The impact of loneliness on compliance with COVID‐19 prevention guidelines

Abstract

Many individuals have been reluctant to follow the COVID‐19 prevention guidelines (e.g., wearing a mask, physical distancing, and vigilant handwashing) set forth by the U.S. Center for Disease Control to reduce the spread of COVID‐19. In this research, we use reciprocal altruism theory to investigate the role of loneliness and its impact on compliance with these guidelines. Our findings indicate that lonely individuals are less willing to comply with COVID‐19 prevention guidelines than non‐lonely individuals. Process evidence suggests that this occurs as loneliness can inhibit an individual's sense of obligation to reciprocate to others. However, we demonstrate that framing information about COVID‐19 through agentic (vs. communal) advertising messaging strategies can offset the negative impact of loneliness on compliance with COVID‐19 prevention guidelines. Thus, marketers and policymakers may want to consider the important role of loneliness when tailoring messaging appeals that encourage compliance with COVID‐19 prevention guidelines.

1. INTRODUCTION

In March of 2020, the World Health Organization declared COVID‐19 a pandemic and a global public health emergency. To slow the virus's spread, government health agencies like the US Centers for Disease Control (CDC) instituted prevention guidelines including urging individuals to physically distance themselves from people outside their immediate household, wash their hands frequently, and later, to wear face masks (CDC, 2021). Research shows that these guidelines have been highly successful in reducing COVID‐19's spread (Eikenberry et al., 2020; Howard et al., 2021; Lio et al., 2020; Moosa, 2020), especially in the presence of new, more contagious variants (Phillips, 2021), inequitable vaccine access (Rouw et al., 2021), and vaccine hesitancy (Machingaidze & Wiysonge, 2021; Pereira et al., 2022). As the pandemic continues, such guidelines remain crucial. Some individuals, however, tend not to follow (or choose to ignore) COVID‐19 prevention guidelines (Belanger & Leander, 2020; Wang, Marmo‐Roman, et al., 2021), sometimes even defying legal mandates (Taylor & Asmundson, 2021). Given the pandemic's severity thus far and the importance of individual behavior in minimizing its impact, it is vital to understand individual differences in compliance with COVID‐19 prevention guidelines. Accordingly, we seek to understand why certain individuals are less likely to comply with these guidelines, and how marketers and policymakers might consider these individuals in their public health messaging. In this research, we suggest that lonely individuals are among those less likely to comply with COVID‐19 prevention guidelines and examine a potential way to mitigate this through an advertising strategy.

An emerging stream of research has begun to explore how individual factors affect compliance with COVID‐19 prevention guidelines (Han, 2021; Kim et al., 2022; Oosterhoff & Palmer, 2020; Uddin et al., 2021; Xu & Cheng, 2021). For instance, some researchers have focused on how factors like an overall increase in vaccine awareness (Verma, 2021) and the use of mobile technology to track the virus (Verma & Mishra, 2020) can mitigate COVID‐19's impact. Other research has explored how various psychological factors can encourage individuals to adopt COVID‐19 prevention guidelines. For instance, Kim et al. (2022) found that hope and fear mediate the relationship between protection motivation and behavioral intentions toward COVID‐19 protection behaviors such as increased hygienic behaviors. In addition, Han (2021) found that extraversion, agreeableness, openness, conscientiousness, and neuroticism all correlated to compliance with one or more COVID‐19 prevention measures. Other research has found that higher levels of social responsibility (Oosterhoff & Palmer, 2020), risk aversion and need for control (Xu & Cheng, 2021), as well as greater trust in government (Uddin et al., 2021), increased compliant behaviors.

However, much remains unknown about how psychosocial factors like loneliness predict compliance with COVID‐19 guidelines. Humans are social creatures, heavily influenced by perceptions of their social environments (Ramya & Ali, 2016)—yet there remains a notable gap in this line of research regarding how psychosocial factors like loneliness impact compliance with COVID‐19 guidelines. The current research focuses on loneliness as the psychosocial factor for two reasons: first, loneliness was highly prevalent even before the pandemic, affecting 35% of adults aged 45 and older (Wilson & Moulton, 2010). Second, while physical distancing and quarantine efforts are effective in preventing the spread of illness (Moosa, 2020), these efforts may also amplify levels of loneliness, a phenomenon widely reported during the COVID‐19 pandemic (Groarke et al., 2020; Killgore et al., 2020; Li & Wang, 2020). Thus, marketers and policymakers need to understand the impact of loneliness on compliance with health‐related behaviors, why this impact occurs, and how to address any subsequent public health consequences.

Using reciprocal altruism theory as a guide, we suggest that investing in collective behaviors like mask‐wearing, washing hands vigilantly, and physically distancing from others, may be less worthwhile for lonely individuals (Roberts, 1998; Trivers, 1971; Van Vugt et al., 2007). Loneliness decreases one's capacity to experience the social benefits of altruistic behavior such as higher social standing and increased access to shared group resources (Miller, 2000; Zahavi, 1975). As such, we might expect lower compliance with COVID‐19 prevention guidelines among lonely individuals as opposed to non‐lonely individuals, because lonely individuals have a lower sense of obligation to reciprocate.

From a practical perspective, identifying and targeting high‐risk groups—like those who feel lonely—with marketing and communication strategies that encourage compliance is important to reducing the severity of COVID‐19. Research indicates that many COVID‐19 public health campaigns have failed to resonate with consumers due to ineffective and often inconsistent communication and advertising strategies (Overton et al., 2021; Robinson & Veresiu, 2021; Sauer et al., 2021; Segal, 2021). To create more effective marketing campaigns, experts are beginning to recommend that marketers use messaging appeals customized to specific consumers, communities, and cultures (Robinson & Veresiu, 2021). We respond to this recommendation by creating and testing messaging strategies that specifically focus on increasing compliance among lonely consumers.

This research makes several contributions to the literature. First, it connects the social affiliation literature with the literature on compliance with COVID‐19, and to our knowledge, is the first to identify a relationship between loneliness and consumers' compliance with COVID‐19 prevention guidelines. We also shed light on why lonely individuals respond less positively to COVID‐19 prevention guidelines by examining how loneliness can lead to a diminished sense of obligation to reciprocate. As such, our research may offer insights into well‐documented phenomena in the cooperation and social affiliation literature, specifically regarding who contributes to the collective and why. Second, this research contributes to reciprocal altruism theory (Trivers, 1971) by bringing it to a new area of inquiry. In doing so, we find that how we view ourselves socially influences our obligation to reciprocate, which in turn affects our willingness to comply with health guidelines. Third, we contribute to marketing practice by showing how to encourage compliance among lonely individuals through advertising strategies that deemphasize the communal nature of adhering to COVID‐19 prevention guidelines.

