«13TH INTERNATIONAL PUBLIC RELATIONS RESEARCH CONFERENCE “Ethical Issues for Public Relations Practice in a Multicultural World” Holiday Inn ...»
Consumers generally form evaluations or reactions to organizations’ motives for engaging in CSR (Du et al., 2007; Klein & Dawar, 2004; Sen et al., 2006). In this light, attribution theory help to explain the way a message is interpreted. Incorporating an attribution model into our theorizing means that when consumers become aware of the organization’s social involvement (e.g., through various communication outlets), they are likely to interpret the message by way of some very specific attribution effects. Previous work in this area (Campbell & Kirmani, 2000; Ellen et al., 2006) has confirmed that consumers’ evaluations of a firm are influenced by these attributions. While several previous investigations have examined the moderating role of these attributions, one recent study by Vlachos et al. (2009) treated attributions as directly influencing consumer responses. For the current study we take a slightly different track, treating the attributions as outcomes rather than antecedents or intervening variables.
Operationalized by Ellen et al. (2006), the consumer attributions used for this study were initially classified as: values-driven, stakeholder-driven, strategic-driven, and egoistic-driven.
Ellen and her colleagues based these attribution effects on two dimensions of CSR: (1) selfcentered and (2) other-centered. For example, if a firms is acting in its own self-interest or with profit-centric motivations (i.e., strategic-driven), attitudes toward the firm will likely diminish or negatively manifest; however if a firm is genuinely socially motivated (i.e., values-driven) or is focused on the society/community (i.e., stakeholder-driven), attitudes toward the organization are likely enhanced (Becker-Olsen et al., 2006). While these four dimensions demonstrated signs of internal consistency and face validity, we adopted the views of Walker et al. (2010) and excluded the egoistic-driven construct. This construct designates a very negative judgment of the firm’s activities but if the negative wording is removed, it closely aligns with the strategic-driven dimension. Recent work by Vlachos et al. (2009) revealed that these attributions influenced consumer trust, repeat purchase intentions, and recommendation intentions. However, we sought (by way of this study) to understand the front-end of the model. Specifically, we examined the effects of information source and valence on (1) self-centered attributions (i.e. strategic-driven), 650 (2) other-centered attributions (i.e. values-driven), and (3) a combination of both (i.e.
Values-Driven Attributions Values-driven attributions indicate that the consumer perceives the corporate action to be altruistic and socially motivated. As discussed above, consumers will likely be more skeptical of information released directly from the company itself (Artz & Tybout, 1999). As such, we predict consumers will question if in fact a company is acting in a truly altruistic manner when publishing their CSR activities internally. Conversely, when exposed to externally published information, individuals have less reason to be skeptical of the message and its contents.
Therefore, Hypothesis 1 is formally proposed:
H1: Values-driven attributions will be lower when CSR information is published internally as opposed to when it is published externally.
We also predict that a proactive CSR initiative (an initiative that precedes information regarding any social misdeeds of a company) will be perceived as being more altruistic in nature than a reactive CSR initiative (an initiative that is performed because of the negative actions of a company). We predict that consumers will perceive this optional giving as being genuinely altruistic. On the other hand, consumers may be skeptical of the underlying motives of a company engaging in a CSR initiative to assuage their social misdeeds. Considering these
points, Hypothesis 2 is formally proposed:
H2: Values-driven attributions will be higher when the message contains proactive rather than reactive CSR information.
Stakeholder-Driven Attributions Firms that are stakeholder-driven work to attend to the interests of all persons coming in contact either directly or indirectly with the firm. Source of information may affect individuals’ stakeholder-driven attributions because consumers may perceive that the content of internally published information is intended for a firm’s stakeholders; the internal communication mechanisms of a firm after all, are geared toward the firm’s stakeholders (Hutton et al., 2001).
However, externally published information is not always intended solely for stakeholders involved in the story. For example, readers of newspapers often read interesting articles that have little if any personal relevance. We predict therefore, that individuals who are exposed to internally published CSR information will believe the content is intended for the firm’s stakeholders and thus will perceive the corporate motives to be more stakeholder-driven.
Formally, Hypothesis 3 is proposed:
H3: Stakeholder-driven attributions will be higher when CSR information is published internally as opposed to when it is published externally.
651 When individuals perceive a firm’s CSR actions as being stakeholder-driven they believe the firm is performing the action because important stakeholders expect the company to act ethically and responsibly. Increasingly, stakeholders are placing more emphases on proactive CSR (Becker-Olsen et al., 2006). Customers, employees, suppliers, and society in general are becoming more accustomed to firms acting in a proactive fashion regarding their social responsibilities. With that said, research has shown that all stakeholders hold firms accountable when they act in an unethical manor (Vlachos et al., 2009). For example, if a company is caught polluting the environment stakeholders will demand the company to rectify their wrongdoings.
While stakeholders may have heightened expectations for firms to be proactive, we predict that all stakeholders will expect a firm to rectify their social wrongdoings. We therefore hypothesize that proactive CSR will be perceived as being less stakeholder-driven than reactive CSR. Thus,
Hypothesis 4 is proposed:
H4: Stakeholder-driven attributions will be lower when the message contains proactive rather than reactive CSR information.
Strategic-Driven Attributions Strategic-driven attributions are placed toward a firm’s CSR initiatives when an individual believes the firm is engaging in the activities to accomplish specific business goals.
As discussed earlier, information published directly by a company is often perceived to be selfserving in nature (Morsing & Schultz, 2006). Considering this we predict that individuals
exposed to internally published CSR information will perceive the firm’s motives to be strategicdriven. Therefore, Hypothesis 5 is proposed:
H5: Strategic-driven attributions will be higher when CSR information is published internally as opposed to when it is published externally.
