Companies implement CSR policies to increase goodwill and improve their reputations [25
], increase profits [26
], and enhance their corporate identity attractiveness [27
]. Previous research has focused on how consumers react to CSR [1
]. Findings show the effect of CSR on consumer responses, such as customer attitudes toward the product [30
], positive product and brand evaluations [31
], consumer–company identification [3
], consumer loyalty [30
], customer donations [32
], and consumer satisfaction [33
], e.g., behaviors such as the one is represented in Figure 1
H2: The positive effect of CSR associations on purchase intention will be greater for generative consumers than for nongenerative consumers.
Study 2 relied on a 2 × 2 between-subjects experiment. We manipulated product transcendence to be either present or absent, as was company commitment to CSR. We created four flyers, one for each condition. A questionnaire was sent randomly to individuals from an online panel. The e-mail stated our affiliation with a European university. To avoid any influence on the respondents’ answers, we gave no explanation of the purpose of the survey. The link included in the e-mail randomly presented one of the four flyers and a self-administered questionnaire. We obtained 476 valid questionnaires (51% male, aged 18 to 62 years, with an average age of 27).
To design the flyers, we went back to Study 1 and selected two similar products with different transcendence score: a seven-day stay in the Sun & Beach Resort (a nontranscendent product) and a seven-day biking trip along the Way of St. James (a transcendent product) (MS&B = 3.11, MSt.James = 4.03; t = 4.48, p < 0.001). The Way of St. James, or “El Camino”, is the ancient pilgrimage route to the Cathedral of Santiago de Compostela in northwestern Spain. Thousands of people from all over the world set out each year to travel the Way. Most of them bike the pilgrimage route for nonreligious reasons: travel, sport, or simply the challenge of weeks of walking or biking. Many consider the experience a spiritual adventure that removes them from the hustle and bustle of modern life. The flyer for this trip stated: “The Way of St. James is a meaningful journey: culture, adventure, challenge, and solitude that will provide life-long memories and experiences to be shared”. The flyer for the Sun & Beach Resort trip stated: “The Sun & Beach Resort: good weather, nice beaches—a perfect place to rest and enjoy yourself, 365 days a year”. The price and length of both trips were the same.
To manipulate company commitment to CSR, we used two fictitious travel agencies with different CSR orientations. The flyer from the company engaged in CSR activities stated: “As a travel agency, the company is committed to sustainable development…”. The flyer from the company not engaged in CSR activities stated: “The company aims to be the leading travel company…”. These statements were taken from real corporate websites (http://www.viajeselcorteingles.es
). The terms “transcendence” and “social responsibility” never appeared on the flyers.
Four hundred seventy-six Spanish individuals answered the questionnaire, which was sent with a flyer and reflected one of the four experimental conditions: transcendent product and company committed to CSR (N = 111), transcendent product and company not committed to CSR (N = 115), nontranscendent product and company committed to CSR (N = 110), and nontranscendent product and company not committed to CSR (N = 140). The questionnaire collected information related to purchase intention (five items; [31
]), attitude toward the product (six items; [38
]), product transcendence (three items, e.g., “I believe this trip is transcendent”), attitude toward the company (three items; [39
]), and CSR associations (four items; [40
]). We also asked for CSR attributions—that is, respondents’ beliefs about why companies engage in CSR activities (five items; [41
])—as a control variable because they can enhance consumer responsiveness to a company’s CSR activities [42
]. Finally, we measured both respondents’ level of support for CSR (three items; [33
]) and their generativity (scale from Study 1; [43
]). All items were rated on a 7-point scale (1 = “strongly disagree”; 7 = “strongly agree”). Product transcendence and CSR associations appeared at the end of the questionnaire.