We investigate the impact of loneliness on compliance with COVID‐19 prevention guidelines across three experiments. Experiment 1A explores whether lonely individuals are less willing than non‐lonely individuals to comply with COVID‐19 prevention guidelines. Experiment 1B expands upon Experiment 1A by exploring whether a lower sense of obligation to reciprocate mediates the effect of loneliness on compliance with COVID‐19 prevention guidelines. Experiment 2 explores whether ads that encourage compliance with COVID‐19 prevention guidelines using agentic‐focused (vs. communally focused) appeals can eliminate the effect of loneliness on willingness to comply with COVID‐19 prevention guidelines.

We begin our discussion by developing our theoretical framework and then present the results of the three experiments. We conclude with our research's theoretical and managerial implications and potential directions for future research.

2. THEORETICAL BACKGROUND

2.1. Loneliness, obligation to reciprocate, and compliance with COVID‐19 prevention guidelines

Humans have an innate drive to connect with others and form social bonds (Baumeister & Leary, 1995). Relationships offered our ancestors protection and access to resources; thus, social bonds increased survival and are considered adaptive (Axelrod & Hamilton, 1981; Bowlby, 1969; Buss, 1990; Moreland, 2013). However, when individuals perceive a lack of social connections in their lives, loneliness can ensue. Typically, people report feeling lonely when they perceive both a misalignment of their relationships with their expectations (Kupersmidt et al., 1999) and a deficit in their current level of social connection (Weiss, 1973). Unfortunately, loneliness can be detrimental to an individual's social functioning (Jones et al., 1982), and it can lead individuals to become less sensitive to situations that may help them regain their sense of connection (Cacioppo & Patrick, 2008). Furthermore, lonely individuals tend to disengage from society (Woodhouse et al., 2012) and exhibit higher levels of shyness (Cassidy & Asher, 1992). Accordingly, given this research, we suggest that the motivation to pursue collective actions likely differs based on an individual's level of loneliness.

Collective actions often require a self‐other trade‐off (Schultz et al., 2017; White et al., 2019): a social dilemma in which individual self‐interest conflicts with the good of the collective (Boone et al., 2010; Cruickshank & Shaban, 2020). Such self‐other trade‐offs may include compliance with COVID‐19 prevention guidelines, which require individuals to take altruistic actions often aimed at keeping others safe such as wearing a mask, physically distancing, and washing hands (Cheng et al., 2020). We note the possibility that individuals are likely motivated by self‐protection in complying with COVID‐19 prevention guidelines; yet previous research demonstrates that risk‐reduction motives remain a weaker motivator than altruism for adhering to guidelines like mask‐wearing and physical distancing (Bir & Widmar, 2021; Nakayachi et al., 2020; Wang et al., 2021b). Accordingly, self‐protective motives do not appear to be as powerful a motivator of compliance with COVID‐19 prevention guidelines as altruism.

There are many reasons why complying with COVID‐19 prevention guidelines draws on altruism. For example, complying with these guidelines can be inconvenient or uncomfortable to the self and following them often necessitates stifling one's immediate desires, to “prioritize behaviors with ill‐defined consequences that are focused on others” (White et al., 2019, p. 30). However, following these altruistic actions may produce a return in the form of social benefits. This assertion is based on reciprocal altruism theory, which holds that individuals have evolved to act altruistically toward others beyond kin, as such acts resulted in reciprocity from others, boosting one's fitness. Thus, the theory states that individuals are motivated by the normative belief that people should help others because they are likely to need help at some later time (Trivers, 1971).

According to reciprocal altruism theory, humans are motivated to cooperate, as it fosters a positive reputation, which allows them to gain greater access to group resources (Nowak & Sigmund, 2005). Consistent with this notion, research has found that people are less inclined to deplete community resources when their reputation is at stake, and that conservationists are more likely to be chosen as group partners (Hardy & Van Vugt, 2006; Milinski et al., 2006). Altruistic acts, like complying with COVID‐19 prevention guidelines, could help to catalyse and maintain social relationships by building on the individual's prosocial reputation (Roberts, 1998; Trivers, 1971; Van Vugt et al., 2007). As such, adhering to COVID‐19 prevention guidelines may demonstrate that one is a cooperative, kind person who is willing to absorb the costs of altruism for the group's welfare. Being perceived as a good group member is also valuable as it is tied to being perceived as a more desirable companion (Cottrell et al., 2007; Stiff & Van Vugt, 2008) and can lead to other social benefits, such as reciprocity from others (Barclay, 2004). However, loneliness, which leads individuals to perceive social deficits (Weiss, 1973) and withdraw from society (Woodhouse et al., 2012), may cause individuals to reassess what they stand to gain from communal efforts in relation to the potential costs of participating in them. This implies that loneliness could impair perceptions of future reciprocity.

We suggest that loneliness may negatively impact compliance with CDC‐recommended COVID‐19 prevention guidelines like mask‐wearing, physical distancing, and vigilant handwashing. We chose to focus on these specific guidelines because following them require significant, repeated compliance on the part of the individual—unlike following other CDC guidelines like self‐quarantining, which is less commonly necessary.

We suggest that repeated compliance with COVID‐19 prevention guidelines may be costly to the self. For instance, individuals are reporting a sense of burnout resulting from mask fatigue, restriction fatigue, or an overall feeling of COVID‐19 fatigue (Berg, 2021). However, according to reciprocal altruism theory, individuals are more likely to suffer some personal cost if they assess the future prospect that their self‐other trade‐off will be reciprocated in return (Trivers, 1971). Yet, lonely individuals may not expect reciprocity in the form of social rewards because they do not expect their behavior to foster beneficial relationships. As such, we suggest that lonely individuals may be unlikely to invest in collective behaviors that decrease the spread of COVID‐19, as the personal costs of doing so likely outweigh any potential benefits that may or may not be gained in the future. In sum, we suggest that lonely individuals will experience a lower sense of obligation to reciprocate, given that they may not accrue the same social benefits from complying with COVID‐19 prevention guidelines. As a result, in light of the self‐other trade‐off motivation for adhering to COVID‐19 guidelines, we propose that lonely individuals will be less willing to comply with them than non‐lonely individuals. Accordingly, we test the following hypothesis:

Hypothesis 1

Lonely (vs. non‐lonely) consumers will be less willing to comply with COVID‐19 prevention guidelines.

