行政院國家科學委員會專題研究計畫 成果報告
公司社會責任對於消費者購買意願之影響
研究成果報告(精簡版)
計 畫 類 別 ： 個別型 計 畫 編 號 ： NSC 100-2410-H-004-170- 執 行 期 間 ： 100 年 08 月 01 日至 101 年 07 月 31 日 執 行 單 位 ： 國立政治大學企業管理學系 計 畫 主 持 人 ： 張愛華 計畫參與人員： 博士班研究生-兼任助理人員：曾祥景 博士班研究生-兼任助理人員：楊偉顥 報 告 附 件 ： 出席國際會議研究心得報告及發表論文 公 開 資 訊 ： 本計畫涉及專利或其他智慧財產權，1 年後可公開查詢中 華 民 國 101 年 10 月 31 日
中 文 摘 要 ： 本研究的目標有兩個:檢驗消費者對企業社會責任形象的前置 因素及檢驗企業社會責任形象如何影響消費者對公司產品的 購買意圖。明確而言，我們檢驗企業社會責任活動的正反性 與類型之交互作用對消費者企業社會責任形象的影響；我們 也進一步拆解企業社會責任形象對消費者購買意圖的效果。 我們使用 2*3+1 的受試者間實驗設計來驗證研究假設。結果 指出企業社會責任活動正反性的效果會隨著不同類型的社會 責任活動而有所不同。明確而言，對於利他型社會責任活動 而言，企業社會責任形象的正反性對企業社會責任形象的影 響是呈現曲關係。也就是說，正面的企業社會責任活動會大 大改善企業社會責任形象，而負面的企業社會責任活動則不 會使企業社會責任形象低於中立組(沒有提及社會責任資訊的 組)。從另一個方面而言，對於規範型與平衡需求型的社會責 任活動，企業社會責任正反性對企業社會責任形象是呈現線 性關係。企業如果做了正面的社會責任活動，他的社會責任 形象會高於中立組，再高於做了負面的社會責任的情況。本 研究進一步驗證了消費者對公司認同在企業社會責任形象與 購買意圖監所扮演的中介角色。 中文關鍵詞： 企業社會責任活動正反性、企業社會責任類型、利他型社會 責任、規範型社會責任、平衡型社會責任、企業社會責任形 象、消費者對公司認同
英 文 摘 要 ： The objectives of this study are two folds: to examine the determinants of consumer＇s corporate social image (CSR image) and to investigate how CSR image works to influence consumer purchase intention. Specifically, we examine the interaction of valence and type of corporate CSR initiatives on consumer＇s CSR images； we further delineate the path of effects of CSR image on consumer purchase intention. A 2*3+1 between-subject experimental design was used to validate the hypotheses. The results indicated that the effects of CSR initiative valence vary with the type of initiatives. Specifically, for altruistic type of CSR, CSR initiative valence has a quadratic influence on CSR image. Positive initiative can greatly improve CSR image, while negative initiative does not result in a CSR images worse than that of a neutral initiative. On the other hand, for both the normative and balancing types of CSR, CSR initiative valence has a linear influence on CSR image. Positive
initiative results in best CSR image, followed by neutral initiative, then negative initiative. This research further validates the mediating role of consumer-company identification on the relationship between CSR images and purchase intention.
英文關鍵詞： CSR initiative valence, CSR initiative types, Altruistic type CSR, Normative type CSR, Balancing need type CSR, CSR images, C-C identification
行政院國家科學委員會補助專題研究計畫
■成果報告
□期中進度報告
（計畫名稱）公司社會責任對於消費者購買意願之影響
計畫類別：■個別型計畫 □整合型計畫
計畫編號：NSC 100 － 2410 － H － 004 － 170
－
執行期間： 2011 年 8 月 01 日至 2012 年 7 月 31 日
執行機構及系所：政治大學企業管理所
計畫主持人： 張愛華
共同主持人： 無
計畫參與人員：曾祥景、楊偉顥
成果報告類型(依經費核定清單規定繳交)：□精簡報告 ■完整報告
本計畫除繳交成果報告外，另須繳交以下出國心得報告：
□赴國外出差或研習心得報告
□赴大陸地區出差或研習心得報告
■出席國際學術會議心得報告
□國際合作研究計畫國外研究報告
處理方式：
除列管計畫及下列情形者外，得立即公開查詢
□涉及專利或其他智慧財產權，■一年□二年後可公開查詢
中 華 民 國 101 年 10 月 31 日
目錄 中文摘要與關鍵詞………II 英文摘要與關鍵詞………...III 報告內容:………1 前言與研究目的………...1 文獻探討………...2 研究方法………...6 結果………...9 討論與結論……….13 參考文獻……….16 附錄一……….20 附錄二……….22 計畫成果自評………..……….23 附錄………..……….24 出席國外研討會心得……….24
中文摘要 本研究的目標有兩個:檢驗消費者對企業社會責任形象的前置因素及檢驗企業社會責任 形象如何影響消費者對公司產品的購買意圖。明確而言，我們檢驗企業社會責任活動的 正反性與類型之交互作用對消費者企業社會責任形象的影響；我們也進一步拆解企業社 會責任形象對消費者購買意圖的效果。我們使用 2*3+1 的受試者間實驗設計來驗證研究 假設。結果指出企業社會責任活動正反性的效果會隨著不同類型的社會責任活動而有所 不同。明確而言，對於利他型社會責任活動而言，企業社會責任形象的正反性對企業社 會責任形象的影響是呈現曲關係。也就是說，正面的企業社會責任活動會大大改善企業 社會責任形象，而負面的企業社會責任活動則不會使企業社會責任形象低於中立組(沒 有提及社會責任資訊的組)。從另一個方面而言，對於規範型與平衡需求型的社會責任 活動，企業社會責任正反性對企業社會責任形象是呈現線性關係。企業如果做了正面的 社會責任活動，他的社會責任形象會高於中立組，再高於做了負面的社會責任的情況。 本研究進一步驗證了消費者對公司認同在企業社會責任形象與購買意圖監所扮演的中 介角色。 關鍵詞: 企業社會責任活動正反性、企業社會責任類型、利他型社會責任、規範型社會 責任、平衡型社會責任、企業社會責任形象、消費者對公司認同
Abstract
The objectives of this study are two folds: to examine the determinants of consumer’s corporate social image (CSR image) and to investigate how CSR image works to influence consumer purchase intention. Specifically, we examine the interaction of valence and type of corporate CSR initiatives on consumer’s CSR images; we further delineate the path of effects of CSR image on consumer purchase intention. A 2*3+1 between-subject experimental design was used to validate the hypotheses. The results indicated that the effects of CSR initiative valence vary with the type of initiatives. Specifically, for altruistic type of CSR, CSR initiative valence has a quadratic influence on CSR image. Positive initiative can greatly improve CSR image, while negative initiative does not result in a CSR images worse than that of a neutral initiative. On the other hand, for both the normative and balancing types of CSR, CSR initiative valence has a linear influence on CSR image. Positive initiative results in best CSR image, followed by neutral initiative, then negative initiative. This research further validates the mediating role of consumer-company identification on the relationship between CSR images and purchase intention.
Keywords: CSR initiative valence, CSR initiative types, Altruistic type CSR, Normative type CSR, Balancing need type CSR, CSR images, C-C identification
INTRODUCTION AND PURPOSES
The importance of corporate social responsibility (CSR) has been increasing.
