Corporate social advocacy and gender equality: how call-to-action messages influence corporate reputation

Abstract

Purpose

Corporations are confronted with growing demands to take a stand on socio-political issues, i.e. corporate social advocacy (CSA), which affects their reputation in the public. Companies use different CSA message strategies, including calling the public to support and act on the issue they advocate. Using reactance theory, the authors investigate the impact of CSA messages with a call to action on corporate reputation in the case of a company's gender equality initiative.

Design/methodology/approach

A one-factorial (CSA message with or without a call to action) between-subjects experiment was conducted by surveying 172 individuals living in Switzerland. The CSA messages were created in the context of gender equality.

Findings

The authors' study indicates that CSA messages with a call to action compared to those without overall harmed corporate reputation due to individuals' reactance, which is higher for CSA messages with a call to action, negatively affecting corporate reputation. The impact of the CSA message strategy with a call to action on corporate reputation remains significant after controlling for issue alignment and political leaning.

Originality/value

Communicating about socio-political issues, especially taking a stand, is a significant challenge for corporations in an increasingly polarized society and has often led to backlash, boycotts and damage to corporate reputation. This study shows that the possible adverse effects of advocating for socio-political issues can be related to reactance. It emphasizes that companies advocating for contested issues must be more cautious about the message strategy than the issue itself.

Citation

Marschlich, S. and Bernet, L. (2024), "Corporate social advocacy and gender equality: how call-to-action messages influence corporate reputation", Corporate Communications: An International Journal, Vol. 29 No. 7, pp. 21-38. https://doi.org/10.1108/CCIJ-10-2023-0142

Introduction

Corporations increasingly face growing demands to take a stand on social issues and more frequently engage in corporate social advocacy (CSA). As an essential corporate communication function, CSA can be conceived as “the taking of a public stance on a controversial social-political issue by corporations” (Dodd and Supa, 2015, p. 287) and is related to corporate reputation. However, prior scholarship is inconsistent on the relationship between CSA communication and corporate reputation. It outlines that CSA can benefit corporate reputation provided it is perceived as authentic and fits the company's identity and public expectations about its values (Lim and Young, 2021). Other research indicates that corporate reputation may suffer from CSA communication (Abitbol et al., 2018).

The inconsistency of CSA's effects on perceptions of or attitudes toward a company may depend on intervening variables, particularly the chosen message strategy. Companies such as Ben & Jerry's or Patagonia use various CSA communication strategies, including asking their stakeholders to take action on petitions to demonstrate support for a social issue. At the same time, companies are afraid of attracting negative coverage or being boycotted due to their CSA engagements. Previous research investigated different factors affecting the perception of CSA and its effects on the perception of a corporation, including the chosen CSA issue (Dodd and Supa, 2014) and value alignment between consumers and the CSA issue (Lim and Young, 2021), but the effect of the communication message strategy has not yet been sufficiently researched.

In response to the rise in corporations that use calls to action, e.g. to support the corporation in response to a specific regulatory change and, in this way, to mobilize stakeholders and the mixed results on CSA effects on corporate reputation, this paper investigates the impact of CSA communication strategy that includes a call to action in the case of a gender equality campaign. Gender equality is one of the most relevant challenges in contemporary society and one pillar of the United Nations' sustainable development goals (SDGs) that many corporations tackle and promote. Our guiding research question is: To what extent and how does CSA communication with a call to action regarding gender equality affect corporate reputation?

Bringing together reactance theory (Brehm, 1966) and previous studies on CSA (e.g. Hong and Li, 2020; McDermott and Lachlan, 2020), we test a model hypothesizing that the effect of CSA with a call to action on corporate reputation depends on the extent of the individual's reactance and identification with the company. With a between-subjects experimental design, we surveyed individuals living in Switzerland to test our hypothesized model. Our results indicate that CSA messages with a call to action regarding gender equality led to higher reactance, negatively affecting the individuals' identification with the company and corporate reputation. Our results emphasize that companies engaging in CSA must be more cautious about the message strategy than the issue itself, as mobilizing strategies that include a call to action may threaten the identification with the company and its reputation. Our research contributes to a better understanding of CSA messaging and lays out a strategic communication approach companies can take in practice.

Literature review

From corporate social responsibility to corporate social advocacy

Corporate social responsibility (CSR) has grown in recent decades to become one of the central research fields in public relations and strategic communication. Many studies assume that, besides their economic function, companies have a broader responsibility to society, including sustainable operations, ethical behavior and philanthropic measures (Carroll and Brown, 2018). According to stakeholder theory, the interests of a company's stakeholders must be fulfilled to ensure tangible and intangible corporate assets in the medium and long term (Freeman, 1984). In this context, CSR communication is supposed to secure or improve the legitimacy and reputation of a company (e.g. Bachmann and Ingenhoff, 2016; Colleoni, 2013; Lock and Schulz-Knappe, 2019). Authors have increasingly drawn attention to political forms of CSR (Ingenhoff and Marschlich, 2019; van der Meer and Jonkman, 2021). Scholars argue that the politicization of companies is related to the increased visibility of corporations due to mediatization and polarization (van der Meer and Jonkman, 2021). The political polarization in society can be observed around socially relevant issues, such as social equality and discrimination, environmental sustainability, or immigration (van der Meer and Jonkman, 2021), on which companies increasingly take a stand, even if it is not related to their core business (Dodd and Supa, 2014). In this new social environment, companies find themselves in a dichotomy between the expectation to speak out on (morally and emotionally charged) social issues with global relevance to retain their social license to operate (van der Meer and Jonkman, 2021) and running the risk of being boycotted if consumer groups do not accept their socio-political commitment (Rim et al., 2020).

