When Corporate Social Responsibility Messages Enter the News: Examining the Effects of CSR-Framed News on Product Purchasing Intentions and the Mediating Role of Company and Product Attitudes

Abstract

Journalists are frequently reporting about new and innovative products in news articles. Oftentimes, journalists use specific company and product information provided by enterprises that highlight a corporation’s social responsibility (CSR) activities. Based on framing theory and theory of reasoned action, an experiment was conducted to examine how types of CSR-framed news (either highlighting ethical, ecological, or philanthropic responsibilities of a company) affect news recipients’ product purchasing intentions. A mediation analysis showed that CSR-framed news indirectly and positively affected individuals’ purchasing intentions via company and product attitudes. No relevant differences could be detected for the specific type of CSR frame. Implications of the results are discussed.

Keywords:

• Corporate Social Responsibility (CSR)
• News Framing Effects
• Public Relations
• Purchasing Intention

Previous research shows that organizations frequently use corporate social responsibility (CSR) strategies in their communications with various audiences and stakeholders (Aguinis & Glavas, 2012; Byun & Oh, 2018; Du, Bhattacharya, & Sen, 2010; Lin-Hi & Blumberg, 2018). That is, organizations regularly use CSR messages in campaigns and other forms of internal and external communication processes (Aguinis & Glavas, 2012) to positively affect an organization’s image, reputation, and credibility (Kim, 2017; Lin-Hi & Blumberg, 2018). In this respect, results showed that “87% of American consumers are likely to switch from one brand to another (price and quality being equal) if the other brand is associated with a good cause, an increase from 66% since 1993” (Du et al., 2010, p. 9). What CSR actually is has been widely debated, and various CSR definitions have been established (Carroll, 1999). Broadly defined, CSR can be understood as a “commitment to improve well-being through discretionary business practices and contributions of corporate resources” (Du et al., 2010, p. 8).

Previous research has extensively examined the effects of CSR communication strategies on various outcome variables, including influences on company and product attitudes as well as purchasing intentions (Aguinis & Glavas, 2012; David, Kline, & Dai, 2005; Kim, 2017; Peloza, Ye, & Montford, 2015). Overall, CSR strategies have been found to positively affect stakeholder perceptions and behaviors (Aguinis & Glavas, 2012; Du et al., 2010).

However, several research gaps persist. First, while research has extensively examined CSR communication effects, e.g., in advertising and in internal and external organizational communication contexts (Aguinis & Glavas, 2012), relatively little is known about the particular effects of CSR-framed messages embedded in ordinary news media coverage (for exceptions, see Byun & Oh, 2018; Flöter, Benkenstein, & Uhrich, 2016; Kim & Lee, 2015; Murray & Vogel, 1997; Sen & Bhattacharya, 2001). That is, how are individuals affected when CSR messages are embedded in an ordinary news story (out of the control of an organization’s public relations practitioners), and do such messages have comparable effects on news consumers similar to campaign messages (Du et al., 2010)? Second, what role does the specific type of CSR framing play in this context (Halme & Laurila, 2009)? Does it matter if a frame emphasizes, e.g., ethical or philanthropic responsibilities of a company (Carroll, 1999), and do particular CSR frames influence news consumers’ product and company attitudes as well as purchasing intentions in different ways (David et al., 2005; Lin-Hi & Blumberg, 2018)? Third, previous research has largely focused on the United States (Murray & Vogel, 1997; Sen & Bhattacharya, 2001), while other cultural contexts (e.g., European countries; see Flöter et al., 2016) have been neglected. Thus, it remains unclear if the effects of CSR messages embedded in the news can be generalized to other cultural contexts.

To close these research gaps, we conducted an experiment in Germany. More precisely, we examined how CSR-framed messages (i.e., ethical, ecological, philanthropic responsibilities of a company) embedded in online news articles affected individuals’ product purchasing intentions. Moreover, aiming at better understanding the underlying processes of CSR effects, we examined if and how company and product attitudes mediated this process.