Furthermore, we expect that the experience of loneliness will decrease the sense of obligation to reciprocate and that lonely individuals will be less willing to comply with COVID‐19 prevention guidelines given that they may not perceive that they will gain the same social benefits from adhering to these behaviors as non‐lonely individuals. Thus, we expect that “sense of obligation to reciprocate” will mediate the effect of loneliness on willingness to comply with COVID‐19 guidelines. Accordingly, we test the following hypothesis:

Hypothesis 2

Sense of obligation to reciprocate will mediate the relationship between loneliness and willingness to comply with COVID‐19 prevention guidelines.

2.2. Agentic and communal appeals

Research finds that messaging framing can greatly impact the appeal and compliance with public policy campaigns across a variety of health domains, including obesity (Puhl et al., 2013), alcohol use (Hassan et al., 2022), smoking (Record et al., 2018), and COVID‐19 (Pink et al., 2020). Therefore, it is important to understand how different types of advertising messaging appeals might encourage compliance with COVID‐19 prevention guidelines among lonely and non‐lonely consumers.

Past research indicates that congruity between a consumer's self‐concept and the content of advertising appeals increases the latter's impact (Hong & Zinkhan, 1995; Newman & Trump, 2019). Bakan (1966) distinguished agentic from communal characteristics in an individual's self‐concept. Agentic characteristics are related to the self, including self‐improvement, self‐esteem, self‐reliance, and control over situations (Bakan, 1966). By contrast, communal characteristics are related to a connection with others, including care for others, nurturance, and social bonding (Bakan, 1966; Bartz & Lydon, 2004). Those individuals who hold agentic characteristics are more likely to be focused on the self and a desire to differentiate themselves from others. On the other hand, those individuals who have communal characteristics are more likely to be focused on others and a desire to maintain social connections. Accordingly, given the self‐other trade‐off associated with complying with COVID‐19 prevention guidelines, we propose that advertising appeals emphasizing agentic instead of communal characteristics will be particularly effective for lonely individuals.

The appeal of communal advertising messages is likely to differ based on levels of loneliness. Complying with COVID‐19 prevention guidelines involves a self‐other trade‐off, as it requires directing resources away from the self and toward others. Thus, adhering to COVID‐19 guidelines may incur some cost or individual sacrifice. Accordingly, given that we expect lonely consumers to experience a lower sense of obligation to reciprocate than non‐lonely consumers, we also expect lonely consumers to be less persuaded by communal appeals that heighten the social aspect of compliance with COVID‐19 prevention guidelines. Thus, there is likely an incongruity between communal appeals that focus on others and lonely individuals, who typically focus on themselves. On the other hand, complying with COVID‐19 guidelines may help to catalyse and maintain social relationships, as it demonstrates to others that one is a cooperative, kind person who is willing to absorb the costs of altruism for the group's welfare, which ultimately leads to social benefits and reciprocity later on. Thus, there is likely a congruity between communal appeals that focus on others and non‐lonely individuals who experience a higher sense of obligation to reciprocate. Accordingly, we test the following hypothesis:

Hypothesis 3

After exposure to a communally focused appeal, lonely (vs. non‐lonely) consumers will be less willing to comply with COVID‐19 guidelines.

On the other hand, we predict that when exposed to agentic‐focused appeals, consumers both higher and lower in loneliness should show no difference in willingness to comply with COVID‐19 prevention guidelines. We propose this because agentic‐focused framing is more congruent with lonely consumers' emphasis on autonomy, self‐focus, and lower sense of obligation to reciprocate. In contrast, since non‐lonely consumers experience a higher sense of obligation to reciprocate, agentic‐focused framing should not negatively affect them and their willingness to comply with COVID‐19 prevention behaviors. In other words, for non‐lonely consumers, we expect that a sense of obligation to reciprocate and the desire to maintain social relations will outweigh any appeal emphasizing a focus on the self. Thus, we predict that lonely and non‐lonely individuals' willingness to comply with COVID‐19 prevention guidelines will be similar. Accordingly, we predict the following hypothesis:

Hypothesis 4

After exposure to agentic‐focused appeals, there will be no difference between willingness to comply with COVID‐19 prevention guidelines among lonely and non‐lonely consumers.

2.3. Overview of the present studies

Three experiments show that loneliness impacts people's willingness to comply with COVID‐19 prevention guidelines. Experiment 1A demonstrates that lonely (vs. non‐lonely) individuals are less willing to comply with COVID‐19 prevention guidelines, while Experiment 1B shows that this effect is mediated by a lower sense of obligation to reciprocate. Experiment 2 shows that when ads encourage compliance with COVID‐19 prevention guidelines using agentic‐focused (vs. communally focused) appeals, the effect of loneliness on willingness to comply with COVID‐19 prevention guidelines is eliminated. These results suggest that when the communal aspect of complying with COVID‐19 prevention guidelines is removed, individuals both higher and lower in loneliness show similar willingness to comply with them.

3. EXPERIMENT 1A

3.1. Materials and methods

Experiment 1 provided an initial test of Hypothesis 1—that is, whether loneliness affects consumers' willingness to comply with COVID‐19 prevention guidelines. We also tested mood as an alternative explanation for our findings, since research has found low mood to affect behavior (Gardner, 1985) and be associated with the feeling that one lacks important relationships (Baumeister & Leary, 1995).

Participants consisted of 147 undergraduate students (M _age = 19.86 years, 62% female) who participated in an experiment in return for course credit. The experiment consisted of a between‐subjects design with one manipulated, independent variable (loneliness: lonely or non‐lonely) and one measured dependent variable (willingness to comply with COVID‐19 prevention guidelines). The experiment was run between the end of October 2020 to the beginning of November 2020 amid the height of the pandemic.