Against this background, new forms of corporate social engagement have emerged, including CSA (Dodd and Supa, 2014, 2015), corporate and chief executive officer (CEO) activism (Chatterji and Toffel, 2019), corporate diplomacy (Marschlich and Ingenhoff, 2021) and political CSR (Ingenhoff and Marschlich, 2019). The distinction between the terms is not always precise, and they are sometimes used synonymously. What most of the approaches have in common is that companies express public support for individuals or groups (Ciszek and Logan, 2018), taking a stand on a social issue (Eilert and Nappier Cherup, 2020) to induce others to do the same and ultimately promote social change (Eilert and Nappier Cherup, 2020). From a strategic communication perspective, communicating about socio-political issues is relevant as it can affect corporate reputation (Abitbol et al., 2018).

Our study relies on CSA, considering it a new form of CSR (Dodd and Supa, 2014; Hong and Li, 2020). The difference between CSA and CSR is that CSA mainly tackles issues that are not connected to the core business (Dodd and Supa, 2014, 2015) and that it involves public statements on social issues, i.e. topics that are discussed in public, which are often polarizing and controversial (Hong and Li, 2020). In contrast, CSR mainly involves philanthropic activities that are non-controversial and non-polarizing. The perspective on the company also differs between the two constructs. Both terms have in common that companies engage in efforts and issues that are not primarily supposed to serve corporate self-interests but at least support the company's stake (see Dodd and Supa, 2014, 2015). While CSR scholars disagree on the extent to which CSR efforts are connected to the company's core business, the CSA literature suggests that the “social-political issues addressed by organizations are divorced from issues of particular relevance to the organization” in the case of CSA (Dodd and Supa, 2015, p. 289). Troy et al. (2023) likewise pointed out that the difference between CSA and CSR messages is that CSA messages include a call for change going beyond the companies' core business.

Ciszek and Logan (2018) defined CSA as “the public support for individuals, groups, ideals, or values that is intended to persuade others to do the same” (p. 118). Eilert and Nappier Cherup (2020) even noted that corporate activism aims to create societal change by influencing a company's institutional environment by taking a stand on issues of public concern. Ciszek and Logan (2018) noted that CSA has much in common with public relations; both advocate for justice and equality. They further concluded that public relations could help “organizations define their values, strategically communicate, and fight for social justice causes” (p. 125). Building on the literature review, we define CSA communication as the communication efforts of companies to take a stance on a socially relevant issue that is (potentially) discussed controversially in the public sphere, thereby expressing societal values and norms. CSA communication aims to contribute to society, for instance, by promoting equality and positively influencing the public's perception of the company.

CSA, corporate reputation and gender equality

When companies take a stand on socio-political issues, this can affect the public perception of companies (Dodd and Supa, 2014; Kim, 2019), including their reputation (Lim and Young, 2021). Corporate reputation can generally be understood as a collective perception of a company resulting from its actions (Fombrun et al., 2000) and its ability to meet public expectations regarding its decisions (Eisenegger, 2005). For companies, it is highly relevant to maintain a (positive or strong) reputation as it can benefit corporate financial outcomes (Roberts and Dowling, 2002) or positively affect relationships with the public and other stakeholder groups (Fombrun and van Riel, 2003).

While some studies situate reputation predominantly at the cognitive level (e.g. Fombrun et al., 2000), others consider emotions part of reputation (e.g. Deephouse, 2000). In the first case, Fombrun et al. (2000) defined corporate reputation as “a cognitive representation of a company's actions and results that crystallizes the firm's ability to deliver valued outcomes to its stakeholders” (p. 87). Integrating cognitive and emotional attributes, Deephouse (2000) defined corporate reputation as the “evaluation of a firm by its stakeholders in terms of their affect, esteem, [and] knowledge” (p. 1093). Walker (2010) systematically reviewed previous definitions of corporate reputation. He pointed out that corporate reputation is reflected in a positive or a negative and critical attitude toward the company, is temporal and comparative and can be issue-specific and thus change over time and across issues. Walsh and Beatty (2007) emphasized that corporate reputation relates to the evaluator's recognition and interpretation of “what the firm stands for” (p. 129). Hence, as a value-driven communication effort, CSA is often moral in nature and concerned with (frequently contested) social issues of public concern and is expected to affect corporate reputation. Building on the literature, we define corporate reputation as individuals' perception of a company reflecting the evaluation of the outcomes of a company's actions.

