Introduction

The two major aspects typically used by consumers to evaluate a company are its contribution to societal issues (e.g., corporate social responsibility) and its operational ability (e.g., performance growth). Many companies may excel in one behavioral aspect but not in others. For instance, according to The Washington Post, Apple’s products are considered to be innovative and of high quality, even though the company has outsourced its manufacturing process to Foxconn in China, where low labor costs and poor working conditions have led to employee suicide (Berg, 2018).

When a company’s morality does not match its competence (i.e., when it is moral but incompetent or immoral but competent), the question of whether morality or competence prevails in relation to determinations of its reputation represents a major managerial issue. Research has shown that consumers’ perceived psychological distance from a company could be an important factor here. Indeed, greater psychological distances lead to the prioritization of morality, whereas nearer psychological distances lead to the prioritization of competence (Chung & Park, 2013, 2017). In the seminal study by Trope and Liberman (2010), psychological distance was separated into the spatial, temporal, social, and hypothetical dimensions, and while Chung and Park (2013, 2017) focused on the social and temporal dimensions, the effects of the other two dimensions—the spatial and hypothetical dimensions—still need further investigation.

In addition to psychological distance, another factor that can potentially determine a company’s reputation when its competence and morality do not match is the question of whether the company is a for-profit or non-profit company. In the case of a for-profit company, its ultimate goal is to gain as much profit as possible, and presumably, its competence should be prioritized over its morality. In terms of a non-profit organization, its aim is to further public interests rather than to pursue profits, indicating that its moral performance should be prioritized over its competence.

Given that only the temporal and social dimensions of a for-profit organization were investigated by Chung and Park (2013, 2017), the present study aimed to fill the identified gap in the literature by examining the moderating effects of spatial and hypothetical distances as well as the type of organization (non-profit organization versus for-profit organization) in relation to consumers’ evaluation and behavioral intention concerning organizations with a competence–morality mismatch.

An organization’s behaviors: competence and morality

People’s attitudes toward an organization are mainly driven by its competence and morality (Chung & Park, 2017). Competence is linked to several essential factors, including the organization’s skills, capacities, and knowledge (Weinert, 2001). In terms of competence, Brown and Dacin (1997) defined corporate ability (CA) as a concentration strategy that focuses on the expertise of the business (e.g., designing products). Different industries have different strategies for enhancing competence. For instance, companies in the service industry can improve their competence by optimizing their service delivery system or adding new services to their offering (Menor & Roth, 2008).

In contrast to competence, an organization’s morality is typically evaluated based on its level of corporate social responsibility (CSR), which refers to an organization’s commitment to using its business practices to enhance social, economic, and environmental well-being (Sen & Bhattacharya, 2001). Companies that pursue CSR focus on social responsibilities (e.g., diversity recruitment) (Brown & Dacin, 1997). Various classifications of acts associated with CSR have been developed in previous studies, and the Kinder, Lydenberg, Domini Research & Analytics (KLD) model is one of the most widely used standards (Du et al., 2017; Sen & Bhattacharya, 2001; Waddock & Graves, 1997). The KLD model features seven essential domains: (1) employee relations (e.g., providing secure and healthy working conditions), (2) human rights (e.g., helping eliminate human trafficking), (3) diversity (e.g., in terms of recruitment and promotion);, (4) community issues (e.g., charity donations), (5) corporate governance (e.g., ensuring transparency and pursuing the best interests of stakeholders and the community), (6) environment (e.g., using renewable materials and energy sources), and (7) product (e.g., ensuring product safety).

While numerous studies have demonstrated the positive influence of CSR on customers’ purchase intention (e.g., Bianchi et al., 2019; Guping et al., 2021; Huang et al., 2014; Sharma et al., 2018), only a few have tested the relative effects of CSR and competence. For instance, Öberseder et al. (2011) used a qualitative method to examine the importance of CSR in relation to consumers’ purchase decisions. Customers reported a favorable view of companies that did well in terms of CSR, although CSR was of lesser importance when compared with other factors (e.g., price, quality, brand, service). According to Mohr et al. (2001), CSR is only considered important by a minority of consumers (21%) when it comes to their regular shopping decisions.