Consumers have rising expectations toward companies. Nowadays, many consumers are switching towards more socially and environmentally responsible products and service; they expect that companies not only offer products of high quality and low price, but also do something beneficial for the society (Freestone and McGoldrick 2008; Handelman and Arnold 1999). Therefore, as market competition becomes more intense and product homogeneity increases, CSR is another battle ground for competitive advantage (Du et al. 2007). Increasing number of companies has therefore invested in CSR activities. CSR activities bring the following benefits to companies. First, they result in better company stock performances (Drumwright 1994; Klein and Dawar 2004). Second, consumers tend to have more positive attitude toward the CSR-doing companies and their products (Brown and Dacin 1997; Creyer and Ross 1997). Third, companies engaged in CSR can get better reputation (Fambrun and Shanley 1990). Fourth, consumers are more loyal to companies engaged in CSR (Bhattacharya and Sen 2003; Du et al. 2007). Fifth, CSR-doing companies have better resistance to negative publicity (Bhattacharya and Sen 2003).
Based on the aforementioned benefits CSR bring to companies, a series of studies have begun to discuss why consumers have positive responses when they know companies engage in CSR activities. Most past studies investigated the question from the perspective of social identity theory (e.g., Ahearne et al. 2005; Berger et al. 2006; Bhattacharya and Sen 2003; Currás-Pérez et al. 2009; Du et al. 2007; Lichtenstein et al. 2004). According to the theory, consumers have positive responses because these CSR-doing companies can satisfy their self-definitional needs. Specifically, CSR image in consumers’ mind may foster consumer’s self-relevant appraisal. The appraisal attracts consumers to the company identity and further facilitates consumer-company identification (C-C identification). C-C identification brings positive outcomes to companies (e.g., better company evaluation and purchase intention) (Bhattacharya and Sen 2003). However, few studies investigated the antecedents of CSR images and the way the antecedents influence consumer’s CSR images. Past studies indicated that the possible antecedents are CSR initiatives that companies engaged in (e.g., Sen and Bhattacharya 2001). The basic assumption is that companies CSR initiatives will enhance their CSR images, and then consumers will reward companies by purchasing more products from these companies (Du et al. 2007; Levy 1999; Lichtenstein et al. 2004; Luo and Bhattacharya 2006). This assumption implies that doing CSR is better than non–engaging or acting in the opposite way. However, this assumption has not been explicitly tested and may contradict to consumer experiences. For example, knowing that a company does not produce harmful products will not lead to a better
CSR images because this CSR initiative is considered to be a must-do; yet, if a company does produce harmful products, its CSR image will be greatly damaged. On the other hand, while consumers will hold better image for the company engaging in philanthropy donation, their company images usually not become deteriorated if learning that a company does not engage in charity donation. Therefore, it seems that other than the valence of CSR initiatives, i.e., engaging in CSR (positive CSR initiatives) versus violating CSR (negative CSR initiatives), the type of CSR should also be considered. The extant research has yet systematically investigated the differential effects of various types of CSR initiatives on consumers. The work of Brunk and Biumelhuger (2011) is a pioneer research in this field of inquiry. They propose a CSR typology which classifies CSR activities into three types: altruism, norms, and balancing need. They further depict the patterns of influence on CSR perception associated with the three types of CSR activities. However, their arguments are based on the findings of a small sample, exploratory study and have not been empirical tested. This study intends to broaden the scope of investigation and examine the differential impacts corresponding to different types of CSR initiates. Specifically, the interaction effect of CSR valence and CSR types on consumer CSR images will be empirically tested. In addition, we extend the past identity-based research stream to study the linkage between CSR images, C-C identification, and purchase intention. The former research has yet addressed the construct of CSR image and its relationship with C-C identification, which is worth study due to the important role of image in consumer response toward a company or brand. The conceptual framework is presented in Figure 1.
Figure 1 Conceptual Framework
LITERATURE REVIEW
Origin of CSR Images and CSR Types
The conceptualization of CSR has not reached a consensus among researchers. Sethi (1975) presented a concept similar to CSR-corporate social performance. Corporate
Valence of CSR (Positive/Negative /Neutral) Type of CSR (Altruism/Norms/ Balancing Need) CSR Images C-C Identification Purchase Intention
social performance can be divided into three levels- social obligation, social response, and social responsibility. Social responsibility was defined in a way that in addition to lawful and economic obligations, companies have to pursue objectives that are beneficial in the long run for the society. Carroll (1979, 1991) indicated that CSR can be categorized into economic, lawful, ethical, and discretionary responsibilities. Van Marrewijk (2003) presented sustainable development model and indicated that CSR was composed of three levels-economic, social, and environmental from bottom to top. Singh et al. (2008) divided CSR into business, ethics and society. However, these researchers define and categorize CSR from the perspective of companies, not from consumers.
Some researchers indicated the perceptual differences exist between consumers and companies on ethical/unethical behaviors (e.g., Galavielle 2004; Crane 2005; Brunk 2010). Consumers perceive these behaviors in a multiple ways and these behaviors influence their attitudes toward the companies and purchase behaviors (Brunk 2010). Sometimes, companies consider their behaviors ethical (positive CSR initiatives) but consumers think these behaviors unethical (negative CSR initiatives). The perceptual differences may impair companies (e.g., consumers’ boycott against Nike’s products after the child labor scandal). Consumer’s perceived CSR valence rather than the objective valence of a firm’s CSR initiative determines the outcomes (e.g., company evaluation, product association, and purchase intention) (Becker-Olsen, Cudmore, and Hill 2006; Marin, Ruiz, and Rubio 2009). Hence, consumer’s perception of CSR initiatives (CSR images) should be considered when the relationship between CSR initiative valence and outcomes is examined. Studies were few on the effect and formation of CSR images resulting from CSR initiatives. This study seeks to fill the knowledge gap.
Past studies indicated that on the whole, the relationship between CSR initiative valence and CSR image is linear. That is, companies engaged in positive (negative) CSR initiatives enhance (lower) their CSR images relative to non-engagement status (Du et al. 2007; Levy 1999; Lichtenstein et al. 2004; Luo and Bhattacharya 2006). We therefore propose the following hypotheses:
H1: When companies have positive CSR initiatives, CSR images are better than neutral initiative. When companies have neutral CSR initiatives, CSR images are better than negative initiatives.
Yet, for specific CSR initiative, its valence may have nonlinear effects on CSR images. Brunk and Blümelhuber (2011) presented three impact dimensions to further divide CSR initiatives according their impacts on CSR perception. The first dimension
is altruism. If companies conduct altruistic CSR initiatives such as philanthropy, they do something that goes beyond business responsibilities. Consumers are happy to see companies doing altruistic CSR initiatives; however, they do not expect companies to conduct altruistic CSR initiatives. If companies conduct altruistic CSR initiatives, they may largely improve CSR images. If not, their CSR images are not worsened. The second dimension is norms. Normative CSR initiatives are deemed minimum requirements that companies must adhere to. If companies do not observe these minimum requirements, their CSR images tend to decrease drastically; yet, if companies meet these minimum requirements, their CSR images are predicted to increase marginally. The aforementioned two dimensions have nonlinear patterns of impact. The third dimension is balancing need. Balancing need CSR initiatives reflect companies’ efforts to balance society and business interests and have linear impact on CSR images. Positive initiatives tend to increase CSR images and negative initiatives lower. We therefore propose the following hypotheses:
H2: The valence of altruistic CSR initiatives has a quadratic relationship on consumer CSR image. The difference of CSR images between positive and neutral initiatives is greater than that between neutral and negative initiatives.