Previous research has examined the relationships between CSA and various constructs, including purchase intentions (e.g. Dodd and Supa, 2014), consumer responses such as boycott intentions (e.g. Rim et al., 2020), corporate and brand attitudes (e.g. Klostermann et al., 2022), corporate reputation (Abitbol et al., 2018; Lim and Young, 2021), or general perceptions of the company (Troy et al., 2023) indicating mixed results. Lim and Young (2021) explored the effects of Ben & Jerry's advocacy of LGBTQ (lesbian, gay, bisexual, transgender and queer) rights on the company's reputation. They found positive effects of CSA on corporate reputation provided that the advocacy efforts are perceived as authentic and fit the brand's identity and the public expectations about the brand's values. Abitbol et al. (2018) examined CSA in the case of Starbucks' Race Together campaign and showed that corporate reputation suffered from the campaign, but financial returns were not affected. The authors explained the negative association with a lack of company-cause-fit.

Perceptions of CSA are related to the issue the company is advocating for (see also Walker, 2010 on the relationship between corporate reputation and issues). In this study, we address the issue of gender equality. Gender equality is a socially relevant issue on a global level. For example, the United Nations has identified gender equality as a central pillar of its Sustainable Development Goals (United Nation, 2023). Many corporations already engage in that issue, e.g. providing information to raise awareness or create initiatives on this topic. Gender equality is, therefore, a highly relevant and widespread social issue and thus relevant for corporate communication practices and society. In addition, it is a public issue that people from different social classes and backgrounds and with varying levels of education and income find relevant. Finally, the issue of gender equality can be seen as political, such that in many countries, policy-makers are enacting various measures to equalize conditions for women and men (e.g. by introducing paternity leave, recommendations on quotas for women in the workforce, especially in management positions, etc.). Therefore, the issue lends itself very well to the present study.

CSA communication strategies: the effect of the call to action

Within the existing literature on CSA and its effects on the perception of companies, too little attention has been paid to the role of communication itself, i.e. the role that different communication strategies may have, with few exceptions. Zhang and Zhou (2023) examined the effect of communicating different CSA positions (pro versus anti) and found that the message type is less relevant to the individuals' reactions to CSA communication. In contrast, the individuals' preexisting attitudes regarding the advocated issue in interaction with the CSA position can lead to cognitive dissonance, negatively affecting consumer reactions to CSA (Zhang and Zhou, 2023). Troy et al. (2023) examined whether CSR and CSA messaging affect corporate perceptions differently. The CSA message, compared to the CSR message, included a call for change. However, the study found no significant differences between both message types and their effects (Troy et al., 2023). Regarding support intentions, prior research found that psychological traits, including individuals' self-efficacy, were positively related to individuals' word-of-mouth intentions to support CSA efforts (Kim et al., 2023). In practice, we can observe that companies such as Ben & Jerry's or Patagonia use various strategies to communicate about socio-political issues in terms of CSA to maintain their reputation. They can make a public statement on the social issue to express their perception of the issue, or they can go one step further and ask the public to take action on the social issue as more and more corporations do in practice. As such, Ben & Jerry's asked their customers several times to actively support their efforts, e.g. by signing a petition to stop mining in the Grassy Narrows Territory in Ontario, Canada, a sovereign and protected area (Ben & Jerry's, 2021).

While previous scholars studied different aspects of CSA that may affect the public responses to CSA (e.g. Dodd and Supa, 2015; Rim et al., 2022; Zhou and Dong, 2022) and research outlined different types of corporate advocacy efforts, including raising political voices by asking for public support regarding a petition (e.g. You and Hon, 2022), little is known about whether a public call to action may affect corporate reputation positively or negatively, if at all. As Vasquez (2022) emphasized, scholarship needs to investigate more the communication of CSA, the call to action, and the public responses to it, “as corporations can use their voice to shape public opinion and influence policy-makers to take action” (p. 12). A call to action makes CSA efforts much more engaging and may lead to societal change if individuals follow it. However, companies often fear attracting negative coverage or being boycotted due to their CSA engagements. We argue that the potential boycott versus support is related to the message strategy they apply to reach their stakeholders, which may cause psychological reactance (Brehm, 1966) and lead to adverse effects. Since CSA efforts with a call to action can more and more be observed in corporate practices, we pose the following research question:

RQ1.

To what extent and how does CSA communication with a call to action regarding gender equality affect corporate reputation?

So far, research on corporate communication efforts that include a call to action lacks significantly. According to Suchman (1995), the effectiveness of the communication efforts that seek to improve perceptions and evaluations of companies depends on the degree of the call for public support. Accordingly, if organizations seek active public support, like through a call to action, the threshold of gaining legitimacy is higher compared to if an organization does not ask for public support because the audience needs to evaluate the company's procedure more accurately, i.e. they are cognitively more engaged when processing the message (Suchman, 1995). Since legitimacy and reputation overlap and can be considered similar regarding audience perceptions (see Deephouse and Carter, 2005), we apply Suchman's (1995) argument to our study and pose the following hypothesis.

H1.

CSA with a call to action leads to a lower corporate reputation than CSA without a call to action.