Cronbach’s alpha tests showed that the scales were reliable (α > 0.88). Reliability of the measures was also assessed using the composite reliability index and the average variance extracted index (AVE). For all the measures, both indices were higher than the evaluation criteria of 0.60 and 0.50, respectively [43
], as shown in Table 1
. Additionally, an application of the procedures suggested by Fornell and Larcker [44
] showed acceptable convergent and discriminant validity for the scales. Convergent validity was assessed by verifying the significance of the t
-values associated with the parameter estimates (Table 1
). All t-values were positive and significant (p
< 0.01). Also, as a first test of discriminant validity, we determined whether the correlations among the latent constructs were significantly less than one. The Φ-matrix (correlations between constructs) is provided in Table 2
. Evidence of the scales’ discriminant validity was found, as none of the confidence intervals of the Φ-values (±two standard errors) included the value of one [45
]. We also verified that the average variance extracted by the underlying construct was larger than the shared variance with other latent constructs. This condition was satisfied, which further supports the discriminant validity of the measures. In summary, all of the scales were found to be both internally consistent and discriminately valid and, as such, gave us confidence to proceed with an estimation of the structural model.
4.2. Results and Discussion
We checked the manipulation through analyses of variance. Analyses showed that we successfully manipulated product transcendence and company commitment to CSR. Respondents perceived the Way of St. James trip as more transcendent than the Sun & Beach Resort trip (MSt.James = 4.50, MS&B = 3.06; F(1, 474) = 111.64, p < 0.001). Similarly, respondents reported higher CSR associations with the company whose flyer mentioned CSR activities than for the company whose flyer did not (MCSR = 4.31, MNoCSR = 3.54; F(1, 474) = 50.06, p < 0.001). A company’s perceived CSR stance did not affect product transcendence (MCSR = 3.86, MNoCSR = 3.66; F(1, 474) = 1.79, p = 0.19). In addition, consumers’ attitudes toward the trips to the Sun & Beach Resort and to the Way of St. James were not significantly different (MS&B = 4.16, MSt.James = 4.31; F(1, 474) = 1.69, p = 0.20), and we found the same lack of significant difference for attitudes toward the companies (MCSR = 4.72, MNoCSR = 4.70; F(1, 474) = 0.026, p = 0.87). These results showed that the sample liked both companies and both products similarly and that only the two variables we manipulated (transcendence and CSR) varied.
We then divided the respondents into nongenerative consumers and generative consumers, using a median split. The resulting mean composite of generativity scores showed significant differences between the two groups (MNGC
= 3.19, MGC
= 4.80; F
(1, 474) = 759.91, p
< 0.001). We tested H2 through an analysis of covariance (ANCOVA), with purchase intention as the dependent variable; consumer generativity, product transcendence, and company commitment to CSR as independent variables; and consumer support of CSR as a covariate (Table 3
The interaction effect of consumer generativity and company’s CSR activities on purchase intention was significant (F
(1, 467) = 4.76, p
< 0.05), confirming H2 (Figure 2
). When the flyer included information about a company’s CSR activities, purchase intention was higher for generative consumers than for nongenerative consumers (MGC
= 4.29, MNGC
= 3.53; F
(1, 467) = 21.49, p
< 0.001). However, when the flyer did not include any reference to the company’s CSR stance, the two consumer groups did not significantly differ in their purchase intention (MGC
= 3.29, MNGC
= 3.06; F
(1, 467) = 1.19, p
The covariate, consumer support of CSR, showed a positive effect on purchase intention (F(1, 467) = 11.75, p < 0.001). Generativity also increased purchase intention (F(1, 467) = 15.70, p < 0.001), while product transcendence did not affect purchase intention significantly (F(1, 467) = 2.11, p = 0.15). However, we found a significant interaction between these two variables (F(1, 467) = 6.40, p < 0.01). This interaction is consistent with the findings in Study 1. Generative consumers showed higher purchase intention than nongenerative consumers for the transcendent product, the Way of St. James trip (MGC = 4.08, MNGC = 3.22; F(1, 467) = 25.05, p < 0.001). In contrast, generative consumers and nongenerative consumers showed similar levels of purchase intention for the nontranscendent product, the Sun & Beach Resort trip (MGC = 3.51, MNGC = 3.33; F(1, 467) = 0.50, p = 0.48).