H3: The valence of normative CSR initiatives has a quadratic relationship on consumer CSR image. The difference of CSR images between negative and neutral initiatives is greater than that between positive and neutral initiatives.
H4: The valence of balancing-need CSR initiatives has a linear relationship on consumer CSR image. The difference of CSR images between positive and neutral initiatives is similar to that between neutral and negative initiatives.
CSR Images and C-C Identification
CSR images reflect the organization’s status and activities related to its perceived societal obligation (Brown and Dacin 1997, p.68). Other than CSR images, company capability (CA) associations are the other dimensions of corporate associations. CA associations refer to the corporation’s expertise in producing and delivering products and services, which are mainly economic and technical in nature. Past studies indicated the importance of CSR images in building consumer-company identification. First, compared with CA associations, CSR images are more enduring and unique (Sen and Bhattacharya 2001). This means CSR images are more useful to companies in building long-term relationship with consumers. Second, many empirical studies indicated that CSR images are likely to trigger consumers’ positive responses (Berens et al. 2005; Du
et al. 2007; Lichtenstein et al. 2004; Sen and Bhattacharya 2001). Marin and Ruiz (2007) further indicated in their empirical research that the influences of CSR images are greater than CA associations on C-C identification. In Marin and Ruiz’s study, the variance of CA associations is small because of the fierce competition. Hence, CSR images are another battleground for competitive advantage. This study adopts the aforementioned viewpoint and only examines the influence of CSR images on C-C identification.
C-C identification is a cognitive state resulting from the process of self-categorization. Members in an organization can develop a collective identification, a sense of community (Ashmore et al. 2004). Members of a certain organization (e.g., a corporation) perceive that they are similar to other in-group members and dissimilar to out-group members (Bergami and Bagozzi 2000). Although consumers do not have a formal relationship with corporations, they still can identify with a corporation like an employee (Bhattacharya and Sen 2003). Social identity theory contends that people do not have to interact intensively or feel strong interpersonal ties to perceive themselves to be part of a group (Brewer 1991). Therefore, consumers can identify with a corporation in certain circumstances.
Bhattacharya and Sen (2003) developed a Consumer-Company identification framework based on theories of social identity and organization identification. Specifically, consumers consider CSR images to be a basis for self-relevant appraisals. Consumers do these appraisals to satisfy self-definitional needs. These appraisals can be divided into three types—self-enhancement, self- similarity, and self-distinctiveness. Self-enhancement indicates that if companies do positive CSR initiatives, they are doing good deeds. Consumers can enhance their self-concept by connecting with those companies. Self-similarity indicated that consumers may appraise the degree of similarity between their own personal identities and company identities. If a match exists, consumers can confirm their self-concepts through identifying with the company. For example, consumers whose personal identity contains altruism may identify with companies doing positive CSR initiatives because personal identity is similar to company identity. Self-distinctiveness indicates that if CSR initiatives are of much value to consumers, they tend to identify with the company. Consumers find a company’s identity attractive through these self-relevant appraisals and further identify with a company. We therefore propose the following hypothesis:
H5: CSR images positively influence C-C identification.
C-C Identification and Purchase Intention
group, they give positive evaluations and participate actively in the group activities because by doing so they can strengthen their self-concept (Tajfel and Turner 1979). Based on the theory, consumers may also have in-group favoritism toward the company they identify with. They show in-group favoritism in a direct way through consuming products and services of the identified companies. Consumers think their companies are likely to prosper if they consume more and the companies’ success is their own success. Past studies further indicated that C-C identification positively influences purchase intention because they have higher company evaluation and commitment (Currás-Pérez 2009). However, some researchers indicated in their empirical results that C-C identification influences purchase intention directly, not through the mediation of affective variables (e.g., Ahearne et al. 2005; Cornwell and Coote 2005). Other researchers indicated empirically that both direct and indirect effects are possible (Currás-Pérez et al. 2009). We therefore propose the following hypothesis:
H6: C-C identification mediates the relationship between CSR image and purchase intention.
METHOD
Selection and Categorization of CSR Initiatives
This study categorizes CSR initiatives based on the valence of CSR initiatives into positive, negative, and neutral, and based on the impact of CSR initiatives on CSR perception into altruism, norms, and balancing need. Hence, we have six types of CSR initiatives plus a neutral initiative.
Brunk (2010) indicated that the origins of CSR perception are multiple and can be divided into six main domains: consumer, employee, environment, overseas community, local community and economy, and business community. Each main domain contains six sub-domains. Mishra and Suar (2010) developed a five dimensional scale (employee, customer, investor, community and environment, and supplier). Both Brunk (2010)’s six sub-domains and Mishra and Suar (2010)’s five-dimension scale indicated examples of CSR initiatives. This study thus selected CSR initiatives referring to Mishra and Suar (2010)’s scale and Brunk (2010)’s sub-domains. A total of 38 initiatives were selected, which includes seven initiatives for consumer, employee, local community and economy, and business community domains, respectively, six initiatives for environment domain, and four initiatives for domain of overseas community. All initiatives were shown in appendix 1. These initiatives are used to discern three types of CSR impact dimensions and the procedure is addressed in the next paragraph.
previously identified CSR initiatives into three types (altruism, norms, and balancing need). One hundred and thirty five valid responses were analyzed. Participants filled out a questionnaire with three sections. The first and second sections seek to describe CSR initiatives in a positive or negative manner (38 initiatives in each section). Participants were asked to check their feelings about each initiative on a scale with anchors ranging from favorable, indispensable, no feeling, tolerable, to unfavorable. The third section is demographics.
Brunk and Blümelhuber (2011) divided CSR activities into altruism, norms, and balancing need. The three categories correspond to three of six quality dimensions presented by Kano et al. (1984). Kano et al. (1984) indicated six types of quality dimensions (i.e., attractive, one-dimensional, reverse, indifference, must-be, and questionable). If a product has attractive dimension, consumers tend to feel satisfied; if not, consumers do not feel dissatisfied. If a product has one-dimensional dimension, consumers tend to feel satisfied; if not, consumers tend to feel dissatisfied. If a product has must-be dimension, because it is the minimum requirement, consumers tend not feel satisfied; if not, consumers tend to feel dissatisfied. The definition of altruistic CSR initiatives corresponds to attractive quality, normative CSR initiatives to must-be quality, and balancing need CSR initiatives to one-dimensional quality. Berger et al. (1993) and Kano et al. (1984) presented a classification table to classify business activities into six quality dimensions. Because there is a correspondence between CSR type and quality dimensions, this study therefore refers to the classification table and classifies CSR initiatives into altruism, norms, and balancing need. Below is an example to illustrate the classification scheme. For a CSR initiative, prohibiting hiring child labor, to be classified as a normative initiative, the following two conditions are met. First, consumers feel the implementation of initiative is reasonable. Second, consumers feel dissatisfied if the company does not implement the initiative.