The mediating role of reactance and the identification with the company

One major challenge for corporate communications is individuals' perception of persuasion intents, e.g. the feeling of individuals that a company, with its self-interests, intends to convince them of its arguments. The phenomenon of persuasion intent has already been studied in CSR communication (Bachmann and Ingenhoff, 2016). Particularly when it comes to companies' socio-political activities and thus also CSA initiatives, people tend to be skeptical, for instance, because individuals perceive self-interest as the underlying motive of a company's social engagement instead of altruism (e.g. Viererbl and Koch, 2022). As Bachmann and Ingenhoff (2016) showed, skepticism toward CSR can lead to reactance, threatening corporate legitimacy, and thus, it may also harm the corporate reputation. According to reactance theory, reactance describes temporary psychological arousal that arises when people feel restricted in their personal freedom and want to counteract this perceived loss of freedom (Brehm, 1966). This may involve adopting a defensive stance toward the received message. Åkestam et al. (2017) described that “an individual experiencing reactance to a stimulus becomes more resistant to persuasion” (p. 797).

Following the argument by Suchman (1995) that the legitimation attempts of companies seeking public support involve a higher (cognitive) engagement of the audience that evaluates the company and its efforts, we argue that CSA with a call to action may cause (more) reactance because individuals might spend more efforts thinking about it and feel persuasion intent. Accordingly, we hypothesize the following:

H2.

CSA with a call to action leads to higher reactance than CSA without.

Previous research outlined that CSR communication and other forms of corporate communication that include political or social issues might enhance feelings of identification with the company (e.g. Eberle et al., 2013), which was found to play a critical role in affecting corporate reputation (Eberle et al., 2013; Kim, 2019). Since CSA, at its core, involves the communication of values and is concerned chiefly with morally charged or even polarizing issues, it seems plausible that CSA communication affects an individual's identification with the company. This argument is further supported by previous research, indicating that the congruence between companies' social and political efforts and individuals' values is central to achieving positive effects of corporate advocacy on perceptions of a company and behavioral intentions toward corporate actions (Weber et al., 2023). At the same time, as Zhang and Zhou (2023) emphasized, consumer-company identification can negatively influence individuals' reactions to CSA through cognitive dissonance in case of incongruence between individuals' attitudes towards an issue and the advocated CSA issue.

Building on social identity theory, Bhattacharya and Sen (2003) argued that some companies “represent and offer attractive, meaningful social identities to consumers that help them satisfy important self-definitional needs” (p. 77), which then results in individuals identifying with the company and their efforts. In this paper, identification with a company is conceived in the sense of “consumer-company identification” (Bhattacharya and Sen, 2003) as the extent to which individuals perceive a congruence between the values that define their self-identity and the companies' efforts (Du et al., 2007). Since CSA messages with a call to action are more encouraging, they might give individuals a more “meaningful social identity” than CSA messages without a call to action. Moreover, Bhattacharya and Sen (2003) explained that a company's identity reflects its core values, which is probably more visible in the case of CSA messages with a call to action. Thus, we argue that they might lead to a higher identification with the company than CSA messages without a call to action.

H3.

CSA with a call to action leads to a higher identification with the company than CSA without it.

As outlined, higher levels of identification with the company were found to increase the perceived reputation of a company in the case of CSR and CSA (Eberle et al., 2013; Kim, 2019), while lower levels of identification can cause cognitive dissonance, which may threaten corporate reputation (Hong and Li, 2020; Zhang and Zhou, 2023). Likewise, previous research indicates that when individuals' opinions differ from the company's stand, which lowers the identification with the company, individuals may perceive the company negatively, resulting in a more negative or lower corporate reputation (Chatterji and Toffel, 2019). This leads to our next hypothesis:

H4.

Identification with the company positively affects corporate reputation.

Exploring whether corporations should take a stand on socially relevant issues, Xu et al. (2022) found that consumer identification with the company mediates the relationship between individuals' psychological distance and behavioral intentions toward the company. Overall, the identification of a consumer and the company increased with higher levels of perceived psychological distance from the company, resulting in increased buycott and boycott intentions. The authors conclude that other mechanisms must play a role in explaining these possibly competing results (Xu et al., 2022). We argue that the psychological distance from the company that was found to affect the identification with a company might be related to psychological states that result from (cognitively) engaging with a CSA message. Therefore, we posit that individuals might feel less identified with the company when they perceive a message as persuasive and thus feel threatened in their personal freedom to decide or act, i.e. their level of reactance increases. In other words, when individuals feel companies want to convince them through their CSA message to follow the cause or support the company, they may react with reactance and disidentify with the company as a counteract. This leads to the following hypothesis:

H5.

Reactance negatively affects the identification with the company.

All hypotheses are summarized in our model (see Figure 1).

Method

Experimental design and procedure

Our research aimed to determine the effects of two CSA message strategies on corporate reputation on an individual's perception level. Thus, we used a single-factor (CSA messages without/with a call to action) between-subjects experimental design. We conducted an experimental online survey in the German-speaking region of Switzerland between September and October 2021. The survey participants were randomly assigned to the experimental (CSA message with a call to action) or the control group (CSA message without a call to action) upon clicking on the survey link via a built-in randomizing function provided by the survey tool (LimeSurvey).

The study received ethical approval from a large university in Switzerland, including the use of real brands for the stimulus material. The chosen brands were contacted via e-mail with the option to opt out of the study, and participants were debriefed about the fictional content of the stimulus messages. At the beginning of the survey, participants were informed about the study and had to give informed consent. Subsequently, participants were asked about their issue alignment with gender equality and the predispositions toward the companies presented later on in the stimulus material. Next, we showed the participants the stimulus containing a manipulated tweet depicting a CSA message with or without a call to action. Upon reading the stimulus material, we asked the participants questions to assess their identification with the company, reactance level, assessment of corporate reputation and demographics. Moreover, we included items to check the manipulation.