Behavioral ambivalence

Ambivalence is characterized by the simultaneous presence of the positive and negative aspects of an event (Conner & Sparks, 2002). An organization can be morally ambivalent, meaning that consumers perceive it to be competent but immoral. For instance, Duhigg and Barboza (2012) described how Apple treated its subcontractors poorly with regard to the cost of building a product, which was meager when compared with the price at which the product was sold to consumers, even though Apple had one of the highest revenues worldwide. This situation created a negative perception of Apple’s morality and a positive perception of its competence among consumers. By contrast, competence ambivalence refers to a situation in which an organization is perceived as moral (e.g., emphasizing environmental protection, decent working conditions, etc.) but incompetent (e.g., poor product quality, low revenue, etc.).

Chung and Park (2013, 2017) tested how both moral and competence ambivalence affected consumers’ attitudes, finding that consumers’ psychological distance in relation to a company was a key factor determining whether morality or competence played a dominant role in influencing consumers’ attitudes toward that company. When the consumer perceived an event to be temporally or socially near, competence prevailed; otherwise, morality prevailed.

Construal level theory and psychological distance

Psychological distance is the perceived distance between an individual and a particular target. According to Trope and Liberman (2010), four dimensions influence the decision-making process concerning psychological distance: spatial distance, which is an individual’s perceived proximity with an event or object; temporal distance, which refers to the temporal interval between the present and the time the target event will occur; social distance, which is described as the perceived social connectedness between the people involved in the target event and the individual; and hypothetical distance, which is the likelihood of an event occurring (Trope et al., 2007). Some studies have shown that the four dimensions of psychological distance—the temporal, spatial, social, and hypothetical dimensions—are interrelated. For instance, using normative, polite language rather than casual, less polite language increases an audience’s perceived distance from the target of the communication (Stephan et al., 2010).

Construal level theory (CLT) concerns the relationship between psychological distance and how an individual thinks about an event abstractly or concretely (Liberman & Trope, 1998). More specifically, the construal level indicates how abstractly an individual thinks about an event, with a higher level of construal indicating a higher degree of abstractness. The main idea is that the further an individual perceives about an event, the more they tend to think abstractly about it. This relationship between the level of construal and the psychological distance can be reciprocal. The level of construal can modify the psychological distance (Soderberg et al., 2015), where the higher the level the construal, the further the psychological distance of the individual from the event or object (and vice versa). The psychological distance can also affect the level of construal. For example, Chou and Lien (2012) proved that spatial distance impacts consumers’ construal level. Their participants’ construal level regarding travel events for spatially distant cities was higher than that for spatially near cities. Meyers-Levy and Zhu (2007) found that when a larger vertical space (e.g., the ceiling height) was obvious, their participants classified objects into broader categories and used more abstract languages. The match between the psychological distance and the construal level facilitates persuasion. For instance, Moon (1999) discovered that respondents were more inclined to change their mind concerning a near source than a spatially distant source after receiving concrete and detailed messages.

In Chung and Park’s (2017) study, the participants were asked to evaluate companies that were highly moral but incompetent when compared with companies that were highly competent but immoral. For the companies that the participants perceived to be socially near, those that were competent (but immoral) were judged more favorably than those that were moral (but incompetent). For the companies that the participants perceived to be socially distant, those that were moral (but incompetent) were judged more favorably than those that were competent (but immoral). The same trend was also observed with regard to purchase intention. However, the effects of spatial distance and hypothetical distance have yet to be tested. Hence, we propose the following:

  • H1: Consumers’ (a) evaluation and (b) behavioral intention favor an organization with competence ambivalence (incompetent but moral) more than one with moral ambivalence (competent but immoral) when the organization is spatially distant, although this favorability reverses when the organization is spatially near.