Effects of CSR-Framed News

Journalists regularly select certain news frames and highlight particular aspects of a topic (while neglecting others) and in this way promote particular issue interpretations and (moral) evaluations (Entman, 1993; von Sikorski & Schierl, 2012). However, news framing is not only relevant from a journalism perspective but also from a public relations perspective. It is well documented that public relations practitioners are highly influential when it comes to shaping news content, assuring “that the story is slanted or framed in a way that is consistent with the source’s preferred framing (i.e., how a client would like to have its story told)” (Hallahan, 1999, p. 228, emphasis in original). Especially, earned media content (information provided by public relations practitioners and disseminated by the news media for free, Stephen & Galak, 2012) in the form of news coverage about a certain company and/or product is usually perceived to be more credible compared to information presented in the advertising section of a newspaper (Vercic & Vercic, 2016). This results in a win-win situation. A company can publicly promote a product in a subtle and credible way for free (not counting the PR costs), while a news outlet receives free (and in the best case) interesting content for its readers.

Research shows that companies actively manage their media image using particularly designed CSR strategies. More socially responsible firms, in turn, receive more media coverage (Cahan, Chen, Chen, & Nguyen, 2015). Furthermore, findings from a correlational study suggest that news media coverage about a company’s CSR activities is “positively associated with shareholder value and improved future operating performance” (Byun & Oh, 2018, p. 68). These findings are supported by experimental data revealing that individuals (i.e., managers) were positively affected by company descriptions when these descriptions consisted of particular CSR information (Murray & Vogel, 1997). Yet the aforementioned experiment used textual descriptions (vignettes) instead of authentic news articles to study respective effects. Sen and Bhattacharya (2001) used a mock news article, and their findings generally supported those of Murray and Vogel (1997). Yet, they did not account for individuals’ product attitudes as a potentially important mediator. Based on these results and other findings generally showing a positive influence of CSR activities on the reputation of companies (e.g., Brown & Dacin, 1997; Lin-Hi & Blumberg, 2018), we formulated our first hypothesis (Figure 1 shows the full theoretical model):

H1:

Exposure to a CSR-framed news article will positively affect participants’ company attitudes compared with the identical news article without a CSR frame (control condition).

Figure 1. Study theoretical model. Examples of the three different articles used as stimulus is available online at https://OSF.IO/9ZQ57/.

In contrast, what type of CSR message elicits the most positive effects remains rather unclear and is rather context sensitive (Becker-Olsen, Cudmore, & Hill, 2006; Du et al., 2010). Thus, we were not able to deduce a hypothesis and formulated a research question:

RQ:

 What type of CSR message will affect participants’ company attitudes in the most positive way?

Previous research in marketing clearly showed that individuals’ perceptions of a company can affect their attitudes toward products manufactured by the company (e.g., Brown & Dacin, 1997; Peloza et al., 2015). That is, individuals who perceive a company in a rather positive way will also evaluate products of the particular company more positively (Brown & Dacin, 1997; Peloza et al., 2015). Accordingly, we expected that this should also be the case in a news media context. H2 reads as follows:

H2:

More positive company attitudes will result in more positive product attitudes.

Furthermore, in line with the theory of reasoned action (Fishbein & Ajzen, 1975) and previous meta-analytical findings (e.g., Sheppard, Hartwick, & Warshaw, 1988; see David et al., 2005), we expected that more positive product attitudes will positively predict purchasing intentions of a product presented in an ordinary news media context. Thus, H3 reads as follows:

H3:

More positive product attitudes will result in increased purchasing intentions.

Method

Participants and Pocedure

We conducted a randomized 1 × 4 controlled laboratory experiment (between-subjects design) at a large German university. Similar to previous research (Sen & Bhattacharya, 2001), we used a student sample (N = 127, age M = 24.41 years, SD = 3.52, age range 18–34 years). We selected male participants only because the product we focused on was men’s running shoes.