For each CSR initiatives, we calculated the number of respondents categorized into attractive quality dimension, must-be quality dimension, and one-dimensional quality dimension. Other quality dimensions were not considered for further analysis in this study. If a CSR initiative has the highest numbers of participants on a certain quality dimension, the CSR initiative is categorized into that dimension. For example, if most respondents consider CSR initiative possess must-be quality dimension, the CSR initiative is classified as a Normative CSR initiative. The results of classification are shown in the Appendix 1. Another sample of 182 consumers was used for cross-validation. 30 of the 38 CSR initiatives had the same classification results, achieving a 79% coincidence rate. The classifications are considered stable.
Experimental Stimuli
The scenario for the experiment describes positive or negative CSR initiatives by a company. These initiatives are one of three types: altruism, norms, or balancing need. The focal company was chosen from hypermarket industry. We chose a hypermarket because most consumers have shopping experiences in this type of stores. The name of the hypermarket is fiction because the use of real brand names may introduce brand-relevant effects. For each type of CSR initiatives, three initiatives of that type were used to describe the store. These initiatives were chosen to fit in the industry nature/practices and were put to a pretest to verify their CSR valences. Positive CSR initiatives describe CSR initiatives in an ethical manner; and negative CSR initiatives depict activities in an unethical manner. Also, a neutral group was created and this group describes in detail product lines sold in the hypermarket that are unrelated to CSR. For each scenario, a brief introduction of the store was offered to increase the realism and the lengths of descriptions are equal. The resulting seven scenarios were compiled in the Appendix 2.
Participants and Design
An online experiment was conducted. The link to the website for experiment was posted on Facebook pages on hypermarkets (e.g., Carrefour, Costo, Greant) and on discussion board of Taiwan’s largest Bulletin Board System on hypermarkets. Two hundred and seventy three consumers having shopping experience in a hypermarket participated in the main experiment. Some data was deleted based on the following rules. First, if consumers filled out the questionnaires in the same positions or in a careless way, their answers were discarded. Second, if consumers failed to survive one of our six manipulation check items, their answers were not used. The resulting 253 valid data was used for analysis. Sample characteristics were as follows. Female consumers (58.8%) are more than male consumers. The age range was mostly 21-40 years old (85.5%). Other than students (52.4%), office workers were included (46.6%) and had wide range of occupations (manufacturing, business, education, and liberal profession). A 2(CSR initiative Valence: positive/negative) 3(type of CSR initiatives: altruistic/normative/balancing need) plus 1(neutral group) between-subject design was used. The dependent variables were CSR images, C-C identification, and purchase intention. Demographic variables (gender, age, occupation, education level, monthly income, and marital status) were used as covariates.
Procedure
Participants were randomly assigned to one of seven experimental conditions once they clicked the link to the online experiment. Participants first read a scenario describing
initiatives of a hypermarket and were asked to complete a questionnaire with three sections. These three sections were in different pages. The first section is items on CSR images, C-C identification, and purchase intention. The second section contains items of manipulation check. Participants were asked to recall whether the store has done the following initiatives. Response scale for the groups of positive and negative CSR initiatives had two options (true/false). Response scale for the groups of neutral CSR initiatives had one more option (don’t know). The online questionnaire was set in a manner that consumers cannot go back to the previous page. The third section was on demographics. After consumers completed all questions, they were thanked and had opportunities to leave their e-mail addresses if they want to learn the results of the study.
Measures
CSR Images. CSR images were measured by five items from previous scales of CSR
images and CSR associations (Alcañiz et al. 2010; Marin and Ruiz 2006). The items are as follows. “X (company name) is highly concerned with the society”, “X fulfills its social responsibilities”, “X puts something back into society”, “I think that X acts with society’s interest in its mind”, “X acts in a socially responsible way”. Items in previous scales that address a specific type of CSR were removed or modified into general orientation because in this study we have three types of CSR instead of one specific CSR initiative in the previous study .
C-C Identification. C-C identification was measured by four items from previous scale
(Currás-Pérez et al. 2009). “The way I am fit in with what I perceive X to be”, “I am similar to what I think X represents”, “I am similar to what I perceive X to be”, “The image I have of X overlaps with my self-images” .
Purchase Intention. Purchase intention was measured by three items from previous
scale (Currás-Pérez et al. 2009). The items are as follows. “It is very likely that I will buy X”, “I will purchase X the next time I need a product”, “ I will definitely try X”
.
For the three measures mentioned above, seven-point Likert scales were used with anchors ranging from strongly disagree to strongly agree.
RESULTS
Manipulation Check
CSR initiatives under different types of CSR initiatives were mostly processed as expected. The success rate of manipulation check is within .85~.93. For example, under negative altruistic CSR initiatives, all participants indicated that the company did not engage in the following activities: developing or offering better new products, making effort to offer innovative services, proactively developing products to improve
environment, offering products that can reduce environmental pollution, helping communities through philanthropic and educational donation, and investing in resource to help develop local communities.
The Valence and Type of CSR Initiatives on CSR Images
This study first tests whether CSR initiative valence positively impacts CSR images. The main effect of CSR initiative valence on CSR images was significant across CSR types (F(2,250)=3.15.8, p<.001). Specifically, positive initiatives has better CSR images than neutral initiatives (Mpo.=5.35, Mneu.=3.5; t(157)=13.56, p<.001).
Neutral initiatives has better CSR images than negative initiatives (Mneu.=3.5,
Mne.=2.09; t(171)=10.61, p<.001). Hence, H1 is supported.
This study further examines the interaction between CSR types and CSR initiative valence on CSR image. Because we want to know whether the influence of CSR initiative valence on CSR image is linear or quadratic under three types of CSR, prior to the analysis, a measure of change of CSR image was created. For each experimental group, participants’ score of CSR image was subtracted from the mean score of CSR image of the neutral initiative (3.5). The absolute values of change of CSR image (ACCSR for short) reflect the difference of CSR images between positive and neutral initiatives or the difference of CSR images between neutral and negative initiatives. This absolute value was used to test hypotheses 2-4. All covariates were not significant and not included in the subsequent analysis (ps>.10). 3 2 ANOVA indicated a significant two-way interaction between the valence of CSR initiatives and the type of CSR initiatives on ACCSR (F(2,168)=5.30, p<.006). Therefore, the main effects of the valence of CSR initiatives on ACCSR were examined under different types of CSR. The results were shown in Figure 2. All cell means were shown in Table 1.
Altruism. The main effect of CSR initiative valence on ACCSR was significant
(F(1,47)=21.67, p<.001). The CSR image difference between positive and neutral initiatives was greater than that between neutral and negative initiatives (2.23 v.s. 1.31). Test for polynomial contrast indicated that CSR initiative valence had a quadratic relationship on CSR image (p<.002). The main effect of CSR valence on CSR image was significant (F(2,125)=75.87, p<.001). Specifically, positive
initiatives were larger than neutral initiatives (Mpo.=5.51, Mneu.=3.5; t(125)=10.04,
p<.001). Neutral initiatives were larger than negative initiatives (Mneu.=3.5,
Mne.=2.45; t(125)=4.57, p<.001).From the above information concerning quadratic
relationship and mean difference, positive CSR initiatives can largely improve CSR images and negative CSR initiatives did not seriously impair CSR images. Hence, H2 was supported.