  • H2: Consumers’ (a) evaluation and (b) behavioral intention favor an organization with competence ambivalence (incompetent but moral) more than one with moral ambivalence (competent but immoral) when the organization is hypothetically distant, although this favorability reverses when the organization is hypothetically near.

Non- and for-profit organizations

The way in which behavioral ambivalence affects an organization’s reputation can also depend on that organization’s operational goals. In fact, organizations can be divided into two types based on their operational goals. Non-profit organizations (NPOs) engage in activities intended to further the public interest, and any surplus generated via profit-making activities must contribute to public welfare. Thus, when behavioral ambivalence occurs within an NPO, its performance in the moral domain is presumably prioritized by consumers. By contrast, a for-profit organization’s primary purpose is to generate as much profit as possible rather than to contribute to society. Profits derived from profit-making activities are funneled into capitalists’ pockets. As a result, a for-profit company may forego expenses related to CSR activities to maximize its profits or improve its product and service quality. Therefore, when behavioral ambivalence occurs within a for-profit company, its performance in the competence domain is presumably prioritized by consumers. Hence, we advance the following hypothesis:

  • H3: Consumers’ (a) evaluations are more favorable and (b) they exhibit a higher behavioral intention toward a non-profit organization with competence ambivalence (incompetent but moral) than one with moral ambivalence (competent but immoral), although this trend reverses for a for-profit organization.

Hypothetical model

Our hypothetical model is illustrated in Fig. 1 below.

Fig. 1
figure 1

Hypothetical model of the present study

Full size image

To manipulate the type of organization, we focused on medical institutions, as they could be non-profit or for-profit according to Articles 3–6 of Taiwan’s Medical Law. It should be noted that we used “behavioral intention” instead of “purchase intention” to indicate each participant’s degree of willingness to pay for a hospital’s service, as we considered the former term to be more appropriate in the context of a medical service.

Experiment 1

Method

The research protocol for Experiment 1 was approved by the Research Ethics Committee for Human Subject Protection of the authors’ institution. A mixed-design experiment with one within-subject factor (behavioral ambivalence: competent and moral ambivalences) and three between-subject factors (type of organization: non-profit vs. for-profit; hypothetical distance: near vs. far; spatial distance: near vs. far) was followed. Eight groups of participants were required for this study based on the three-factor between-subject design. In terms of the within-subject factor of behavioral ambivalence, each participant was exposed to two different types of ambivalence, with the presentation order being counterbalanced (half were initially exposed to an organization with competence ambivalence, while the other half were initially exposed to a morally ambivalent organization), leading to 16 experimental conditions.

Participants

A power analysis software package—namely, G*Power 3 (Faul et al., 2007)—was used to calculate the minimum sample size required based on Chung and Park’s (2017) study. Based on their results, the effect size η2partial of the interaction between behavioral ambivalence and psychological distance on consumers’ evaluation was 0.04. Thus, to achieve a power of 0.8 in the mixed-methods experiment, with behavioral ambivalence being the within-subject factor, the required sample size was estimated to be 50 to manifest the necessary interaction between behavioral ambivalence and one dimension of psychological distance. As we manipulated two dimensions of psychological distance and the type of organization, we planned to use four times the required sample size, resulting in an estimated sample size of 200.

Procedure

We used Google Forms with Google Apps Script to create the experiment, with a self-coded function for random assignment. A convenience sampling method was applied. The questionnaire link was posted to a Facebook group, where the majority of users were residents living in the area surrounding the authors’ institution. The respondents were randomly assigned to one of 16 experimental conditions after clicking on a red button labeled “Please click here to generate a new questionnaire.” A consent form was shown first, and the respondents had to agree to the consent form and answer the exclusion questions (“Are you over 20 years old?” and “Are you a subordinate of any of the researchers involved in this study?”), as required by the Research Ethics Committee of the authors’ institution. In accordance with the Medical Law, we explained that there are non- and for-profit medical institutions. The respondents read a piece of simulated news used to manipulate an organization’s behavioral ambivalence (moral or competence ambivalence). The news stories were all fictitious and created solely for research purposes.