Sample

The initial sample consisted of 215 participants. In the data cleaning process, those participants who did not complete the manipulation check were excluded, as well as outliers. The final sample consisted of 172 participants from the German-speaking part of Switzerland, wherein 83 people were in the experimental group and 89 were in the control group. The sample comprised 46.5% female, 51.7% male and 1.7% non-binary participants. People aged 21 to 86 completed the survey (M = 42.9). 50.0% of participants held at least a university degree at Bachelor's level. The participants' political leanings were assessed on a scale from strongly left-leaning (1) to strongly right-leaning (5), which resulted in a mean of 2.52, indicating a reasonably politically balanced but slightly left-leaning. The participants were recruited via an email containing a generic description of the study and a link to the survey on LimeSurvey. No incentive was used for recruitment.

A randomization check was conducted to assess the balance of the participants across the two groups (experimental and control groups). Chi-square tests were conducted, revealing no significant association for gender (χ²(2) = 0.73, p = 0.69), completed education level (χ²(6) = 2.85, p = 0.72), or income (χ²(5) = 3.80, p = 0.58). An independent samples t-test was conducted which revealed no significant difference in the mean age between the groups, t(177) = 0.46, p = 0.65. These findings provide evidence of successful randomization, i.e. participants' gender, level of completed education, income and age were evenly distributed across the experimental conditions.

Stimulus and pretest of the stimulus

Considering how ubiquitous strategic communication of companies via their social media channels is, we used a manipulated Twitter post as the experimental stimulus material. To maintain external validity, two actual and well-known companies were chosen as the authors of the tweets. The companies Dior and Armani were picked as a result of the pretest, considering their widespread notoriety and range of products that could appeal to female and male customers alike, as well as their well-documented issue alignment with previous campaigns surrounding gender equality. Dior's creative director, Maria Grazia Chiuri, has gone public with her feminist views, going as far as creating the well-known “We Should All Be Feminist” t-shirt for the brand. Giorgio Armani of the namesake brand has also gone on record stating his favorability for gender-neutral fashion that doesn't objectify women's bodies, thus taking a public feminist stand (cf. Di Lello, 2019; Yotka, 2016). The plausibility of the stimulus messages being associated with previous communication from the chosen brands is expected to create external validity, thus reducing confounding factors in the survey results.

The tweet in the control group was: “A study shows that if men received their yearly pay on January 1st, women would have to work for free until February 20th (source: bit.ly/EqualPay). On this #EqualPayDay, our company renews its commitment to give men and women equal pay for equal work [1].” In contrast, the tweet containing a call to action (experimental group) stated: “To take a stand against the gender pay gap, all female employees of our company will receive a #FeminismBonus to mark this year's #EqualPayDay. Join us and sign the petition on bit.ly/FeminismBonus to encourage other companies to do the same!” The messages were pretested (N = 10) to ensure the manipulation would be perceived as intended. To avoid any bias provoked by social cues, the number of likes, comments and retweets was kept consistent across all stimulus materials and reflected the typical range of likes, comments and retweets found on actual posts of the companies.

Measures

Independent variable

CSA messages

The independent variable of our study is CSA messages as represented in the stimulus. The variable was coded as a dummy variable (CSA message without a call to action = 1; CSA message with a call to action = 2). For this study, a CSA call-to-action message is conceived as a company's message that includes a statement asking the broader public to support their CSA effort by taking action, i.e. signing a petition.

Dependent variable

Corporate reputation

We conceived corporate reputation as individuals' perception of a company and its actions. As previous literature suggested, the construct was measured using six items reflecting the overall perception of a company in terms of attitudes towards the company, satisfaction with the company's outcomes, willingness to recommend the company and the perceived trustworthiness of the company (Lim and Young, 2021; Walsh and Beatty, 2007). For instance, we included the item “I am very positive towards the company.” All items were rated on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree; M = 2.69, SD = 0.41, Cronbach's Alpha = 0.85). For an overview of the measures, see Table 1.

Mediating variables

Identification with the company

Identification with a company is defined as the extent to which an individual can identify with a company, building on the perceived congruence between corporate efforts and the individual's values (cf. Du et al., 2007). Following prior studies (Dodd and Supa, 2015; Hydock et al., 2020), the construct was assessed based on different aspects, including a perceived fit of the company with one's values, the favorability of the company and agreement with the values represented by the company and measured with five items. For instance, we included the item “I fully share the values that the company represents.” All items were rated on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree; M = 2.38, SD = 0.78, Cronbach's Alpha = 0.79).

Reactance

This study conceives reactance as a resistance to perceived persuasion intent. For measuring the individuals' reactance, we used three items that were slightly adapted from previous literature (Åkestam et al., 2017), including the item “The statement of the company annoys me.” All items were rated on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree; M = 2.33, SD = 1.10, Cronbach's Alpha = 0.78).