Quality news outlets regularly report about new and innovative products like sports shoes, e.g., the Washington Post’s Innovations Section (Overly, 2016). We constructed an authentic news article based on original articles. We used a fictitious brand (Vibe) and photo of a running shoe (that participants were unfamiliar with) allegedly produced by a new United States-based company. The article was assigned to the German quality newspaper Frankfurter Allgemeine Zeitung, which regularly reports about product innovations (the stimulus material is available at: https://osf.io/9zq57/). Similar to previous research (e.g., Sen & Bhattacharya, 2001), we used a fictitious brand and product. We aimed at establishing a cause-effect relationship and strived for avoiding other influences such as particular prior attitudes that may have undermined the purpose of the study.

In his seminal work, Carroll (1999) identified four different types of corporate responsibilities (economic, ethical, legal, philanthropic). More recently, other scholars have pointed out that consumers (with different cultural backgrounds) tend to differentiate between different types of organizational responsibilities and do not perceive all types of responsibilities as equally important (Maignan, 2001) or—as in the case of “economic” and “legal” responsibilities—as part of the CSR construct. Furthermore, ecological responsibilities have become more important (Aguinis & Glavas, 2012). Halme and Laurila (2009) have conceptually differentiated three types of CSR: philanthropy (actions outside a company’s core business, e.g., donations), corporate responsibility integration (actions close to existing business, e.g., upgrades of a company’s environmental, ethical/social performance), and corporate responsibility innovation (e.g., enlarging core business, developing new products/services to alleviate social/environmental problems). In the present study, we focused on the first two types. Based on Halme and Laurila’s (2009) CSR types and conceptualizations by Carroll (1999), Sen and Bhattacharya (2001), as well as Hallahan’s (1999) conception of framing in public relations research, we deduced three central types of CSR frames from the literature: ethical, philanthropic, and ecological frames.

Group 1 participants (n = 34) were exposed to the news article without a specific CSR frame (control condition). Group 2 participants (n = 30) read the identical article additionally consisting of an ethical CSR frame stating that Vibe shoes are manufactured under fair working conditions and that Vibe is concerned about paying workers fair wages. Group 3 participants were exposed to an otherwise identical article with an ecological CSR frame (n = 32) highlighting Vibe’s ecological engagement and its environmentally friendly manufacturing processes. Group 4 participants (n = 31) received the article with a philanthropic CSR frame highlighting Vibe’s social engagement and its commitment in sports leisure projects for disadvantaged children. Participants read the article at their own reading rate and were told that the article was published recently. Subsequently, participants completed a questionnaire (see Measurement section) and were then thanked and debriefed.

Measurement

All items were measured on a 7-point Likert-type scale. Company attitude was measured with nine items (e.g., innovative–not innovative, likeable–not likeable, trustworthy–not trustworthy, 1 = very negative, 7 = very positive, M = 3.43, SD = 0.99, Cronbach’s α = .86) based on Sen and Bhattacharya (2001). Product attitude was measured with 11 items (e.g., likeable–unlikeable, high quality–low quality, useful–useless, 1 = very negative, 7 = very positive, M = 3.43, SD = 0.99; α = .83) based on Spears and Singh (2004). Purchasing intention was measured using the following item based on Spears and Singh (2004) from 1 (very unlikely) to 7 (very likely): “If you would need a new pair of running shoes would you be willing to buy the Vibe V14?” (M = 2.39, SD = 1.69).

Also, the questionnaire contained several control variables. That is, we measured individuals’ interest in sports, how often they actively go for a run, their general CSR support, and general media trust to ensure that all effects on purchase intention were assessed independently of these variables.

Finally, a manipulation check was conducted. Participants were asked how much information they detected regarding ethical (M = 1.61, SD = 2.24), ecological (M = 2.14, SD = 2.10), and philanthropic aspects (M = 1.40, SD = 2.08) in the article. ANOVAs with Bonferroni-adjusted post hoc analyses revealed that individuals exposed to ethical information perceived significantly more ethical information (M = 5.00, SD = 1.26) compared to participants in the other groups, F(3, 124) = 111.70, p < .001, η²_part = .40. Ecological information (versus the other groups) resulted in significantly higher perceptions of ecological information (M = 4.53, SD = 1.91), F(3, 124) = 49.61, p < .001, η²_part = .17, and philanthropic information (versus the other groups) in significantly higher perceptions of philanthropic information (M = 4.13, SD = 2.14), F(3, 124) = 55.29, p < .001, η²_part = .49. The results revealed a successful manipulation.