We used fictitious hospitals known as M and C, and we manipulated morality using CSR-related criteria based on the KLD classification and competence using competence-related criteria (e.g., increasing profit and providing high-quality services). We chose four domains of the KLD model (as mentioned in An organization’s behaviors: competence and morality section)—employee relations, diversity, community issues, and corporate governance—to design our experimental materials, as they were most suitable for incorporation into our experimental scenario. We excluded the “human rights” and “environment” dimensions because they did not align with our initial experimental context—namely, a hospital—where topics such as “human trafficking” and “environmental pollution” were irrelevant. The respondents read a simulated piece of news about Hospital M or C that depicted the positive and negative behaviors of the relevant hospital in terms of morality and competence. The type of organization was manipulated as a non- or for-profit hospital in the simulated news.

In the moral ambivalence experimental condition, the hospital’s behavior was immoral but highly competent (e.g., it exploited employees but provided high-quality services). Conversely, the hospital’s behavior was incompetent but highly moral in the competence ambivalence experimental condition (e.g., it had low user satisfaction but donated to charity regularly).

We manipulated the spatial distance factor by presenting the respondents with information about a hospital located either far from them (e.g., in Europe) or near to them (e.g., in Taiwan). To manipulate the hypothetical distance, we verbally changed the likelihood of the existence of the hospital, similar to the approach used in previous studies (Huang et al., 2021; Polman et al., 2018; Zhang & Wang, 2009). In the hypothetically near condition, the respondents read a piece of simulated news featuring assertive words and phrases implying that the target hospital certainly existed (e.g., “According to information provided by the health authority, the non-profit Hospital M…”). In the hypothetically far condition, the respondents read a piece of stimulated news featuring words and phrases conveying weaker arguments, implying that the target hospital did not exist (e.g., “Imagine there is a situation, suppose/if/assuming the non-profit Hospital M…”). All the experimental scenarios presented to the respondents are listed in Appendix 1. After reading the news, the respondents had to answer the subsequent questions.

Measurement

Measurement of the factors of interest to this study proceeded as follows.

Behavioral ambivalence

To verify the manipulation of an organization’s behavioral ambivalence, the respondents were asked two questions about their perception of the target hospital: (1) After reading this news, how moral (corporate social responsibility) do you think for-profit (non-profit) Hospital C (M) is, and (2) After reading this news, how competent do you think for-profit (non-profit) Hospital C (M) is? The responses to both questions were provided on a seven-point scale ranging from 1 = extremely negative to 7 = extremely positive.

Type of organization

To verify the manipulation of the type of organization, the respondents were required to respond to the following question: “Is the organization mentioned above a non-profit one or a for-profit one?”

Spatial distance

The spatial distance was measured by the following question: “How near do you feel the distance between you and the for-profit (non-profit) Hospital C (M) is?” The answer to this question was given on a seven-point scale ranging from 1 = extremely near to 7 = extremely far.

Hypothetical distance

The hypothetical distance was measured by the following question: “How likely do you feel that the for-profit (non-profit) Hospital C (M) really exists?” This question was answered using a seven-point scale ranging from 1 = very unlikely to 7 = very likely.

Evaluation

The measurement of the respondents’ evaluation was based on Chung and Park’s (2017) study and determined using a seven-point scale. Depending on the experimental condition, the three questions (Cronbach’s alpha = 0.82) were as follows: “According to the above news, how much do you favor for-profit (non-profit) Hospital C (M)?”, “According to the above news, how good do you think for-profit (non-profit) Hospital C (M) is?”, and “According to the above news, how desirable do you think for-profit (non-profit) Hospital C (M) is?”

