

activity. The results showed significant indirect effects of the conversation orientation on

offline political activity, β = 0.03, p < 0.01, 95% CI [.042, .162], and online political

activity, β = 0.03, p < 0.01, 95% CI [.024, .101], through boycotting. There were also

significant indirect effects of the conformity orientation on offline political activity, β =

0.05, p < 0.001, 95% CI [.099, .253], and online political activity, β = 0.05, p < 0.001,

95% CI [.064, .169], through boycotting. Furthermore, the results showed significant

indirect effects of the conversation orientation on offline political activity, β = 0.03, p <

online political activity, β = 0.06, p < 0.001, 95% CI [.070, .177], through buycotting

(See Table 10).

While a full mediation occurs when the direct effect becomes nonsignificant in

the presence of the indirect effect, partial mediation is present when the direct effect is

reduced, but still significant (Hair et al., 2010). As the results for H2a and H2b indicated,

orientation) and political activity (offline and online political activity) (See Figure 2).

These results suggest that interactions within the family setting can motivate children to

engage in political consumption during adulthood, which subsequently makes them more

likely to take part in other types of political activity during adulthood.

political and public affairs, allowing children to foster the confidence and competence to

effectively participate in politics. The family communication orientations are broad

categories that help distinguish family communication contexts, but families do not fit

perfectly within the categories; they all communicate in a unique manner. High

conformity families may be overprotective and stifle conversation, which can lead to

Mangleburg, 2005; Ledbetter, 2015). This may explain the relationship between high

conformity families and political consumption, as this family communication schema has

been proven to effectively promote political information processing behaviors.

This study further theorized that family communication during childhood would

political activity during adulthood. Due to having more regular interactions and the

exposure to a variety of topics and viewpoints, children raised in this type of environment

may participate in discussions about political and social issues, which can help them form

an opinion and lead to subsequent participation in various political activities. Adding to

this from a cognitive processing standpoint, this type of family environment can teach

children how to process complex information with reduced levels of anxiety (Ledbetter &

Schrodt, 2008). Given today’s polarizing political landscape and the overwhelming

amounts of information available, political discussions can be challenging to participate

in and may induce anxiety for some, especially when trying to balance one’s own

political beliefs and interests with the desire to maintain friendly interpersonal

relationships (Pennington & Winfrey, 2021). Yet, having honest and frequent

conversations with other family members who may express disagreement can better

engage in various political activities.

The positive relationship between high conformity families and offline political

participation has also been backed by prior research. Children who must adhere to family

norms and expectations are accustomed to refraining from expressing controversial

opinions, and thus, may be able to consume political information from various news

sources or social groups to acquire political information and engage in political

conversations. Through this action, they may become more inclined to participate in

offline political activities, including voting, wearing campaign-related apparel, or

working for a political party or candidate. Another reason that high conformity families

may encourage greater levels of offline political participation is the idea that these family

structures can be overbearing for children who must obey their parents’ wishes. This can

create a hostile family environment and make adolescents more prone to engage in

deception or rebellion (Bristol & Mangleburg, 2005). If discussions about controversial

issues were consistently shut down, and children were expected to adhere to their parents’

belief system, they may want to branch off as adults and express their own viewpoints via

offline political activity.

The present study also found that childhood family communication was positively

some possible explanations for this finding. First, prior studies have examined and

documented the shift among younger generations from traditional forms of political

participation to less conventional methods. The integration of politics into digital

mediums can impact teen’s political engagement, as well as entire networks of friends

and acquaintances (Bode et al., 2014). Young people are taking advantage of the

interest and becoming less engaged in traditional party-based politics, online political

activity seems to be a feasible avenue for them to increase political engagement.

Therefore, it is important to recognize that if family communication inspires political

participation, that may take root predominantly online for younger generations. This

finding may also point to the appealing characteristics of online political activity,

including the ease of it and the ability to quickly reach a vast audience. By offering

greater convenience, online activities afford citizens greater control over how they choose

to engage in politics. Therefore, if high conversation and high conformity family

communication environments promote children’s involvement in politics as they get

older, these individuals may elect to participate in online forms of political activity

because they are less time consuming and can be completed from any location.

Another reason for this finding is that online political participation may be a

Zúñiga et al., 2010; Oser et al., 2013). Therefore, individuals do not necessarily

specialize in one form of political activity over the other. Although people may begin to

participate in new online political activities, they do not automatically serve as a

substitution for disengagement in offline political activity. The present study’s results

provide evidence for what Gil de Zúñiga et al. (2010) deemed “hybrid participation,”

effective path to engaged citizenship. As family communication can lead children to

engage in more traditional modes of political activity during adulthood, the expanding

definition of participation means that family communication can also cultivate online

political participation.

The results indicated that political consumers are generally more active in various

forms of political participation. This relationship is upheld by previous literature finding

that political consumers are more active in both conventional and unconventional

participation (Strømsnes, 2009) and online civic participation (Ward & de Vreese, 2011).