Control variables

Disposition towards these companies

Considering that the stimulus used for the experimental design contained actual and well-known companies, the disposition of the participants towards these companies was measured as a control variable before the stimulus was shown with one item for each company (“How do you regard [company]?”) using a 5-point semantic differential scale ranging from “very negatively” to “very positively” (cf. McDermott and Lachlan, 2020; Dior: M = 3.23, SD = 0.87; Armani: M = 2.95, SD = 0.88).

Issue alignment

Additionally, we measured the issue alignment of participants concerning the gender pay gap as a control variable with five items. For this, we asked the participants to assess the issue of gender equality regarding its benefits, desirability, personal significance, appeal and value (cf. Zaichkowsky, 1985) before the stimulus was shown. All items (e.g. “The issue of gender equality is crucial for me.”) were rated on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree; M = 3.02, SD = 0.42, Cronbach's Alpha = 0.81).

Political leaning

The participants' political leaning was assessed with one question, asking, “Where on the political spectrum would you most likely place yourself?” and their responses reached from strongly left-leaning (1) to strongly right-leaning (5) (M = 2.52., SD = 0.87).

Company

Finally, we included a dichotomous variable indicating the company included in the stimulus (1 = “Dior,” 2 = “Armani”). Since the results of a multivariate analysis of variance (MANOVA) with “company” and “CSA message” as independent variables and identification with company, reactance and corporate reputation as dependent variables indicated no significant (p > 0.05) mean differences between the companies or the interaction between companies and the CSA message regarding the dependent variables, we did not include the variable “company” in the final mediation analysis.

Manipulation check

Two items were used to check whether the manipulation was successful. The participants were asked to assess whether the message they saw was “controversial” and “risky” on a 5-point Likert scale (“1” strongly disagree, “5” strongly agree). The results of the independent t-test indicate that the mean values of these items varied significantly (t(170) = -3.61, p < 0.001) between the control and the experimental groups (“controversial”: M_control = 2.39, M_exp = 3.14; “risky”: M_control = 1.88, M_exp = 2.67), suggesting that the manipulation was successful.

The correlation matrix can be found in Table 2.

Results

To investigate whether CSA communication with a call to action compared to CSA messages without such a call affects corporate reputation and what role the identification with the company and individuals' reactance play, a mediation was performed using SPSS 27, PROCESS Model 6 (Hayes, 2013), with 5,000 bootstrap samples. The outcome variable was corporate reputation, and the independent variable was the CSA message (without/with a call to action). In the mediation analyses, we controlled for predispositions toward the company, issue alignment, gender and political leaning.

H1 posited a direct effect of the independent variable on corporate reputation. Results show that the independent variable does not significantly predict corporate reputation (b = −0.03; p = 0.55; see Figure 2), indicating no direct effect of the CSA message (1 = without a call to action, 2 = with a call to action) on corporate reputation. Thus, we need to reject H1. Hypothesis H2 posited that the CSA message with a call to action leads to higher levels of reactance than the one without. Our analysis showed a significant positive effect of the independent variable on reactance (b = 0.69; p < 0.001). These results imply that the CSA message with a call to action led to higher reactance than the CSA message without, supporting hypothesis H2.

Furthermore, H3 suggested a positive and direct effect of the independent variable on the identification with the company. However, our analysis showed a significant negative effect (b = −0.20, p = 0.047). Thus, we need to reject H3. This means that the CSA message with a call to action compared to a message without a call to action decreases the identification with the company. Hypothesis H4 posited a positive effect of the identification with the company on corporate reputation. The analysis showed that identification with the company significantly positively predicts corporate reputation (b = 0.26; p < 0.001). This provides support for H4. Lastly, hypothesis H5 assumed that higher levels of reactance negatively affect the identification with the company. Our results indicate a significant negative effect of reactance on the identification with the company (b = −0.29; p < 0.001). Thus, H5 is also supported by our data.

We controlled for predisposition toward the company, political leaning, gender and issue alignment. The results showed that the independent variable's effect on corporate reputation remains significant (p < 0.001) even if including the control variables in the model, explaining 30.5% of the variance of corporate reputation. The mediation model analyses show that gender (b = 0.05; p = 0.38) and the predisposition towards the company (b = −0.03; p = 0.48) did not significantly predict corporate reputation.

However, the mediation analysis indicated that issue alignment predicts corporate reputation significantly and positively (b = 0.16; p = 0.03), but no significant coefficients could be found for issue alignment predicting reactance (b = 0.33; p = 0.14) or the identification with the company (b = −0.08; p = 0.41). Since we could not find a significant direct effect of CSA messages on corporate reputation, the relationship between issue alignment and corporate reputation can be disregarded.

Finally, results show that the control variable, political leaning, significantly positively predicted reactance (b = 0.30; p = 0.001) and corporate reputation (b = 0.06; p = 0.04), whereas political leaning did not significantly predict the identification with the company (b = 0.04; p = 0.55). Since we could not find a significant direct effect of CSA messages on corporate reputation, the relationship between political leaning and corporate reputation can be disregarded. However, the relationship between political leaning and reactance is relevant to this study as it shows that right-leaning individuals showed more reactance to the CSA messages than left-leaning ones. Thus, we ran a moderation analysis to see how political leaning affects the relationship between CSA messages (the independent variable) and reactance (the dependent variable in the moderation analysis). Results indicate no significant interaction effect of political leaning and CSA messages on reactance (b = −0.24; p = 0.19). This means that political right-leaning increases reactance compared to left-leaning but irrespectively of the CSA message.