Behavioral intention

The measurement of the respondents’ behavioral intention consisted of three items adopted from Liu et al. (2021), which were rated on a seven-point scale ranging from 1 = very unlikely to 7 = very likely. Depending on the experimental condition, the three questions (Cronbach’s alpha = 0.91) were as follows: “According to the above news, when you need medical treatment, how likely would you consider for-profit (non-profit) Hospital C (M)?”, “According to the above news, when you decide to go to the hospital, how likely would you choose for-profit (non-profit) Hospital C (M)?”, and “According to the above news, how likely would you recommend the for-profit (non-profit) Hospital C (M)?”

Demographic variables

Other demographic variables—namely, gender, year of birth, education level, region, and monthly income—were also included in the experimental forms.

Results

The data for this experiment were collected from November 5, 2021 to November 12, 2021. Before we closed the link to the online experiment, a valid sample of 360 (from 408) responses were recorded, exceeding the minimum required sample size calculated above, as our online experiment automatically collected responses. Thus, we included all the valid response in the analysis (65.8% females and 34.2% males, Mage = 30.63) (Table 1).

Table 1 Demographical information concerning the respondents in Experiment 1 (n = 360)
Full size table

Manipulation checks

The following manipulation checks were performed in Experiment 1.

Type of organization

To ensure that the respondents had completely read and understood the experimental scenarios, we asked them whether the scenario they had read about involved a non-profit or a for-profit hospital. As each respondent read two scenarios—one regarding moral ambivalence and the other regarding competence ambivalence—they all answered the question twice. If their response was different from what they actually saw (e.g., they read descriptions of a non-profit hospital but responded that they had read descriptions of a for-profit hospital), their response was excluded, and this approach led to a final sample size of 360.

Behavioral ambivalence

We performed a Wilcoxon signed-rank test to check whether the manipulation of behavioral ambivalence was valid. In terms of the rating of competence, the hospital with competence ambivalence (Hospital C, x̄ = 2.71) was perceived to be less competent (z = -12.29, p < .001, r = .65) than the hospital with moral ambivalence (Hospital M, x̄ = 4.72). With regard to the rating of morality, Hospital M (x̄ = 2.48) was perceived to be less moral (z = -14.62, p < .001, r = .77) than Hospital C (x̄ = 5.13). Thus, the manipulation of behavioral ambivalence was successful.

Spatial distance

Each participant had to twice report how far they perceived the hospital to be from them, once for Hospital M and once for Hospital C. For each participant, Hospitals M and C were located either both in Taiwan or both in Europe in the same survey questionnaire featuring different ambivalence scenarios. Therefore, the perceived spatial distance was indexed by the averaged rating values of the spatial distance for the two hospitals. A Mann–Whitney U test revealed that the perceived spatial distance for the spatially near hospital (x̄ = 3.76) was significantly lower (U = 11833.5, z = -4.44, p < .001, r = .23) than that for the spatially far hospital (x̄ = 4.56). Thus, the manipulation of spatial distance was successful.

Hypothetical distance

Similar to the approach for spatial distance, the perceived hypothetical distance was indexed by the averaged rating values of hypothetical distance for Hospitals M and C. A Mann–Whitney U test revealed that there was no statistically significant difference (U = 15422, z = − 0.77, p = .44, r = .04) between the perceived hypothetical distances of the hypothetically near (x̄ = 5.09) and far hospitals (x̄ = 4.99). Hence, the manipulation of hypothetical distance was not successful.

To examine the effect of hypothetical distance, given that the manipulation proved unsuccessful, we applied the second-best strategy and collapsed the respondents in the hypothetically far and near groups, thereby splitting the respondents according to their averaged rating values of the hypothetical distance. These values were 1.5, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, and 7.0. We tried to separate them into three groups with roughly equal numbers in each group, which led to rating values of 1.5–4.0 being assigned to the “unlikely” group (these respondents tended to think that the hospital was unlikely to exist), 4.5–5.0 to the “medium” group, and 5.5–7.0 to the “likely” group (these respondents tended to think that the hospital was very likely to exist). For the avoidance of confusion, we termed this factor the “relative hypothetical distance.”