Data Analysis

A mediation model with two mediators using the PROCESS macro in SPSS was conducted. Experimental condition was dummy coded with the control group as reference group. Participants’ company attitude was modeled as the first mediator and product attitude as the second mediator. Purchasing intention was entered as the dependent variable. Therefore, it was possible to examine the respective news article’s effect on purchasing intention and if company and product attitude mediated this relationship (Figure 1). The 95% bias-corrected bootstrap confidence intervals based on 5,000 bootstrap samples were used for statistical inference of indirect effects. Also, the aforementioned control variables were entered as covariates.

Results

First, it was investigated if the news article with a CSR message affected individuals’ company attitudes (H1). That was the case. Ethical (b = 6.34, SE = 2.11, p = .003), ecological (b = 4.39, SE = 2.01, p = .031), and philanthropic CSR-framed news (b = 4.46, SE = 2.04, p = .032) positively and significantly affected participants’ company attitudes compared with the control condition (Table 1). To answer our RQ, we adjusted our dummy coding (not shown in Table 1). First, we used the ethical condition as the reference group. The results neither showed significant effects for the ecological condition (b = 1.94, SE = 2.09, p = .356) nor for the philanthropic condition (b = 1.87, SE = 2.13, p = .381). Second, with the ecological condition as the reference group (versus the philanthropic, b = –0.07, SE = 2.07, p = .972 and the ethical condition (b = –1.94, SE = 2.09, p = .356), we found no significant effects as well. Third, no significant effects emerged when the philanthropic condition served as the reference group (see aforementioned results).

Table 1 Ordinary Least Squares Path Analysis, Unstandardized Coefficients, N = 127

Variables	Company Attitude		Product Attitude		Purchasing Intention
	b	SE	b	SE	b	SE
Ethical CSR Frame	6.34**	2.11	3.76	1.83	–0.33	0.68
Ecological CSR Frame	4.39*	2.01	2.58	1.71	0.14	0.63
Philanthropic CSR Frame	4.46*	2.04	4.89**	1.74	–0.31	0.66
Company Attitude			0.68***	0.08	0.04	0.04
Product Attitude					0.14***	0.03
General CSR Support	–0.07	0.16	–0.08	0.13	0.24*	0.11
Interest in Sports	1.31*	0.59	0.81	0.52	0.05	0.18
Active Sports Participation	–0.12	0.41	–0.05	0.34	–0.21	0.13
Media Trust	1.94	0.55	–1.04*	0.48	0.05	0.18
Adj. R^2	.16		.42		.29

***p < .001; **p < .01; *p < .05.

Next, we examined if more positive company attitudes resulted in more positive product attitudes (H2). This was the case. In line with H2, the results revealed that more positive company attitudes positively and significantly predicted product attitudes (b = 0.68, SE = 0.08, p < .001). Furthermore, the examination showed (supporting H3) that more positive product attitudes positively and significantly predicted purchasing intension (b = 0.14, SE = 0.03, p < .001). Also, a direct and positive effect of the philanthropic CSR condition on product attitudes could be detected (b = 4.64, SE = 1.76, p = .009). Moreover, the results showed a significant indirect effect of exposure to CSR-framed news on purchasing intention through company and product attitude (b = 0.64, CIs [0.23, 1.25]). In total, the predictors explained 16% of the variance of company attitudes, 42% of the variance of product attitudes, and 29% of the variance on purchasing intentions.