Overall, the indirect effect of the independent variable through reactance and identification with the company on corporate reputation is significant and negative (indirect effect: −0.052, standard error (SE) = 0.02, lower level of the 95% confidence interval (LLCI): −0.0927 and upper level of the 95% confidence interval (ULCI): −0.0231), indicating that the CSA message with a call to action overall leads to a lower corporate reputation than CSA messages without a call to action.

Discussion

In this study, we investigated the effect of CSA communication on corporate reputation in the case of a company's promotion of gender equality and to what extent reactance and identification with the company influence this effect. Overall, the results demonstrated that CSA messages with a call to action lead to higher reactance of individuals compared to CSA messages without such a call, which in turn lowers the identification with the company and the perceived corporate reputation. Hence, individuals' reactance, i.e. temporary psychological arousal that arises when people feel restricted in their personal freedom (Brehm, 1966), is highly relevant to consider when it comes to CSA messaging. Our results imply that a call to action in a CSA message leads to a more defensive stance toward the received message (cf. Åkestam et al., 2017). Moreover, our study implies that an individual's reactions to restore his or her freedom (i.e. reactance) may be accompanied by disassociation with the message source -- in our example, in the form of a disidentification with the company, which, however, harms the perception of the company.

Previous research already showed that in the case of CSR messages, reactance is a significant factor and is, for instance, increased by the perceived persuasive intention (Bachmann and Ingenhoff, 2016; Viererbl and Koch, 2022) and consequently negatively affects the perception of corporate legitimacy (Bachmann and Ingenhoff, 2016). Our study confirms and extends these findings by demonstrating the effect of the communication message strategy on reactance and indirectly on corporate reputation, in this way, significantly contributing to previous CSA scholarship (Abitbol et al., 2018; Ciszek and Logan, 2018; Eilert and Nappier Cherup, 2020; Rim et al., 2020). Our study indicates that messages with a call to action may be processed differently, i.e. people are cognitively more engaged when processing a message with a call to action compared to those without such a call as individuals and thus may evaluate the message more closely. As Suchman (1995, p. 575) argued, if “an organization seeks protracted audience intervention […] [and to] mobilize affirmative commitments, however, it must also ‘have value’ either substantively, or as a crucial safeguard against impending non-sense.”

Furthermore, our study offers possible explanations for different reactions to CSA, e.g. boycott versus support or negative versus positive perception of a company. Irrespective of the issue alignment, political leaning and the predispositions of the individuals surveyed, the reactance was lower in the case of the CSA message without a call to action and the identification with the company and the perception of it in terms of corporate reputation were higher. Thus, it can be assumed that the mobilizing character of CSA communication in the case of the included call to action is less favorable for corporate reputation. Although our results remained significant when controlling for political leaning, political leaning had a significant effect on the tested model and was found to affect an individual's reactance; people who politically lean more towards right were found to react more with reactance across both the control and the experimental group. This result is in line with previous studies in political communication, which, for instance, found that right-leaning individuals, in comparison to left-leaning ones, resisted messages encouraging support for governmental measures against climate change (Zhou, 2016) and encouraging support for a university health campaign to fight the spread of COVID-19 (Dillard et al., 2023). Also, in the context of corporations, scholars noted that social issues management must consider individuals' political ideology as social issues are often polarizing and connected to political ideology (Coombs and Holladay, 2018). To be perceived as legitimate, companies need to be authentic and compelling in their messages to legitimize their engagement in advocacy and activist efforts (Chatterji and Toffel, 2019). Our result implies that companies need to consider individuals' political ideology when communicating CSA efforts – irrespective of the chosen communication strategy, as it may trigger reactance, which threatens corporate reputation.

Although CSA efforts with a call to action are likely to lead to more significant behavioral change as more consideration is put into it, which would be the goal of CSA in terms of societal change (Eilert and Nappier Cherup, 2020), this seems less applicable when companies are the source of the advocating message. This might be related to the fact that people are more skeptical about companies due to past scandals, misconduct, greenwashing and the like and tend to assume that they act in their self-interest and want to persuade or even manipulate the public.

Implications, limitations and future research

Our results are highly relevant, with companies becoming increasingly involved in political issues and public debate (van der Meer and Jonkman, 2021) due to politicization and globalization (Scherer and Palazzo, 2011). Communicating about political issues, especially taking a stand on political issues, is particularly challenging for companies in an increasingly polarized society and has often led to backlash, boycotts and damage to corporate reputation in practice. This study shows that possible adverse effects of advocating for socio-political issues can be related to reactance. Moreover, the study implies that corporate communication strategies that are less emotionalizing and possibly less activating and instead present corporate engagement in political issues in a more informative and factually objective way are less risky for companies and may even contribute to their reputation.