Norms. The main effect of CSR initiative valence on ACCSR was not significant
(F(1,57)=.11, p>.74). The CSR image difference between positive and neutral initiatives was similar to that between neutral and negative initiatives (1.75 v.s. 1.68). Test for polynomial contrast indicated that CSR initiative valence did not reveal a significant quadratic relationship on CSR image (p>.59), but have a significant linear relationship on CSR image (p<.001). The main effect of CSR valence on CSR image was significant (F(2,135)=94.70, p<.001). Specifically, positive initiatives result in better CSR images than neutral initiatives (Mpo.=5.14, Mneu.=3.5; t(135)=7.98,
p<.001). Neutral initiatives result in better CSR images than negative initiatives
(Mneu.=3.5, Mne.=1.94; t(135)=8.72, p<.001), contradicting to the prediction of
non-significant difference of CSR images. From the above information concerning linear relationship and mean difference, positive CSR initiatives seem to improve CSR images and negative CSR initiatives did impair CSR images. Hence, H3 was not supported.
Balancing need. The main effect of CSR initiative valence on ACCSR was not
significant (F(1,64)=3.79, p>.06). The CSR image difference between positive and neutral initiatives was similar to that between neutral and negative initiatives (1.90 v.s. 1.57). Test for polynomial contrast indicated that CSR initiative valence did not reveal a significant quadratic relationship on CSR image (p>.08), but a significant linear relationship (p<.001). The main effect of CSR valence on CSR image was significant (F(2,142)=142.99, p<.001). Specifically, positive initiatives were larger than neutral initiatives (Mpo.=5.35, Mneu.=3.5; t(142)=11.19, p<.001). Neutral
initiatives were larger than negative initiatives (Mneu.=3.5, Mne.=2.02; t(142)=8.83,
p<.001). From the above information concerning linear relationship and mean
difference, positive CSR initiatives seem to improve CSR images and negative CSR initiatives did impair CSR images. Hence, H4 was supported.
Table 1. Cell Means for All Experimental Conditions
CSR Type Positive Negative
Altruistic 5.51 (1.08) 2.45 (.81) Normative 5.14 ( .73) 1.94 (.92) Balancing Need 5.35 ( .75) 2.02 (.61) Note. The cell mean for the neutral group is 3.5 (.89). The cell standard deviations are shown in the parentheses.
CSR images-CC identification-purchase intention link
The results of a series of regression models indicated as follows (n=253). CSR images positively influenced C-C identification (beta=.78, t(251)=19.97, p<.001) and C-C identification positively influenced purchase intention (beta=.80, t(251)=19.19,
p<.001). Hence, H5 was supported. To test H6, Baron and Kenny (1986)’s
three-stage procedure was used to examine the mediating role of C-C identification between CSR images and purchase intention. The first stage indicated that CSR image positive influenced purchase intention (beta=.76, t(251)=18.78, p<.001). The second stage revealed that CSR image positively influenced C-C identification (beta=.78, t(251)=19.97, p<.001). The third stage indicated that both CSR image (beta=.43, t(250)=7.02, p<.001) and C-C identification (beta=.46, t(250)=7.49,
p<.001) positively influenced purchase intention. Also, the influence of CSR images
on purchase intention was lowered but significant when C-C identification was controlled in the model (from .76 to .43). The results of the mediation analysis indicated that C-C identification partially mediates the relationship between CSR images and purchase intention.
Figure 2. The Interaction of CSR Type and Valence on CSR Image
DISCUSSION AND CONCLUSION
Summary
This study examines the antecedents of CSR images and their patterns of influences. Specifically, we examined the interaction of valence (positive/negative) and type (altruism/norms/balancing need) of CSR initiatives on CSR images. The results confirm our hypotheses on altruistic and balancing need CSR initiatives, but not on normative CSR initiatives. Specifically, for altruistic CSR, CSR initiative valence has a quadratic relationship on CSR images. The mean difference of CSR images between positive and neutral initiatives is larger than that between neutral and negative initiatives. Positive initiatives can greatly improve CSR images and negative initiatives do not hurt CSR images very much. For normative and balancing need CSR, valence of CSR initiatives has a linear effect on CSR images. Positive CSR initiatives have better CSR images than neutral and negative initiatives. Neutral CSR initiatives have better CSR images than negative initiatives. The results on altruistic and balancing need CSR are consistent with the viewpoints of Brunk and Blümelhuber (2011) based on their qualitative study but the result on normative CSR is not.
This study also extends the well-established theoretical and empirical link of CSR initiative-CC identification-purchase intention by studying the role of CSR image in the former part of the path. The results indicate that CSR images positively influenced C-C identification and C-C identification positively influenced purchase intention. The mediating role of C-C identification was examined and the results indicate that C-C identification partially mediates the relationship between CSR
C S R I M A G E CSR Initiative Valence
images and purchase intention. We thus demonstrate a more complete linkage of CSR initiatives, CSR images, C-C identification, and purchase intention.
Theoretical Implications
Our results show a linear relationship between CSR initiative valence and CSR images for normative CSR. However, Brunk and Blümelhuber (2011) found a quadratic relationship in their qualitative research. The inconsistent results of normative CSR initiatives between ours and Brunk and Blümelhuber (2011)’s deserve further discussion. Brunk and Blümelhuber (2011) indicated that when consumers see a company does normative CSR initiatives, CSR images do not improve a lot because these initiatives are minimum requirements. However, our results indicated that engage in normative CSR initiatives can improve CSR images. The inconsistent result could be attributed to different CSR orientations across different cultures. Some researchers contend that people in different cultures have different CSR orientations, which results in different expectation about the appropriate types of CSR initiatives (e.g., Burton et al. 2000; Pinkston and Carroll 1996). For example, American or European people have higher expectation of lawful responsibilities (normative CSR) than China and Hong-Kong people (Burton et al. 2000; Chen 1995). Therefore, it seems that for American or European people, doing normative CSR does not largely improve CSR images because companies are expected to do so. The reverse is true for Chinese or Hong-Kong people. If companies conduct normative CSR initiatives, this exceeds expectation of consumers, and according to the disconfirmation paradigm, this may delight consumers and make consumers deeply impressed. Since this study was conducted in Taiwan, which is culturally similar to China and dis-similar to European countries where the findings of Brunk and Blümelhuber (2011)’s derived, the inconsistent findings may be explained by different CSR orientations across cultures.