In this experiment, participants were led to believe they would take part in two separate experiments to reduce demand effects, although in actuality, participants only took part in one experiment. This methodological approach is common in experimental research to reduce demand effects and the likelihood that participants will make a connection between the independent variable and the dependent variable (Das et al., 2020; Finkelstein & Fishbach, 2012; Wang et al., 2022). In the “first” experiment, participants were randomly assigned to one of two loneliness conditions (lonely vs. non‐lonely) based on Jiao and Wang's (2018) manipulation, where participants were asked to write about either feeling socially isolated or socially connected. Specifically, participants randomly assigned to the lonely condition were asked to write about a time in their life when they felt socially isolated, while participants randomly assigned to the non‐lonely condition were asked to write about a time when they felt socially connected (see Appendix A). Once participants finished the writing task, they were thanked for completing the “first” experiment and were asked to proceed to the second experiment. In the “second” experiment, participants were told they would answer a few questions about the COVID‐19 pandemic, followed by some general questions. After reading the description of “Experiment 2,” participants were asked in the next week, to what extent do you plan to (1) practice social distancing; (2) thoroughly wash your hands with soap and water (scrub back and front of hands for 20 s); and (3) wear a mask when you are around others (people outside of your immediate household) based on 5‐point scale (1 = Never to 5 = Always). These items were built based on CDC recommendations and were combined to create a scale measuring participants' willingness to comply with COVID‐19 prevention guidelines (α = .67). The measure served as our dependent variable. Finally, participants completed the PANAS scale to assess mood (positive affect, α = .87; negative affect, α = .87; Watson et al., 1988), answered some demographic questions, and then were debriefed.

3.2. Results

We conducted a one‐way analysis of variance (ANOVA) to test the effect of loneliness (contrast coded: lonely = 1, non‐lonely = −1) on willingness to comply with COVID‐19 prevention guidelines. As expected, individuals in the lonely (vs. non‐lonely) condition were less willing to comply with COVID‐19 prevention guidelines (M _lonely = 4.23 vs. M _non‐lonely = 4.46; F[1, 145] = 4.74, p = .03). These results fully support Hypothesis 1.

Neither positive nor negative mood were significant predictors of willingness to comply with COVID‐19 prevention guidelines (p's > .10). These results provide evidence to rule out mood as an alternative explanation.

3.3. Discussion

Results from Experiment 1A provide initial support for our hypothesis (Hypothesis 1) that lonely (vs. non‐lonely) individuals display less willingness to comply with COVID‐19 prevention guidelines while also ruling out mood as an alternative explanation.

4. EXPERIMENT 1B

4.1. Materials and methods

Experiment 1B examined whether sense of obligation to reciprocate mediates the relationship found in Experiment 1A. Specifically, we tested whether lonely (vs. non‐lonely) individuals are less willing to comply with COVID‐19 prevention guidelines due to a lower sense of obligation to reciprocate (Hypothesis 2).

Participants in Experiment 1B consisted of 186 undergraduate students (M _age = 19.67 years, 57% female) who participated in an experiment in return for course credit. The experiment consisted of a between‐subjects design with one manipulated, independent variable (loneliness: lonely or non‐lonely) and one measured dependent variable (willingness to comply with COVID‐19 prevention guidelines). The experiment was run at the beginning of November 2021 at a university requiring students to comply with mask and social distancing guidelines.

As with Experiment 1A, participants were led to believe they would take part in two separate experiments to reduce demand effects—although in actuality, participants only took part in one experiment. Using the same loneliness manipulation from Experiment 1A (Jiao & Wang, 2018), participants were randomly assigned to one of two conditions (lonely vs. non‐lonely). Once participants finished the writing task, participants moved on to the “second experiment” and answered the dependent variable, which consisted of the same willingness to comply with COVID‐19 prevention guidelines scale from Experiment 1A (α = .72). Next, participants answered the five‐item reciprocity scale (from Jami et al., 2021), where participants indicated at that moment the extent to which they felt “obligated,” “indebted,” “appreciative,” “thankful,” and “grateful” on a 7‐point scale (1 = Feel very little to 7 = Feel a lot) (α = .92). This scale served as our mediator. Finally, participants answered some demographic questions and were debriefed.

4.2. Results

4.2.1. Willingness to comply with COVID‐19 prevention guidelines

We conducted a one‐way ANOVA to test the effect of loneliness (contrast coded: lonely = 1, non‐lonely = −1) on willingness to comply with COVID‐19 guidelines. As expected, individuals in the lonely (vs. non‐lonely) condition were significantly less willing to comply with COVID‐19 prevention guidelines (M _lonely = 2.87 vs. M _non‐lonely = 3.19; F[1, 184] = 6.43, p = .01). As with Experiment 1A, these results fully support Hypothesis 1.

4.2.2. Mediating role of obligation to reciprocate

We predicted that the degree to which participants experienced a lower sense of obligation to reciprocate would mediate the effect of loneliness on willingness to comply with COVID‐19 prevention guidelines. To test this prediction, we conducted a 5000‐sample bootstrapped regression analysis using PROCESS Model 4 (Hayes, 2022). As expected, the pathway from loneliness to willingness to comply with COVID‐19 prevention guidelines through obligation to reciprocate was not significant (indirect effect = −0.05, 95% CI [−0.10, −0.01]), demonstrating full mediation (see Figure 1). These results support Hypothesis 2 and our theoretical framework, suggesting that lonely (vs. non‐lonely) participants display less willingness to comply with COVID‐19 prevention guidelines because they experience a lower sense of obligation to reciprocate.

FIGURE 1.

Experiment 1B: Lonely (vs. non‐lonely) individuals are less willing to comply with COVID‐19 prevention guidelines as mediated by obligation to reciprocate. *p < .05, **p < .01

To ensure that a potential interaction between X (loneliness) and M (obligation to reciprocate) did not affect the results of our mediation model, we reran the mediation test by selecting the X by M interaction option in Hayes PROCESS Model 4 (Hayes, 2022). Consistent with our theory, we did not find any interactive effect of X and M in our analysis (F[1,182] = 0.73, p = .39) and the pathway from loneliness to willingness to comply with COVID‐19 prevention guidelines through obligation to reciprocate was still not significant with the inclusion of the interactive effect of X and M (indirect effect = −0.05, 95% CI [−0.10, −0.01]), demonstrating full mediation. These results suggest that our effect still holds even when accounting for the interaction of X and M in our mediation model.