Our study complements existing studies on strategic communication that examine the effect of (corporate) communication strategies on reactance (e.g. Bachmann and Ingenhoff, 2016) and has significant implications for this research field. Our results can be applied to other communication interventions on issues other than gender equality and possibly to communication interventions or corporate initiatives other than CSA. Furthermore, in an increasingly politicized and polarized society (cf. van der Meer and Jonkman, 2021), our study is of interest to communication practitioners and companies increasingly expected to speak out on social and political issues. Our study suggests that when communicating on a political or social issue, mobilizing communication that includes a call to action should be considered carefully to avoid provoking reactance, which may threaten the identification with the company and, ultimately, the company's reputation.

As with any experimental study focusing on specific effects, our study has limitations that open up avenues for future research. First, our sample consisted of people living in one specific country, which lowers the generalizability of our results. Future research could test our model in other cultural contexts, considering the country's level of gender equality. Further, considering that gender equality is a highly politicized and moral issue, social desirability might have affected how participants responded to the survey. Our choice of equal pay as the central issue of this experiment may further impact the results since we cannot exclude issue-dependent effects. Despite that, we do not expect the effect of call-to-action messages to vary as studies that have centered on different issues, such as the inclusion of transgender people, have previously concluded that CSA has shown to increase the willingness to engage with the company and its cause (Lim et al., 2023) and that people who do not identify directly with the cause will still critically engage with the CSA message (Fried and Opree, 2023). Thus, we expect our findings about the impact of call-to-action messaging to stay consistent across different socio-political issues, which future research could investigate further.

Our stimulus could also represent a possible limitation, as it utilizes real companies that respondents might have preconceived notions of. To counteract this potential issue, we controlled for the predisposition toward the companies to preserve the added value of external validity a real company brings to the stimulus material. Building on an extensive pretest with different brands and considering real-world examples of these brands, we chose Dior and Armani since these two brands are associated with having pro-gender-equality positions. This could have led to individuals perceiving the CSA messages as more credible and authentic than in the case of a company advocating for gender equality that is less associated with promoting gender equality in real life. However, our purpose was to test the different effects of CSA messages with versus without a call to action, and for this purpose, we did not want to create stimulus materials that might have led to more confounding effects (e.g. the perceived credibility or authenticity of the CSA message). Another methodological issue comes with our research design, as the stimulus material was only shown for two companies and on one social media platform, and non-textual cues were excluded from the stimulus, limiting the study's external validity. Different methodologies or methods, including qualitative approaches that explore the CSA strategies for other socio-political issues in more in-depth or mixed-methods designs, would broaden the insight provided by this study.

Finally, we only measured the effect of CSA communication at one point and only manipulated one message strategy. Since reputation is formed over time, longitudinal studies would allow us to measure the long-term effects of CSA on corporate reputation. Also, longitudinal research designs would help measure changes in attitudes and behavioral intentions concerning socio-political issues – after all, CSA is about contributing to society and social change.

Figures

Figure 1

Hypothesized model on the effect of CSA messages on corporate reputation

Figure 2

Path model of the hypothesized model

Table 1

Variables with items and sources

Variables and items		Sources
Identification with the company*
idc_1	I fully share the values that the company represents	Based on Dodd and Supa (2015) and Hydock et al. (2020)
idc_2	I am very familiar with the company
idc_3	The company suits me very well
idc_4	I find the company very likeable
idc_5	I think it is a very good idea to buy products from this company
Corporate reputation*
corprep_1	The company is very trustworthy	Based on Lim and Young (2021) and Walsh and Beatty (2007)
corprep_2	I am generally very satisfied with the company
corprep_3	I would recommend this company to my friends and family
corprep_4	I have a very positive attitude towards the company
corprep_5	I am very critical of the company. [reverse]
corprep_6	I feel that the company is particularly self-serving. [reverse]
Reactance*
react_1	The statement of the company annoys me	Based on Åkestam et al. (2017)
react_2	I am very annoyed by the fact that the company has commented on a political issue
react_3	I find the company's statement extremely manipulative
Issue alignment*
issueal_1	The issue of gender equality is very important to me	Based on Zaichkowsky (1985)
issueal_2	The issue of gender equality is very boring for me. [reverse]
issueal_3	The issue of gender equality means nothing to me. [reverse]
issueal_4	Gender equality is extremely desirable to me
issueal_5	The topic of gender equality is very important to me
Disposition towards these companies**
dispo_comp	How do you regard [company]?	Based on McDermott and Lachlan (2020)
Political leaning***
polleaning	Where on the political spectrum would you most likely place yourself?

Note(s): All items were asked in German language

* These items were judged on a Likert scale ranging from “strongly disagree” (1) “strongly agree” (5)

** This question could be responded to on a semantic differential scale ranging from “very negatively” (1) to “very positively” (5)

*** This question could be responded to on a scale ranging from strongly left-leaning (1) to strongly right-leaning (5)

Source(s): Table by authors

Table 2

Correlation matrix of the variables

	1	2	3	M	SD
1. Reactance	1			2.33	1.10
2. Identification with the company	−0.49*	1		2.38	0.78
3. Corporate Reputation	−0.15*	0.53**	1	2.69	0.41

Note(s): N = 172, *Correlation is significant (p < 0.05) **Correlation is significant (p < 0.001) (two-sided); 1 “strongly disagree” to 5 “strongly agree”

Source(s): Table by authors

Note

1. For this article, the tweets were translated into English; in the study, they were in German.