The results of mediating analysis indicated that C-C identification partially mediates the relationship between CSR images and purchase intention. This means that CSR images can directly influence purchase intention. This result is consistent with Lichtenstein et al. (2004), but inconsistent with Currás-Pérez et al. (2009) and Currás-Pérez (2009). The different results may be due to the fact that consumers have different levels of cognitive appraisals. Lichtenstein et al. (2004) used mail survey and consumers completed the questionnaires at home. Because consumers are not watched over by researchers or interviewers, some consumers may answer the questionnaire seriously and use more cognitive appraisals. For these consumers, self-relevant appraisals are likely to occur when they see CSR initiatives and C-C identification may fully mediate the relationship between CSR images and purchase
intention. Conversely, other consumers may not use a lot of cognitive appraisals while answering the questions; they instead adopt a heuristic rule: doing CSR initiatives is good, and should be rewarded. For these consumers, CSR images may directly influence purchase intention. This study used a research setting similar to Lichtenstein et al. (2004)’s because consumers are not supervised by others when filling out questionnaire. They may use more or less cognitive appraisals. This means CSR images may directly influence purchase intention or indirectly through C-C identification. On the other hand, consumers in Currás-Pérez et al. (2009) and Currás-Pérez (2009)’s studies may be triggered to do more self-relevant cognitive appraisals. For in Currás-Pérez et al. (2009) study, the test product is cosmetics. The product category is important for self-expression and may trigger consumers’ self-relevant appraisals (Belk 1988). Similarly, in Currás-Pérez (2009)’s study, insurance service is test product category which has high credence quality. Because credence goods have high perceived risk, consumers may do more appraisal concerning the perceived value of the service to them, a self-distinctiveness appraisals. Also, both Currás-Pérez et al. (2009) and Currás-Pérez (2009)’s studies used personal interview as a data collection method. Consumers may be watched over by the interviewer and do more cognitive appraisal when filling out questionnaires. In summary, the research setting (e.g., the product category and data collection method used) may determine whether C-C identification partially or fully mediates the relationship between CSR images and purchase intention. When research setting makes consumers do more cognitive appraisals, C-C identification may play a strong mediator.
Managerial Implications
Because CSR images positively influence purchase intention, managers have to improve CSR images through doing more positive CSR initiatives. Other than doing more positive CSR initiatives, managers have to know the fact that CSR initiatives have three types (altruism, norms, balancing need) and have unequal impacts on CSR images. Based on our findings, the following suggestions are offered to managers. Managers have to ensure that their business activities do not violate normative or balancing need of CSR initiatives to avoid worsen of CSR images. For example, Nike hired child labor and caused severe boycott from consumers again its products. In addition, managers should do more positive altruistic CSR initiatives. Compared to other types of CSR, altruistic initiatives can most effectively improve CSR images. For example, McDonald’s donated some of its profit to build a foundation for the homeless children. The principle for companies to follow is stated below: do positive CSR initiatives, especially altruistic CSR; stay away from negative CSR initiatives of
normative or balancing- need type. The Appendix 1 lists all types of CSR initiatives and can be used as a guide for managers to improve CSR images of their companies.
Conclusion, Limitation, and Future Research
This study contributes to the literature by extending the research stream on CSR images-CC identification-purchase intention link to include antecedents of CSR images. Specifically, we empirically examine the interaction of the valence and type of CSR initiatives on CSR images. Literature on the formation of CSR images from the perspective of consumers is few and is qualitative in nature. Past researches were mostly done in the offline contexts. This study empirically confirms the CSR images-CC identification-purchase intention link in the online context and this result adds external validity. This study contributes to the practice by attempting to classify CSR initiatives according to the CSR valence and impact. The resulting classification can be used for managers to improve their companies’ CSR images.
This study has the following limitations. This study indicated that if a research setting makes consumers do more (less) cognitive appraisals, consumers may perform more (less) self-relevant appraisals based on CSR images. In this manner, C-C identification is likely to (partially) fully mediate the relationship between CSR images and purchase intention. This claim should be further tested by subsequent researchers. Also, this study used an online experiment to test our hypotheses and this may cause concern about data quality. However, reliability coefficients for all constructs are all above .95 and manipulation check is assured. They provide some evidences for data quality. This study attempts to classify CSR initiatives based on valence and impact. However, further empirical studies should be conducted to test the reliability of the classification. This study has manipulation check only for the valence of CSR, not for the type of CSR (i.e., altruism, norms, and balancing need). However, we did a pretest to divide CSR initiatives into three types and the two samples indicate evidences of the stability of the classification. More strict manipulation check should be conducted to ensure validity.
REFERENCE
Alcañiz, E. B., R.C. Caceres, and R.Currás-Pérez (2010)," Alliances between Brands and Social Causes: The Influence of Company Credibility on Social Responsibility Image," Journal of Business Ethics, 96(2), 169-186.
Ahearne, M., C.B. Bhattacharya, and T.W. Gruen (2005), "Antecedents and Consequences of Consumer-Company Identification: Expanding the Role of Relationship Marketing," Journal of Applied Psychology, 90(3), 574-585. Ashmore, R.D., K. Deaux and T. McLaughlin-Volpe (2004), "An Organizing
Framework for Collective Identity: Articulation and Significance of Multidimensionality," Psychological Bulletin, 130(1), 80-114.
Baron, R.M. and D.A. Kenny (1986), "The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations," Journal of Personality and Social Psychology, 51(6), 1173-1182.
Becher-Olsen, K.L., B.A. Cudmore and R.P. Hill (2006), “The Impact of Perceived Corporate Social Responsibility on Consumer Behavior,” Journal of Business
Research, 59(1), 46-53.
Belk, R. (1988), "Possessions and the Extended Self," Journal of Consumer Research, 15(September), 139-168.
Berger, I., R. Blauth, D. Boger, C. Bolster, G. Burchill, and W. DuMouchel (1993), “Kano's Methods for Understanding Customer-Defined Quality,” Center for
Quality Management Journal, 2(4), 3-36.
Berger, I., P. Cunningham and M. Drumwright (2006), "Identity, Identification, and Relationship Through Social Alliances," Journal of the Academy of Marketing
Science, 34(2), 128-137.
Bergami, M. and R. P. Bagozzi (2000), "Self-Categorization, Affective Commitment, and Group Self-Esteem as Distinctive Aspects of Social Identity in the Organization," British Journal of Social Psychology, 39(4), 555-577.
Bhattacharya, C.B. and S. Sen (2003), "Consumer-Company Identification: A Framework for Understanding Consumers' Relationships with Companies,"
Journal of Marketing, 67(April), 76-88.
Brunk, K.H. (2010), "Exploring Origins of Ethical Company/Brand Perceptions: A Consumer Perspective of Corporate Ethics," Journal of Business Research, 63(3), 255-262.
Brunk, K.H. and C. Blümelhuber (2011), "One Strike and Your're Out: Qualitative Insights into the Formation of Consumers' Ethical Company or Brand Perceptions," Journal of Business Research, 64 (2), 134-141.
Brown, T.J. and P. A. Dacin (1997), "The Company and the Product: Corporate Associations and Consumer Product Responses," Journal of Marketing, 61(1),68-84.
Burton, B.K., J.L. Farh, and W. H. Hegarty (2000), “A Cross-Cultural Comparison of Corporate Social Responsibility Orientation: Hong-Kong vs. United States Students,” Teaching Business Ethics, 4(2), 151-167.
Carroll, A.B. (1979), "A Three-Dimension Model of Corporate Performance,"
Academy of Management Review, 4(4), 497-505.
Moral Management of Organizational Stakeholders," Business Horizons, 34(4), 39-48.
Chen, C.C. (1995), “New Trends in Rewards Allocation Preferences: A Sino-U.S. Comparison,” Academy of Management Journal, 38(2), 408-428.
Cornwell, B. and L. Cotte (2005), “Corporate Sponsorship of A Cause: The Role of Identification in Purchase Intent,” Journal of Business Research, 58(3), 268-276.
Crane, A. (2005) "Meeting the Ethical Gaze: Challenges for Orienting to the Ethical Market," In Harrison R, Newolm T, Shaw D. The Ethical Consumer. London: Sage, 173-87.