Hypothesis testing

We used non-parametric statistical methods to test whether the spatial distance (H1), hypothetical distance (H2), and/or type of organization (H3) moderated the effects of an organization’s behavioral ambivalence on the respondents’ evaluation and behavioral intention. First, we computed the “difference in evaluation” and “difference in intention” by subtracting the evaluation and intention ratings for Hospital M from Hospital C (Hospital C - Hospital M). Our hypotheses would be supported if the evaluation or intention values significantly differed between the spatially near and far locations (H1), the three different relative hypothetical conditions (H2), or the non-profit and for-profit organizations (H3). As we tested six hypotheses, the Bonferroni correction was applied so a statistical significance would be indicated by a p value less than 0.008.

Testing of H1

For the dependent variable of evaluation (H1a), a Mann–Whitney U test revealed that the difference in the evaluation values was statistically significantly lower in the spatially near group (x̄C−M = -0.15) when compared with the spatially far group (x̄C−M = 0.29), where U = 13444.5, z = -2.75, p = .006, r = .14 (see the left panel in Fig. 2). Consistent with hypothesis H1a, the respondents favored the organization with competence ambivalence when it was spatially far, although this favorability reduced when it was spatially near. For the dependent variable of behavioral intention (H1b), a Mann–Whitney U test showed that the difference in the intention values did not significantly differ between the spatially near group (x̄C − M = -1.47) and the spatially far group (x̄C − M = -1.30), where U = 15,623, z = − 0.54, p = .59, r = .03 (see the right panel in Fig. 2). Thus, H1b was not supported.

Fig. 2
figure 2

Mean values for evaluation (left) and behavioral intention (right) in the different conditions of Experiment 1. Error bars indicate 95% confidence intervals

Full size image

Testing of H2

It should be noted that our manipulation of hypothetical distance was not successful; therefore, we used another strategy to separate the data into the “unlikely,” “medium,” and “likely” groups according to the respondents’ ratings of the perceived hypothetical distance. A Kruskal–Wallis test revealed that the evaluation values across the three groups of relative hypothetical distance did not significantly differ (H (2) = 4.54, p = .10, η2H = 0.01) among the “unlikely” (x̄C−M = 0.04), “medium” (x̄C−M = 0.26), and “likely” (x̄C − M = -0.50) groups. The difference in behavioral intention also showed no significant difference (H (2) = 4.32, p = .12, η2H= 0.01) among the “unlikely” (x̄C−M = -1.46), “medium” (x̄C−M = -1.12), and “likely” (x̄C − M = -1.58) groups. Thus, neither H2a nor H2b was supported.

Testing of H3

A Mann–Whitney U test demonstrated that the evaluation values did not significantly differ (U = 16,150, z = − 0.03, p = .98, r = .001) between the non-profit organization (x̄C−M = 0.05) and the for-profit organization (x̄C−M = 0.11). The difference in the behavioral intention values between the non-profit organization (x̄C − M = -1.44) and the for-profit organization (x̄C − M = -1.33) also did not differ (U = 16008.5, z = − 0.17, p = .87, r = .009). Hence, neither H3a nor H3b was supported.

Other findings

In addition to the three hypotheses focusing on how behavioral ambivalence’s effect was moderated, we also identified its main effect. If we only compared the respondents’ ratings concerning their evaluation and intention between Hospital C and Hospital M, a Wilcoxon signed-rank test revealed a generally higher behavioral intention (z = -11.57, p < .001) for the moral ambivalence hospital (x̄ = 4.36) than the competence ambivalence hospital (x̄ = 2.98), with an effect size of r = .61. However, the generic benefit of moral ambivalence was not observed in the evaluation (z = -1.63, p = .10, r = .09).

Experiment 2

A potential problem with Experiment 1 was the failure to successfully manipulate hypothetical distance. We separated the respondents into three relative hypothetical distance groups based on their responses, and the effect was not significant in relation to the respondents’ evaluations or behavioral intentions. It is possible that hypothetical distance might be irrelevant in terms of determining a consumer’s evaluation of a company; alternatively, it might be important, but our experimental manipulation did not elicit sufficient differences in the hypothetical distance to elicit different degrees of evaluation and behavioral intention. To overcome this issue, we modified the experimental scenarios in Experiment 2.