Discussion

The aim of the present study was to examine how CSR-framed news affects news recipients’ purchasing intentions of a newly introduced product. Corroborating and extending previous results (e.g., Byun & Oh, 2018; Sen & Bhattacharya, 2001), the findings showed that CSR-framed news positively increased male German readers’ purchasing intentions of a presented product (running shoes) via company and product attitudes. No differences emerged regarding the type of CSR framing, indicating that a company’s involvement in CSR activities mattered to readers; however, the specific kind of activity did not matter. The results show that the identical product may benefit from different types of CSR-framed news as long as (one may argue) individuals perceive the social issue to fit the product/company (Du et al., 2010).

Importantly, the identical article (control) without any CSR information did not result in more positive company and product evaluations, although the company/product details—as well as the visual presentation of the product—were identical across conditions. Thus, the CSR frames positively affected male individuals’ company and product attitudes and eventually—in line with the theory of reasoned action (Fishbein & Ajzen, 1975; Sheppard et al., 1988)—increased purchasing intentions. In contrast to Sen and Bhattacharya’s (2001) findings (who studied CSR effects in connection with inkjet printers), our results showed that individuals’ general CSR support positively affected purchasing intentions for running shoes; CSR support may have interacted with the specific type of product. Furthermore, the results showed that media trust was a negative predictor of product attitude. One explanation for this result is that individuals with a low level of media trust may have processed the article more critically and may have tended to perceive the product presentation as a form of sponsorship or product placement. This may have triggered reactance and then negatively affected the product itself (Boerman, Van Reijmersdal, & Neijens, 2012). However, this is just speculation, and more research is needed.

Moreover, the present study examined the effects of CSR-framed news in a European context (Germany), indicating that respective effects extend to different (non-U.S.) cultural settings (Sen & Bhattacharya, 2001) and countries with different (compared to the United States) media systems (Hallin & Mancini, 2004).

The results have several implications. First, companies can systematically influence news consumers’ purchasing intentions when providing respectively framed information to journalists (e.g., on the company’s Web page). Providing such information increases the likelihood that journalists report about a specific company and that the particular CSR information is actually used in an article (e.g., Vercic & Vercic, 2016). Second, CSR information is relevant for journalists because it offers new angles of reporting and provides additional information to readers (Cahan et al., 2015). Yet merely adopting CSR frames from public relations practitioners can also be considered problematic. News consumers may consider this form of reporting as uncritical and enterprise-friendly, which may have negative consequences on recipients’ news source evaluations. Future research should examine this assumption in depth.

There are several limitations to this study. First, similar to previous research (Sen & Bhattacharya, 2001), we used a student sample (males only); although young males are certainly part of the target group of the product examined (running shoes), future studies should try to replicate the present findings with older and female samples to test if the effects can be generalized. Second, similar to previous studies (Murray & Vogel, 1997), we used a fictitious company/product; future studies should test the role of CSR-framed news using real and other companies/products (e.g., using field experiments). Third, future studies should examine the effects of different CSR frame types in depth, examining interactions with specific prior consumer attitudes. Fourth, the mean value of purchasing intention for the fictitious product was rather low (M = 2.39). A one-way t-test revealed that it significantly differed from the midpoint of 4 on the 7-point scale, t(126) = 4.61, SD = 1.69, p < .001; this should be taken into account when interpreting the results, and future research—as mentioned previously—should try to replicate the findings using different (real) products. Fifth, more comparative approaches should be realized to test if Americans (e.g., versus Germans) are generally more/less supportive of CSR information presented in the news and how this affects CSR-framed information. Sixth, a direct effect of the philanthropic frame on product attitudes emerged; thus, future research should test additional mechanisms that might explain effects on purchasing intention. Seventh, we used a quality news outlet; future studies should try to replicate the effects with less credible (boulevard) news media.

Conclusion

The present study shows that CSR-framed news can significantly affect recipients’ purchasing intentions. When news consumers are exposed to a corporation’s CSR activities via news frames in the media, this information positively affects individuals’ company attitudes, which in turn results in more positive product attitudes and eventually in increased product purchasing intentions.

Disclosure statement

No potential conflict of interest was reported by the authors.