4.3. Discussion

Experiment 1B suggests that lonely (vs. non‐lonely) individuals experience a lower sense of obligation to reciprocate, which in turn, reduces their willingness to comply with COVID‐19 prevention guidelines. In total, our findings provide evidence for the proposed effect (Hypothesis 1) and support for the underlying process mechanism demonstrated through our mediator (obligation to reciprocate) (Hypothesis 2). These results suggest that lonely individuals, due to a lower sense of obligation to reciprocate, are a particular consumer segment that may necessitate targeted marketing strategies to encourage them to comply with COVID‐19 prevention guidelines. Thus, the purpose of Experiment 2 was to provide additional support for our underlying process explanation (obligation to reciprocate) by demonstrating how advertisement appeals can be used to eliminate the negative impact of loneliness on compliance with COVID‐19 prevention guidelines.

5. EXPERIMENT 2

Experiment 2 provided a test of Hypotheses 3 and 4—that is, whether the use of communal or agentic advertising appeals would impact consumers' willingness to comply with COVID‐19 prevention guidelines. Specifically, we expect that when COVID‐19 prevention guidelines are advertised using communally focused appeals, lonely (vs. non‐lonely) individuals will be less willing to comply with COVID‐19 prevention guidelines (Hypothesis 3). In contrast, when COVID‐19 prevention guidelines are advertised using agentic‐focused appeals, the negative effect of loneliness on willingness to comply with COVID‐19 prevention guidelines should be eliminated (Hypothesis 4). This should provide further support for the underlying process mechanism (obligation to reciprocate) via moderation.

Another purpose of Experiment 2 was to test ostracism as an alternative explanation for our findings on loneliness. Other research on consumer behavior has focused on social exclusion (not loneliness), which is most often operationalized through ostracism (Duclos et al., 2013; Lu & Sinha, 2017; Tunçel & Kavak, 2021; Ward & Dahl, 2014). Prior research has suggested that ostracized people and lonely people engage in different behavioral patterns. For instance, Jiao and Wang (2018) point out that those who are ostracized still see social connection in a positive way and are interested in engaging with others (Maner et al., 2007). In contrast, those who are lonely are more pessimistic and less eager to participate in social interactions (Cacioppo et al., 2006). Furthermore, ostracism serves as an antecedent to a loneliness, but it often takes time and multiple sequences of rejection for the effects of loneliness to appear (Rawat et al., 2022); thus, one can be rejected and experience a decrease in self‐esteem (Tunçel & Kavak, 2021) but not feel lonely, especially if given the opportunity to reaffiliate with the group following an experience of ostracism (Asher & Weeks, 2014). Therefore, we do not predict that ostracism will serve as an alternative explanation for our findings.

5.1. Pretest materials and methods

We conducted a pretest for Experiment 2 with 160 undergraduate students (M _age = 20.32 years, 41% female) who participated in an experiment in return for course credit. First, participants were randomly assigned to view an ad with an appeal that was manipulated to be either agentic or communally focused (adapted from Newman & Trump, 2019; see Appendices B and C for stimuli). After viewing the ad, participants completed the manipulation checks of the agentic and communally focused appeal types by responding to a six‐item measure adapted from Trapnell and Paulhus (2012). Based on a 7‐point scale (1 = Not at all to 7 = Completely), participants were asked to indicate the extent to which they perceived the appeal in the ad as representing the following agentic and communal qualities: (1) achievement; (2) status; and (3) superiority (agentic items) (α = .80) and (1) compassion; (2) civility; (3) loyalty (communal items) (α = .88). To assess the impact of the advertising appeals on self‐construal, participants answered the independent (α = .89) and interdependent (α = .86) self‐construal scale (Singelis, 1994). Finally, participants answered some demographic questions and then were debriefed.

5.2. Pretest results

A one‐way ANOVA test confirmed that participants assigned to the agentic‐focused appeal condition perceived the ad as significantly more agentic than those assigned to the communally focused appeal condition (M _agentic = 4.03 vs. M _communal = 3.62; F[1, 158] = 3.77, p = .05). Conversely, participants assigned to the communally focused appeal condition perceived the ad as significantly more communal than those assigned to the agentic‐focused appeal condition (M _communal = 4.73 vs. M _agentic = 4.08; F[1, 158] = 8.60, p < .01). This suggests that we successfully manipulated the ad as having either a communally focused or agentic‐focused appeal type.

We found no effect of our appeal type manipulations on independent (M _agentic = 5.00 vs. M _communal = 4.83; F[1, 158] = 1.53, p = .22, NS) or interdependent (M _agentic = 4.81 vs. M _communal = 4.81; F[1, 158] = .00, p = .95, NS) self‐construal. This suggests that we successfully manipulated a situational agentic or communally focused appeal type and not an appeal type that heightened independent or interdependent self‐construal.

5.3. Experiment 2 materials and methods

Experiment 2 consisted of 210 undergraduate students (M _age = 20.42 years, 61% female) who participated in an experiment in return for course credit. The experiment used a between‐subjects design with one manipulated, independent variable (appeal type: agentic or communally focused), one measured independent variable (loneliness), and one measured dependent variable (willingness to comply with COVID‐19 prevention guidelines). The experiment was run during March 2021 amid the height of the pandemic.

As with the first two experiments, participants were led to believe they would take part in two separate experiments to reduce demand effects—although in actuality, participants only took part in one experiment. In the “first” experiment, participants were asked to view an advertisement that was manipulated to use either an agentic or communally focused appeal type, in line with the pretest. Once participants finished viewing the ads, they were thanked for completing the “first” experiment and were asked to proceed to the second experiment. The “second” experiment consisted of the same willingness to comply with COVID‐19 prevention guidelines scale (α = .70) from Experiments 1A and 1B. This measure served as our dependent variable. Afterwards, participants took part in a filler task in which they rated a series of brand logos, and then completed the UCLA Loneliness Scale (Russell et al., 1980) (α = .88) and a social exclusion scale from Smith (2011) (α = .90). Finally, participants answered some demographic questions and then were debriefed.

5.4. Experiment 2 results

5.4.1. Willingness to comply with COVID‐19 prevention guidelines

Using Hayes PROCESS Model 1 (Hayes, 2022), we conducted a 5000‐sample bootstrapped multiple regression analysis to test the effect of the advertisement's appeal type (contrast coded: agentic‐focused appeal = 1, communally focused appeal = −1) and loneliness (continuously measured variable), as well as their interaction, on willingness to comply with COVID‐19 prevention guidelines. As expected, the analysis revealed a significant two‐way interaction of loneliness and the advertisement appeal type (β = .23, t[1, 206] = 2.06, p = .04) (Figure 2).