Method

Participants

The research protocol for this experiment was approved by the Research Ethics Committee for Human Subject Protection of the authors’ institution. This experiment used a two-way 2 (behavioral ambivalence: competence and moral ambivalence) × 2 (hypothetical distance: near vs. far) mixed design. As in Experiment 1, behavioral ambivalence was manipulated in Experiment 2 as a within-subject factor, while hypothetical distance was manipulated as a between-subject factor.

To calculate the required sample size, we used G*Power 3 (Faul et al., 2007), setting the effect size η2partial (based on a parametric statistical method) for evaluation as 0.010 and for behavioral intention as 0.009, which was based on the interaction effect between the hypothetical distance and behavioral ambivalence observed in Experiment 1. To achieve a power of 0.8, the minimum sample size should be 198 and 218 for the respondents’ evaluation and behavioral intention, respectively. Thus, we chose 218 as the final sample size.

Procedure

We manipulated behavioral ambivalence in the same way as in Experiment 1. Briefly, the respondents read a piece of news about a company. The company’s behavior was immoral but competent in the moral ambivalence condition, whereas it was incompetent but moral in the competence ambivalence condition.

We manipulated the hypothetical distance by presenting the respondents with news about either a real company or a fictitious company. In the hypothetically near (high possibility) condition, the respondents read a piece of news about Apple, while in the hypothetically far (low possibility) condition, the respondents were presented with a piece of news about a fictitious company named Coconut. In addition to the different company names, the wording describing the company’s behavior was stronger and more assertive in the hypothetically near condition than in the hypothetically far condition. The news stories presented to the respondents are provided in Appendix 2. All of the news stories were fictitious and created solely for research purposes. 

Results

The data for this experiment were collected from June 8, 2022 to July 7, 2022. Before we closed the link to the online experiment, a valid sample of 250 (from 268) responses was collected, exceeding the required minimum sample size of 218. Thus, we included all the valid responses in the analysis (59.6% females and 40.4% males, Mage = 29.84) (Table 2).

Table 2 Demographical information concerning the respondents in Experiment 2 (n = 250)
Full size table

Manipulation checks

We asked the respondents three questions to ensure that they had completely read and understood the experimental scenarios: “Which company was in the news that you just read about?”, “What kinds of products does the company focus on?”, and “What media reports are there about the company?” As each respondent read two scenarios—one in which the ambivalence concerned morality and the other in which the ambivalence concerned competence—they all answered these three questions twice. If any of the respondent’s response was different from what they actually saw in the news, the data would be excluded.

Behavioral ambivalence

For the rating of competence, a Wilcoxon signed-rank test showed that the competence ambivalence scenario (x̄ = 3.47) was perceived to be less competent (z = -3.24, p = .001, r = .20) than the moral ambivalence scenario (x̄ = 4.11). For the rating of morality, the moral ambivalence scenario (x̄ = 2.06) was perceived to be less moral (z = -12.49, p < .001, r = .79) than the competence ambivalence scenario (x̄ = 5.31). Thus, the manipulation of behavioral ambivalence was successful.

Hypothetical distance

As in Experiment 1, each participant answered the questions regarding the perceived hypothetical distance twice (once for the competence ambivalence scenario and once for the moral ambivalence scenario). A Mann–Whitney U test was performed on the averaged rating values for the perceived hypothetical distance. Here, Apple (x̄ = 4.81) was perceived to be more likely to exist (U = 6355.5, z = -2.59, p = .01, r = .16) than Coconut (x̄ = 4.43). Hence, our manipulation of the hypothetical distance was successful.

Hypothesis testing

Similar to Experiment 1, we first computed the “difference in evaluation” and “difference in intention” by subtracting the evaluation and intention ratings for the moral ambivalence scenario from the competence ambivalence scenario. H2 would be supported if the difference in evaluation or intention differs between the hypothetical far and the hypothetical near condition. The hypothesis testing in Experiment 2 proceeded as follows.