A reactive CSR strategy involves a company spontaneously reacting to some negative event (Wagner et al., 2009). Firms that deploy a reactive CSR strategy simply respond when negative information is made public. On the other hand, firms that deploy a proactive CSR strategy expend a great deal of recourses to plan and implement such a strategy. Considering the characteristics of these two strategies we predict that individuals exposed to proactive CSR information will perceive the company to be more calculating in their CSR endeavors and
therefore will attribute the action as being strategically driven. Thus, Hypothesis 6 is proposed:
H6: Strategic-driven attributions will be higher when the message contains proactive rather than reactive CSR information.
a 2 (CSR valence: proactive vs. reactive) x 2 (information source: internal vs. external) between subjects experimental design. Following their consent to participate, the subjects received an experimental packet consisting of a questionnaire, instructions, a fictitious press release, and a series of questions in response to the press release. To prevent against demand artifacts (i.e.
subjects predicting the objective of the experiment) subjects were informed that the goal of the study was to assess the effectiveness of press releases in disseminating information; study participants were not informed about the experimental manipulations (Sawyer, 1975).
The press release consisted of information about a recent CSR initiative of a fictitious company (i.e., Mayetta Food and Beverage Corp). A fictitious company was used to minimize any confounds due to preexisting attitudes toward the firm (Brown & Dacin, 1997). An Internet search was conducted to verify that this company did not exist.
Independent variables. Similar to Wagner et al. (2009), in the proactive condition the press release explained that the company voluntarily donated money to assist in the aid of an environmental issue. In the reactive condition the company appeared to be irresponsible regarding the same environmental issue and agreed to pay a sum of money to compensate for damages. The information source manipulation was achieved with a header above the press release which stated the following press release was either taken directly from the company itself (i.e. Mayetta Food and Beverage Corp.) or from an external news outlet (i.e. The Associated Press). For priming purposes and gauge whether they successfully processed the manipulation, the subjects were asked what organization published the press release. Of the 230 initial subjects, 215 (93%) successfully processed the manipulation. An analysis of variance (ANOVA) between these two groups indicated no significant difference on any of the three dependent outcomes. Thus, all 230 subjects were retained. Except for the described manipulations, all other aspects of the press release were invariant.
Dependent variables. The dependent measures consisted of three consumer attribution dimensions (i.e., values-driven attributions, stakeholder-driven attributions, and strategic-driven) adopted from the work of Ellen et al. (2006). Each scale consisted of three items which were averaged to form the operational measures used in the analysis. As in previous administrations, the three scales were internally consistent (α=.86values-driven; α =.85stakeholder-driven; α =.85strategic-driven) according to the criteria suggested by Lance et al. (2006). The items used in this study can be found in the appendix.
Results and Discussion
A series of one-way ANOVAs were run to assess the effects of CSR valence and information source on the attributions individuals assigned to both proactive and reactive CSR initiatives. Following is a discussion of the results of these ANOVAs separated by each attribution and listed in order of our hypotheses.
Values-driven attributions. To test Hypothesis 1, a one-way ANOVA with source of information as the between subject factor was analyzed. Contrary to our prediction, source had no effect on values-driven attributions (F(1,228)=.358, p.10). To test Hypothesis 2, a one-way ANOVA with valence as the between-subjects factor and values-driven attributions as the dependent variable was analyzed. The valence effect was significant (F(1,228)=141.94, p 653.001). Individuals exposed to the proactive CSR initiative perceive the act to be more valuesdriven (M=4.87) than individuals exposed to the reactive CSR initiative (M=3.07). This result indicates that consumers perceive firms engaging in proactive (versus reactive) CSR as acting in more of an altruistic manner. Therefore as predicted, firms that voluntarily participate in CSR are perceived by consumers as being values-driven. Table 1 contains the cell means and F-test scores for the values-driven attributions.
While source of the press release was predicted to significantly influence values-driven attributions, there is a possible explanation for the non-significant result. The proactive manipulation was extremely strong and as a result had a large influence on individuals’ perception of a values-driven attribution (F(1,228)=141.94). This strong manipulation may have overpowered any affect the source of the information may have had. Additionally, the source manipulation was relatively weak (i.e., a header preceding the press release). This particular manipulation may not have been strong enough to account for any additional variance not accounted for by the valence of the information. These equivocal results suggest that valence influenced the values-driven attributions individuals assigned to a firm’s CSR actions.
Consequently, Hypothesis 2 was supported while Hypothesis 1 was not.
Stakeholder-driven attributions. To assess the effect of source of information on stakeholder-driven attributions (Hypothesis 3) a one-way ANOVA was performed. The analysis indicated that source did affect stakeholder-driven attributions (F(1,228)=3.73, p.10).
Individuals exposed to the press release published by the company perceived the action to be more stakeholder-driven (M=4.50) than individuals exposed to a press release published by an external source (M=4.20). Essentially, individuals exposed to internally published information believed the company was participating in CSR because of pressures from relevant stakeholder groups (i.e. customers, employees, and society). Presumably, the perception might be that communication of CSR via internal mechanisms signals to stakeholders that the company is working for them; and they subsequently anticipate some action to be taken. As such, the motives assigned to internal corporate communications are perceived as being more stakeholderdriven than if the information came from an external source.
Hypothesis 4 was again tested using a one-way ANOVA with valence as the betweensubjects factor and stakeholder-driven attributions as the dependent variable. The results of this analysis indicated the effect of valence on the outcome significant (F(1,228)=6.78, p.01). As predicted then, individuals exposed to a reactive CSR initiative perceived the act to be more stakeholder-driven (M=4.55) than those exposed to the proactive CSR initiative (M=4.15).