FIGURE 2.

Experiment 2: The effect of higher (vs. lower) loneliness as a function of advertisement appeal type on willingness to comply with COVID‐19 prevention guidelines

As anticipated, when the advertisement used a communally focused appeal, we found a negative association between loneliness and willingness to comply with COVID‐19 prevention guidelines. Specifically, as an individual's level of loneliness increased, they were significantly less willing to comply with COVID‐19 prevention guidelines when exposed to an ad using a communally focused appeal (β = −.34, t = −2.24, p = .03). Most importantly, though, when the ad used an agentic‐focused appeal, the difference in willingness to comply with COVID‐19 prevention guidelines between individuals higher and lower in loneliness was no longer significant (β = .12, t = 0.73, p = .47). These results fully support Hypotheses 3 and 4.

5.4.2. Advertisement appeal type

To understand whether exposure to the advertisement's appeal type (communal vs. agentic) influenced feelings of loneliness, we conducted a one‐way ANOVA test. As expected, we found no significant effect of appeal type on loneliness (M _agentic = 1.68 vs. M _communal = 1.65; F[1, 208] = .20, p = .66). Accordingly, these findings do not support the conclusion that the advertisement's appeal type manipulation affected feelings of loneliness.

5.4.3. Ostracism

To understand whether feelings of ostracism could account for our findings, we conducted a 5000‐sample bootstrapped multiple regression analysis using PROCESS model 1 (Hayes, 2022). This allowed us to test the effect of appeal type (contrast coded: agentic‐focused appeal = 1, communally focused appeal = −1) and ostracism (continuously measured variable), as well as their interaction, on willingness to comply with COVID‐19 prevention guidelines. As expected, there was no significant interaction of ostracism and appeal type (β = .04, t = 1.00, p > .10) and no significant main effect of ostracism on willingness to comply with COVID‐19 prevention guidelines (β = −.07, t = −1.51, p > .10). Accordingly, these findings do not support the alternative explanation that ostracism accounts for our earlier results.

5.5. Discussion

Experiment 2 provides additional support for our process explanation (obligation to reciprocate) while also revealing a means to eliminate the negative impact of loneliness on willingness to comply with COVID‐19 prevention guidelines. Specifically, we find further evidence that individuals who experience higher (vs. lower) levels of loneliness are less willing to comply with COVID‐19 prevention guidelines, but only when these guidelines are promoted using communally focused (vs. agentic‐focused) appeals (Hypothesis 3). Thus, our results suggest that lonely (vs. non‐lonely) individuals may be less motivated to comply with COVID‐19 prevention guidelines that heighten communal (vs. agentic‐focused) aspects because they experience a lower sense of obligation to reciprocate. Furthermore, we demonstrate that advertisements using agentic‐focused appeals can lead lonely individuals to be as willing to comply with COVID‐19 prevention guidelines as non‐lonely individuals (Hypothesis 4). Thus, the use of agentic‐focused appeals may be an effective method to motivate compliance with COVID‐19 prevention guidelines among lonely individuals. Experiment 2 also ruled out ostracism as an alternative explanation for our results.

6. CONCLUSION

Across three experiments, we find that lonely individuals are less willing to comply with COVID‐19 prevention guidelines. Our results suggest this occurs because lonely (vs. non‐lonely) individuals experience a lower sense of obligation to reciprocate (Experiment 1B). In addition, we demonstrate that the negative impact of loneliness on compliance with COVID‐19 prevention guidelines can be offset through advertising messaging appeals (Experiment 2). Specifically, we find that individuals higher (vs. lower) in loneliness are more willing to comply with these guidelines when information about COVID‐19 is framed using an agentic as opposed to communally focused appeal (Experiment 2). Thus, we suggest that using agentic‐focused appeals may be an effective method to motivate compliance with COVID‐19 prevention guidelines among lonely individuals. Alternatively, lonely individuals may be less responsive to appeals emphasizing the social or communal nature of complying with COVID‐19 prevention guidelines. Overall, these results raise many important theoretical and managerial implications, leading to potential areas of future research.

6.1. Theoretical contributions

This research is the first to establish a relationship between loneliness and consumers' compliance with COVID‐19 prevention guidelines. In doing so, this research fills a gap in the emerging stream of research that has begun to explore individual factors impacting the willingness to comply with COVID‐19 prevention guidelines (Han, 2021; Jordan et al., 2021; Kim et al., 2022; Oosterhoff & Palmer, 2020; Uddin et al., 2021; Xu & Cheng, 2021). Specifically, this research is unique as it focuses on loneliness and provides an initial understanding of how social psychological factors, also referred to as psychosocial factors, influence compliance with COVID‐19 prevention guidelines. This emphasis on loneliness as the primary psychosocial component is important given that social factors are crucial drivers of consumer behavior (Ramya & Ali, 2016; Youn et al., 2022). Furthermore, there is a potentially reinforcing effect of COVID‐19 on loneliness. During the pandemic, individuals were asked to physically distance and quarantine, which limited socialization, increased isolation, and ultimately increased levels of loneliness (Groarke et al., 2020; Killgore et al., 2020; Li & Wang, 2020). Thus, given the rise of loneliness before the COVID‐19 pandemic (Jeste et al., 2020) and during it (Groarke et al., 2020; Killgore et al., 2020; Li & Wang, 2020), loneliness and other psychosocial factors may have far‐reaching implications for the ability to successfully combat COVID‐19 and encourage compliance with other public health behaviors.

This research extends our understanding of how loneliness relates to consumer decision making by demonstrating why lonely consumers may be less likely to follow COVID‐19 prevention guidelines. Specifically, we add to literature on reciprocal altruism theory (Roberts, 1998; Trivers, 1971; Van Vugt et al., 2007) by finding that loneliness can lead to a lower sense of obligation to reciprocate. Thus, by assessing how loneliness can impair a sense of obligation to reciprocate, we may be able to explain why some individuals are reluctant to follow COVID‐19 prevention guidelines. Our research suggests that lonely individuals may be less responsive to the social nature of complying with COVID‐19 prevention guidelines, given that they may be less likely to obtain the social benefits from doing so. That is, lonely individuals may feel that the self‐other trade‐off inherent in complying with COVID‐19 prevention guidelines is too costly and that their acts of altruism may potentially go unnoticed, and therefore, unreciprocated. Thus, we find evidence that loneliness can impair perceptions of future reciprocity.