Creyer, E.H. and W.T. Rose, Jr. (1997), "The Influence of Firm Behavior on Purchase Intention: Do Consumers Really Care About Business Ethics," Journal of
Consumer Marketing, 14(6), 421-432.
Currás-Pérez, R (2009) “Effects of Perceived Identity Based on Corporate Social Responsibility: The Role of Consumer Identification with the Company,”Corporate Reputational Review, 12(2), 177-191.
Currás-Pérez, R., E. Bigné-Alcañiz, and A. Alvarado-Herrera (2009), “The Role of Self-Definitional Principles in Consumer Identification with A Socially Responsible Company,” Journal of Business Ethics, 89(4), 547-564.
Drumwright, M. (1996), "Company Advertising with A Social Dimension: The Role of Non-Economic Criteria," Journal of Marketing, 60(4), 71-87.
Du, S., C.B. Bhattacharya and S. Sen (2007), "Reaping Relational Rewards from Corporate Social Responsibility: The Role of Competitive Positioning,"
International Journal of Research in Marketing, 24(3), 224-241.
Fombrun, C. and M. Shanley (1990), "What's in A Name? Reputation Building and Corporate Strategy," Academy of Management Journal, 33(2), 233-258.
Freestone, O. and P. McGoldrick (2008), "Motivations of the Ethical Consumer,"
Journal of Business Ethics, 79(4), 445-467.
Galavielle J.P. (2004), "Business Ethics is A Matter of Good Conduct and of Good Conscience," Journal of Business Ethics, 53(1-2), 9-16.
Handelman, J. and S. Arnold (1999), "The Role of Marketing Actions with A Social Dimension: Appeals to the Institutional Environment," Journal of Marketing, 63(3), 33-48.
Kano, N., N. Seraku, F. Takahashi, and S. Tsuji (1984), “Attractive Quality and Must-Be Quality,” Journal of the Japanese Society for Quality Control, 14(2), 39-48.
Attributions and Brand Evaluations in A Product-Harm Crisis," International
Journal of Research in Marketing, 21(3), 203-217.
Levy, R. (1999), Give and Take, Cambridge: Harvard Business School Press.
Lichtenstein, D., M. Drumwright and B. Braig (2004), "The Effect of Corporate Social Responsibility on Customer Donations to Corporate-Supported Nonprofits," Journal of Marketing, 68(October), 16-32.
Luo, X. and C.B. Bhattacharya (2006), "Corporate Social Responsibility, Customer Satisfaction, and Market Value," Journal of Marketing, 70(October), 1-18. Marin, L. and S. Ruiz (2006), “I Need You Too! Corporate Identity Attractiveness for
Consumers and The Role of Social Responsibility,” Journal of Business Ethics, 71(3), 245-260.
Marin, L., S. Ruiz and A. Rubio (2009), “The Role of Identity Salience in the Effects of Corporate Social Responsibility on Consumer Behavior,” Journal of Business
Ethics, 84(1), 65-78.
Mishra, S. and D. Suar (2010), “Does Corporate Social Responsibility Influence Firm Performance of India Companies?,” Journal of Business Ethics, 95(4), 571-601.
Pinkston, T.S. and A.B. Carroll (1996), “A Retrospective Examination of CSR Orientations: Have They Changed?,” Journal of Business Ethics, 15(2), 199-206.
Putrevu, S. and Kenneth R. Lord (1994), "Comparative and Noncomparative Advertising: Attitudinal Effects under Cognitive and Affective Involvement Conditions," Journal of Advertising, 13(2), 77-91.
Sethi, S.P. (1975), “Dimensions of Corporate Social Performance: An Analytical Framework,” California Management Review, 17(3), 58-64.
Sen, S. and C.B. Bhattacharya (2001), "Does Doing Good Always Lead to Doing Better? Consumer Reactions to Corporate Social Responsibility," Journal of
Marketing Research, 38(May), 225-43.
Singh, J., M. del M. G. de los Sanchez, and I.R. del Bosque (2008), “Understanding Corporate Social Responsibility and Product Perceptions in Consumer Markets: A Cross-Cultural Evaluation, Journal of Business Ethics, 80(3), 597-611.
Tajfel, H. and J.C. Tuner (1979), “An Integrative Theory of Intergroup Conflict,” in W. Austin and S. Worchel (eds.), The Social Psychology of Intergroup Relations (Brooks/Coke, Monterrey, CA).
van Marrewijk, M. (2003), “Concepts and Definitions of CSR and Corporate Sustainability: Between Agency and Communion,” Journal of Business Ethics, 44(2-3), 95-105.
Appendix 1. Results of Classification for Each Initiative
Items (CSR initiatives) Sample 1 (A,O,R,I,M,Q)
Sample 2 (A,O,R,I,M,)
CSR Type
The company engages in providing products of high quality.* (12,50,0,14,58,0) (19,86,0,23,52,2) The company engages in product research & development and
innovation.* (44,33,0,30,27,0) (39,72,0,36,34,0)
The company designs products suitable for minority groups. (43,22,0,48,21,0) (36,54,0,56,35,1) Indifference The company makes product quality and price in right proportion. (3,40,0,16,75,0) (6,53,1,22,99,1) Norms The company makes the price and specification of products correct. (6,39,0,18,70,1) (4,63,1,15,98,1) Norms The company launches ads without cheating. (5,28,1,25,74,1) (5,55,0,28,93,1) Norms
The company has a good policy of customer satisfaction. (30,49,0,22,33,0) (22,67,1,35,56,1) Balancing Need The company emphasizes employee safety and health in the work
environment. (9,52,0,17,56,0) (6,75,0,20,81,0 ) Norms The company prohibits overtime working. (24,41,0,47,22,0) (16,51,1,64,49,1) Indifference The company gives salaries no less than the minimum requirement
by law. (5,27,1,17,84,0) (5,48,1,19,106,3) Norms
The company emphasizes employee training and development. (22,44,0,42,26,0) (35,71,0,37,39,0) Balancing Need The company enacts corporate policies to ensure the right of female
and minority group workers. (11,50,0,24,49,0) (8,68,0,40,66,0) Balancing Need The company does not discriminate employees based on gender,
pregnancy, and marital status. (6,57,0,18,52,1) (5,73,1,32,60,2) Balancing Need The company makes employees free to form clubs, initiate collective
negotiation, and file suiting.* (15,35,0,44,40,0) (19,55,0,49,58,1) The company engages in making its operational procedure less
resource and energy consumption.* (24,41,1,47,21,0) (31,62,2,47,39,1) The company reduces resource waste in the procedure of product
manufacturing, transporting, and using.* (25,43,0,47,19,0) (25,63,0,48,46,0)
The company does waste disposition and recycling well. (16,44,0,37,37,0) (15,77,1,35,55,1) Balancing Need The company enacts and implements specific environmental policies
and plans. (25,35,0,47,27,0) (33,53,0,54,42,0)
The company has emergency plans for environmental disasters. (11,52,0,28,42,1) (14,66,0,38,64,0) Balancing Need The company actively invests in products and technology that can
improve the environment. (45,26,0,42,21,0) (50,49,0,48,33,2) Altruism The company encourages its employees devote time and effort to