We also confirmed previous findings [22
] for the positive influence of company commitment to CSR on purchase intention (MCSR
= 3.94, MNoCSR
= 3.18; F
(1, 467) = 38.01, p
< 0.001). CSR activities influenced purchase intentions of both nongenerative consumers (MNoCSR
= 3.06, MCSR
= 3.53; F
(1, 467) = 9.59, p
< 0.005) and generative consumers (MCSR
= 4.29. MNoCSR
= 3.29; F
(1, 467) = 39.00, p
< 0.001). These results, in addition to the ones commented two paragraphs above concerning the significant difference in purchase intention between generative and nongenerative consumers when the flyer also refers to CSR activities (F
(1, 467) = 21.49, p
< 0.001) and the lack of difference when the flyer does not include that information (F
(1, 467) = 1.19, p
= 0.28), confirm that the two lines shown in Figure 2
are not parallel and, therefore, that the increase in purchase intention is higher for generative consumers than for nongenerative consumers.
We had no expectations of the interaction effect of product transcendence and company commitment to CSR on purchase intention. This effect was not significant (F(1, 467) = 0.49, p = 0.48), nor was the three-way interaction among generativity, product transcendence, and company commitment to CSR (F(1, 467) = 0.20, p = 0.65).
We then considered only the respondents who received the flyers with a reference to CSR activities undertaken by the company (N = 221). We divided this subsample into two groups, using a median split. We called respondents with negative CSR attributions skeptics (i.e., consumers who do not believe that companies that claim to value CSR are truly committed to it; N = 102) and those with positive CSR attributions believers (i.e., consumers who do believe that companies that claim to value CSR are truly committed to it; N = 119). The mean score of CSR attributions significantly differed between the two groups (MSkep. = 2.76, MBeli. = 4.74; F(1, 212) = 316.17, p < 0.001).
We ran an ANCOVA with purchase intention as the dependent variable and consumer generativity, product transcendence, and CSR attributions as independent variables, while consumer support of CSR served as a covariate (Table 4
). The results showed that the covariate, support of CSR (F
(1, 212) = 8.79, p
< 0.01), and CSR attributions (MSkep.
= 3.46, MBeli.
= 4.35; F
(1, 212) = 12.83, p
< 0.001) were significant variables in predicting purchase intention. Generativity was significant (F
(1, 212) = 12.13, p
< 0.001), as was its interaction with product transcendence (F
(1, 212) = 5.36, p
< 0.05). However, neither product transcendence (F
(1, 212) = 1.77, p
= 0.19) nor the interaction of generativity with CSR attributions (F
(1, 212) = 1.11, p
= 0.30) was significant.
A surprising result for this subsample was the significant interaction between product transcendence and CSR attributions (F
(1, 212) = 4.30, p
< 0.05). Product transcendence did not affect purchase intention when the company engaged in CSR activities and consumers were CSR believers (MS&B
= 4.35, MSt.James
= 4.35; F
(1, 212) =0.11, p
= 0.74). However, consumers who were CSR skeptics showed lower purchase intentions for nontranscendent products than for transcendent products (MS&B
= 3.27, MSt.James
= 3.72; F
(1, 212) = 5.68, p
< 0.05) when the company engaged in CSR activities (see Figure 3
). Therefore, the variable product transcendence is relevant not only when considering consumer generativity but also when considering consumers’ CSR attributions. Product transcendence increases purchase intentions of both CSR skeptics and generative consumers. Finally, the three-way interaction among generativity, transcendence, and CSR attributions was not significant (F
(1, 212) = 0.97, p
In summary, the results from Study 2 confirm H2 as well as the results from Study 1, while also explaining how generativity influences consumer behavior related to CSR. We found that while a company’s commitment to CSR affects both nongenerative and generative consumers, the increase in purchase intention is higher for generative consumers than for nongenerative consumers. This means that generative consumers (because they are concerned about future generations, buy life insurance, invest in high-quality education, and purchase sustainable products) are even more sensitive than nongenerative consumers to CSR initiatives.
When consumers make decisions that involve companies with a high commitment to CSR, purchase intentions are higher for believers than for skeptics. Product transcendence interacts not only with generativity but also with CSR attributions. When consumers are skeptical about CSR activities, product transcendence becomes an asset (e.g., even if a company advertises that it sells paper produced with raw materials from sustainable forests, CSR skeptics will prefer recycled paper to regular paper manufactured by the company). In contrast, CSR believers are indifferent to product transcendence (e.g., CSR believers will value the company’s recycled and regular paper equally).