In addition, we extend research showing the importance of emphasizing congruity between a consumer's self‐concept and the use of advertising appeals (Hong & Zinkhan, 1995; Newman & Trump, 2019; Price et al., 2018). Our research uniquely reveals the benefits of congruity between loneliness and the use of agentic versus communal messaging appeals within the public health domain. Specifically, we show that lonely consumers may not respond well to communal appeals that encourage a connection with others, care for others, nurturance, and social bonding. However, lonely consumers may respond well to agentic appeals that are more congruent with their emphasis on the self. Thus, loneliness is an important factor when considering how to design advertisements that encourage public health compliance, such as adherence to COVID‐19 prevention guidelines.

6.2. Managerial and societal implications

Our findings have the potential to guide efforts to mitigate the current COVID‐19 pandemic as well as future crises. In this research, we show an additional and unintended consequence of the increasing trend of loneliness: specifically, that loneliness may lead individuals to be less willing to comply with COVID‐19 prevention guidelines. This is particularly troubling when considering the success of these guidelines in combating COVID‐19 (Eikenberry et al., 2020; Howard et al., 2021; Lio et al., 2020; Moosa, 2020) and the presence of consumers who have been reluctant to comply with them (Belanger & Leander, 2020; Wang et al., 2021a). Thus, for a variety of reasons, marketers and policymakers would benefit from understanding how to better market behaviors that comply with community health initiatives—especially among consumer segments that may be particularly resistant to them, such as those who are lonely.

Based on our research, we offer recommendations on how to target a growing segment of consumers that report feelings of loneliness. Our findings suggest that lonely individuals are less willing to comply with COVID‐19 prevention guidelines because they experience a lower sense of obligation to reciprocate. Thus, marketing advertisements that appeal to the communal or social nature of compliance may be less persuasive and effective for lonely individuals. Instead, when promoting health compliance guidelines to lonely individuals, we suggest that marketers may want to use agentic messaging, which focuses more on the individual's own well‐being and safety rather than that of the group, as lonely individuals feel a lower sense of obligation to reciprocate.

Given our findings, we recommend that marketers and policymakers consider using segmentation strategies when encouraging compliance with COVID‐19 prevention guidelines. Our results suggest that a uniform, targeted approach to compliance with COVID‐19 prevention guidelines may not be beneficial and could potentially backfire with some consumer segments. This may help to explain why some public health campaigns encouraging compliance with COVID‐19 prevention guidelines have been so unsuccessful (Robinson & Veresiu, 2021; Sauer et al., 2021). In the future, marketers may wish to segment and target different campaigns to specific groups who feel higher levels of loneliness, such as those who live alone or are disabled, divorced, or widowed. In addition, new advances in artificial intelligence and big data are providing policymakers and marketers with the ability to detect loneliness via individuals' smartphones (Khatchadourian, 2015), which may allow for tailored messaging strategies that target lonely consumers. Going further, marketers may also utilize sentiment analysis techniques through social media platforms, like Twitter, to track how different segments of consumers (e.g., lonely vs. non‐lonely consumers; rural vs. urban consumers) react to marketing campaigns and advertising strategies.

6.3. Future research

Future research within marketing should explore how loneliness impacts a variety of consumer behaviors—including behaviors related to healthcare and public health, whether pandemic‐related or not, and consumer behaviors in general (Chen et al., 2021; Jiao & Wang, 2018; Rippé et al., 2018, 2022). For instance, does loneliness lead to decreased or increased vaccination rates? We chose not to focus on this specific question in our research, as we wanted to explore COVID‐19 prevention guidelines that require significant, repeated sacrifices for the individual on behalf of the collective. Since people are vaccinated infrequently, it could be argued that individual vaccination requires less of a sacrifice for the individual than complying with COVID‐19 prevention guidelines that disrupt daily life (daily mask‐wearing, social distancing, and vigilant handwashing). Based on our research, whether a relationship exists between loneliness and vaccination likely depends on the motivational drivers of vaccination rates, and how much a sense of obligation to reciprocate is present. Thus, future research should explore whether a relationship exists between loneliness, vaccination, and related consumer health behaviors. For example, future research could investigate whether loneliness exacerbates other unhealthy behaviors that were prevalent during the COVID‐19 pandemic, including the overconsumption of alcohol and unhealthy snacks (Gordon‐Wilson, 2022; Wright & Schultz, 2022). In addition, future research may wish to explore the reinforcing relationship between COVID‐19 prevention guidelines and loneliness. In our research, we studied the impact of loneliness on COVID‐19 prevention guidelines. However, it is likely that following COVID‐19 prevention guidelines exacerbates loneliness. We do note that in our research, we predominantly manipulated loneliness, with the exception of Experiment 2, where loneliness was measured. Therefore, it is unlikely that willingness to comply with COVID‐19 guidelines caused loneliness to increase in our experiments, and thus accounted for our results. However, future research should seek to further disentangle the relationship between these two variables.

Although it is one of the predominant methodologies in the social sciences to test for mediation (Sarstedt et al., 2020), our use of Hayes PROCESS (Hayes, 2022) could not eliminate a limitation that arises with mediation models in general. That is, given the mediator and dependent variable are measured, as opposed to manipulated, the causal relationship between the mediator and the dependent variable cannot be fully established (Hayes, 2022; Qin & Yang, 2021), even with other statistical approaches to mediation like causal mediation. Since we did not include any confounding variables in Experiment 1B, we were unable to run a sensitivity analysis with our data to understand the extent to which confounding variables affect the relationship between the mediator and the dependent variable. Thus, the causal relationship between our mediator (obligation to reciprocate) and dependent variable (willingness to comply with COVID‐19 prevention guidelines) in Experiment 1B cannot be unequivocally established. However, we did use experimental design in Experiment 1B. According to Hayes, “No other design [experimental design] gives a researcher more confidence in the claim that differences between groups defined by X on some variables of interest are due to X rather than something else” (Hayes, 2022, p. 128). Since causality is a function of theory and research design (Hayes, 2022), future research should consider and test for other confounders that influence the relationship among variables in the mediation model. Future research might also incorporate longitudinal designs, where the mediator (obligation to reciprocate) and dependent variable (willingness to comply with COVID‐19 prevention guidelines) would be measured across multiple occasions.