community service. (35,19,3,65,12,0) (48,35,3,72,24,0) Indifference The company complies with law and regulation of the local
government. (6,21,0,30,77,0) (7,48,0,32,95,0) Norms The company helps communities through philanthropic and
educational donation. (55,20,1,48,10,0) (56,47,1,55,23,0) Altruism The company supports and participates in social welfare activities. (39,19,0,68,8,0) (54,49,2,60,17,0) Indifference The company supports public policies beneficial for the society (e.g.,
developing environmental products and the energy-saving policy). (40,29,0,50,15,0) (47,51,2,54,28,0) Indifference The company prohibits bribery and gift receiving among its people. (11,29,0,55,38,1) (22,39,0,62,55,4) Indifference The company actively initiates approaches to solve societal problems
using its expertise (e.g., a computer company introduces
informational technology to the farmers in a far away region).* (36,43,0,38,17,0) (55,40,1,62,23,1)
The company reports its CSR initiatives or plan for sustainability. (24,21,0,70,19,0) (31,32,1,83,35,0) Indifference The company makes its financial information transparent. (3,26,0,48,57,0) (13,44,1,39,85,0) Norms The company ascertains its suppliers keep facilities that are safe and
are not harmful to health and environment. (19,30,0,38,47,0) (7,59,1,44,71,0) Norms The company’s transaction with suppliers complies with business
ethics.* (10,30,0,51,43,0) (11,40,1,55,74,1)
The company prohibits its suppliers hiring child labor. (8,45,0,30,51,0) (10,68,1,23,80,0) Norms The company asks its suppliers offer reports on CSR initiatives or
sustainability. (18,26,0,66,24,0) (21,37,3,77,44,0) Indifference The company does not override competitors by selling in a price
lower than cost. (8,29,1,55,40,1) (18,44,5,58,56,1) Indifference The overseas subsidiaries of the company prohibit child labor. (9,39,0,29,57,0) (5,52,1,44,80,0) Norms The overseas subsidiaries of the company prohibit any behaviors
against human rights. (18,34,0,28,54,0) (8,54,1,50,69,0) Norms The overseas subsidiaries of the company do not waste local
resource. (9,38,0,40,47,0) (13,55,0,54,60,0) Norms The overseas subsidiaries of the company do not bribe local
government.* (15,23,0,56,40,0) (12,48,0,53,69,0) Notes: (A,O,R,I,M,Q) indicates the initials of the quality dimensions by Kano et al. (1984): (Attractive,One-Dimension,Reverse,Indifference,Must-be,Questionable). The number corresponding to the initials indicates how many respondents were classified as a certain quality dimension for each CSR initiatives. * indicates the inconsistent results for the classification between the first and second samples and the corresponding CSR type is not shown.
Appendix 2. Experimental Scenario
Initiatives Scenario
Positive altruistic ---The company has been engaging in new product research & development and service innovation. For example, it launched new lines of clothes that used reused materials and organic cotton. To protect the environment, the company does not sell clothing made up of animal furs or from endangering animals. To help community development, in addition to philanthropic donation, the company purchases dairy products from local farms and this facilitates the ease of life and stability of production for the local community
Negative altruistic …The company seldom engages in new product research & development and service innovation. It even sold clothing made up of animal furs and endangering animals. This does harm on the environment. The company has never invested in philanthropic donation or local community development even though it has open for a lot of years.
Positive normative ….The company offers employees a safe and healthy environment. It assesses whether its suppliers do harm to the environment on a year basis and ascertain the job environment of its suppliers comply with government regulation. It avoids cooperating with suppliers hiring child labor.
Negative normative ….The company has poor work environment for employees. It has never assessed whether its suppliers use products that can do harm to the environment. Even when one of its suppliers hired child labor, the company did not voice out or exhort that supplier.
Positive balancing need
….The company always offers high quality products and emphasizes customer satisfaction. It actively offers job opportunities or experiences to female and minority groups. It gives these people the same promotion opportunities as others. It has launched a plan for waste reduction and recycling. During the implementation, many unwanted clothes and household goods were recycled.
Negative balancing need
….Based on a survey in 2006, the company provides products of poor quality and seldom conducts survey on customer satisfaction. There are few female and handicapped employees in the company. It is almost impossible for them to get a promotion. The company has never engaged in waste reduction and recycling plan.
Neutral altruistic, normative, and balancing need
….Specifically, clothing product category includes casual clothes, children’s wear, and swimming suit. Food category includes frozen foods, canned foods, hot foods, snacks, and the like. Household goods contain TV sets, washing machines, kitchenware, bicycles, and the like. There are also other product categories with rich product lines in the store.
Note. Each scenario begins with “ X company is a hypermarket that attracts customers by using a pricing strategy “everyday low price”. The company was founded in 1960 and has around 2000 employees now. It keeps clothing, food, household goods and the like.”
計畫成果自評
本研究探討公司社會責任對於消費者購買意願之影響，過去的研究立論之基 礎在於「有從事社會責任活動」對於司較有利，而在社會責任方面有瑕疵的資訊 對於公司不利，然而這些論點並未考慮各類社會責任行動的性質差異。且過去對 於社為責任的分類較多著重公司的觀點，如分類為經濟的、倫理的、慈善的以及 法律層面。卻未能思考從消費者感受上的差異予以區隔。Brunk and Biumelhuger (2011) 以小樣本的質化研究初步獲得一個分類架構，本研究認為值得進行較廣泛 而具系統性的分類研究，並探討各類社會責任行動對於消費者的企業倫理印象 （CSR image）上之影響。具體而言，本研究首先經由兩波調查研究，針對六個 CSR 領域探討消費者對於 38 項 CSR 作為的歸類，共可分為「利他型」、「規範型」、 以及「平衡需求型」三大類。不同類的社會責任的執行與否、違逆與否，對於消 費者的企業倫理印象效果不同。兩次分類的結果顯示歸類頗為穩定。 本研究又進一步以實驗法探討 CSR 正、反性(valence)與類型的交互作用。以 2（正、反性）* 3（利他型、規範型、平衡需求型）+1（中性，無 CSR 資訊）的 受試者間實驗，驗證不同類型的 CSR 的影響型態確有差異。然而並非所有假設都 獲得支持，然而這些發現對於當今的 CSR 的效果的知識已有一些貢獻。 本研究並進一步探討企業倫理印象的後果。此部分亦延伸過去以「社會認同 理 論 」（ social identity theory ） 所 獲 致 的 「 消 費 者 — 公 司 認 同 」 （Customer-corporate identification，簡稱 CC identification)在 CSR 作為 與消費者反應之中介效果的發現，將消費者的企業社會責任印象與上述各項構念 連結起來，形成一個更完整的 CSR 作用關係鍊。 本研究的結果對於廠商在 CSR 經營的策略有參考之價值。且在選擇 CSR 作為 時，可參考本研究對於 38 項常見的 CSR 之性質歸類。且在判斷 CSR 效果時有更豐 富的資訊可供參考，可以據此選擇有效的、消費者重視的，並避免違逆消費者的 期待。 本人已研究畫之發現撰寫成衣篇期刊投稿文章，由於還有附帶資料之收集， 故估計可以有二至三篇的文章可以發表。 本研究進行過程中，訓練大學生、碩士生以及博士生參與研究，引發他們對 於該項主題的興趣，並提升了其相關的知識水準與研究能力。亦可為本研究的另 一項價值創造—教育與人才的發展。