We acknowledge an additional limitation of our research, which is that certain confounding factors could have reduced the validity and generalization of our findings. For example, while we found evidence for the effect of loneliness on compliance with COVID‐19 prevention guidelines across three experimental studies, the act of self‐quarantine could impact the frequency with which individuals wear masks and wash their hands. In addition, individuals with certain chronic illnesses may not be able to wear masks at all. Accordingly, future research on loneliness could test whether confounding variables like these impact compliance with COVID‐19 prevention guidelines.

Additional research should also be conducted on how a sense of obligation to reciprocate affects a variety of consumer behaviors. For instance, a sense of obligation to reciprocate likely drives compliance in other healthcare behaviors, such as whether a lonely individual may be more inclined to show up late for an appointment, more inclined to cancel an appointment last minute, or not show up at all because they experience less of sense of obligation to healthcare providers. Exploring this question is important since little research exists about the psychological factors affecting patient appointment compliance, even though 23% of medical appointments result in no‐shows (Dantas et al., 2018; Milne, 2010). No‐shows can often adversely affect a patient's health outcomes, delay diagnosis or treatment, and reduce effective clinic capacity (Hwang et al., 2015; LaGanga & Lawrence, 2012). By measuring factors like loneliness, researchers might be able to develop better predictive models for patient scheduling and appointment overbooking. Furthermore, given that lonely individuals may feel a lower sense of obligation to reciprocate, it is possible that this group is more likely to participate in negative consumption behaviors like the panic buying that was so prevalent during COVID‐19 pandemic's early months (Billore & Anisimova, 2021; Prentice et al., 2022).

In addition, given our findings that lonely individuals show a lower sense of obligation to reciprocate, future research may want to test the use of agentic vs. communal messaging appeals within healthcare settings. For instance, based on our research, lonely individuals may prefer agentic appeals emphasizing how not showing up to appointments could impact their own health (e.g., longer waits and reduced care for them). However, non‐lonely individuals may prefer communal appeals that emphasize social motives for attending their appointment, such as how not showing up will inconvenience staff, or how tardiness could lead to longer waits for other patients.

Loneliness may also be a particularly interesting construct for exploring how consumers navigate non‐health related behaviors, such as those within the digital world. Given our findings that lonely consumers experience a lower sense of obligation to reciprocate, this may impact how consumers leave reviews, interact with brands through social media, and respond to targeted or individual marketing messages. Would a lonely consumer, for example, be less willing than a non‐lonely consumer to leave a positive review after a good service encounter as they may not experience as much of an obligation to reciprocate? Future research may also wish to explore how relationship marketing affects lonely consumers. For example, will lonely consumers be less concerned about being good members of a brand community or responding well to brands that create spaces and marketing offerings that foster a sense of belonging with other consumers? In other words, marketing toward and responding to lonely consumers may prove difficult, and future research should be conducted on this particular segment of consumers.

In closing, given the pervasive rise of loneliness within modern society (Jeste et al., 2020), it is crucial that a variety of stakeholders (researchers, marketers, policymakers, and health care groups) better understand the impact of loneliness on consumer behavior. This is especially important within the public health domain given the global devastation caused by the COVID‐19 pandemic and the likelihood that its impact will reach far into the future. We hope that our findings will inspire additional research about loneliness and specifically on how to encourage better compliance with consumer behaviors that are beneficial to society.

FUNDING INFORMATION

This research was supported by an internal research grant from the Arthur F. and Patricia Ryan Center for Business Studies at Providence College.

CONFLICT OF INTEREST

The authors declare no conflict of interests.

ETHICS STATEMENT

The authors confirm that the appropriate approval of using human subjects was obtained through the IRB of Providence College and all studies conform to recognized standards of research. The authors also confirm informed consent was obtained by all human subjects.

Biographies

Ainslie E. Schultz received her PhD from the University of Arizona. She is an Assistant Professor of Marketing in the Arthur F. and Patricia Ryan Center for Business Studies at Providence College. Ainslie's research focuses on ways to improve consumer decision making. The goal of her research is to investigate questions that have implications for consumers' well‐being and to use that knowledge to advise organizations about responding to consumer needs, reengaging consumers with existing products and services, or nudging consumers toward better decisions. Ainslie has published her research in the Journal of Business Ethics, Journal of Consumer Research, and Journal of Business Research. Prior to starting her career in academia, she worked as an analyst for IMS Research.

Kevin P. Newman received his Ph.D. from the University of Arizona. He is an Associate Professor of Marketing in the Arthur F. and Patricia Ryan Center for Business Studies at Providence College. Kevin uses experimental research techniques to better understand consumer behavior issues relating to corporate social responsibility, ethical branding, moral identity and behavior, self‐control, and how consumers cultivate their identities through brands. He has published articles in these research areas at the Journal of Consumer Psychology, Journal of Business Ethics, Marketing Letters, and Psychology & Marketing.

APPENDIX A.

A.1. Loneliness manipulations

From Jiao and Wang (2018).

A.1.1. Lonely condition

Please recall, in as much detail as you can, a time when: You felt very lonely (e.g., feeling isolated, not having a high sense of intimacy, companionship, friendship, togetherness, or feelings of belonging).

In the space below, describe in as much detail as possible, a time that you felt lonely: Where were you? When was that? What were you doing? And why did you feel so lonely?

A.1.2. Non‐lonely condition

Please recall, in as much detail as you can, a time when: You felt very connected (e.g., having a high sense of intimacy, companionship, friendship, and feelings of belonging and being loved).

In the space below, describe in as much detail as possible, a time that you felt connected: Where were you? When was that? What were you doing? And why did you feel so connected?

APPENDIX B.

B.1. Experiment 2 agentic‐focused appeal

APPENDIX C.

C.1. Experiment 2 communally focused appeal

Schultz, A. E. , & Newman, K. P. (2023). The impact of loneliness on compliance with COVID‐19 prevention guidelines. International Journal of Consumer Studies, 47(1), 59–73. 10.1111/ijcs.12884

DATA AVAILABILITY STATEMENT

The data that support the findings of this research are available from the corresponding author upon reasonable request. Data were collected at Providence College.