Testing of H2

A Mann–Whitney U test revealed that the difference in evaluation values did not significantly differ (U = 7546.5, z = − 0.45, p = .65, r = .03) between the hypothetically near group (x̄C−M = 1.48) and the hypothetically far group (x̄C−M = 1.38). Moreover, the difference in behavioral intention did not significantly differ (U = 7161.5, z = -1.13, p = .26, r = .07) between the hypothetically near group (x̄C − M = 0.59) and the hypothetically far group (x̄C − M = 0.28).

Nevertheless, there was a generic higher evaluation of the competence ambivalence condition than the moral ambivalence condition. A Wilcoxon signed-rank test revealed that the evaluation of the competence ambivalence scenario (x̄ = 4.24) was significantly higher (z = -10.78, p < .001, r = .68) than that of the moral ambivalence scenario (x̄ = 2.81) (see the left panel in Fig. 3). For the behavioral intention, the rating was also higher (z = -3.19, p = .001, r = .20) in the competence ambivalence scenario (x̄ = 3.59) than in the moral ambivalence scenario (x̄ = 3.17).

Fig. 3
figure 3

Mean values of the evaluation (left) and behavioral intention (right) in the different conditions of Experiment 2. Error bars indicate 95% confidence intervals

Full size image

General discussion

Key findings

This study sought to test the interaction effect between psychological distance and behavioral ambivalence in relation to consumers’ evaluations and behavioral intentions. Thus, we postulated three hypotheses and conducted two online experiments to examine consumers’ attitudes toward an organization’s behavior.

In Experiment 1, the respondents evaluated the organization with competence ambivalence more favorably than the organization with moral ambivalence when the organization was perceived to be spatially far, whereas the preferentiality was reversed when the organization was perceived to be spatially near. Our manipulation of the hypothetical distance was not successful; thus, we used a correlational method that compared the evaluations of respondents with different levels of perceived hypothetical distance, but their differences did not reach statistical significance. In terms of behavioral intention, there was only a main effect, where the respondents’ behavioral intention was higher for the organization with moral ambivalence than for the organization with competence ambivalence.

In Experiment 2, no significant interaction effect was found between behavior ambivalence and hypothetical distance in relation to the respondents’ evaluation, even though our manipulation of the hypothetical distance was successful. The evaluation value of the company with competence ambivalence (moral but incompetent) was in general higher than that of the company with moral ambivalence (competent but immoral). The resulting trend was the same for behavioral intention.

Based on both our data and Chung and Park’s (2013, 2017) data, we can conclude that psychological distance may affect how people evaluate a company’s behavior, even though it does not apply to all the dimensions of psychological distance. Spatial distance, temporal distance, and social distance may affect people’s evaluation, albeit the effect of hypothetical distance is not as strong as that of the other dimensions.

Discrepancies within and between studies

In Experiment 1, two dependent variables—the respondents’ evaluation and behavioral intention—were measured. Our hypothesis concerning the interaction between psychological distance and behavioral ambivalence was manifested in relation to the respondents’ evaluation but not their behavioral intention. Thus, we speculated that the two dependent variables involved different decision-making processes.

The measurements of the respondents’ evaluation involved their perceived values concerning an organization, which were related to a high level of construal (Eyal et al., 2008, 2009; Trope & Liberman, 2010). This is probably why a higher evaluation for the organization with competence ambivalence in the spatially far condition was observed, while in the spatially near condition, the trend toward a higher evaluation for the morally ambivalent institution was much lower, because the mental processes underlying the “evaluation” might have increased the overall level of construal, leading to prioritization of morality, for even the spatially near condition.

Different to people’s evaluation, which is represented at the abstract and high levels, low-level, concrete, or peripheral features are emphasized at the final purchase decision stage (Dhar & Kim, 2007). Competence is associated with the concrete and low levels (Chung & Park, 2017); therefore, our results showed only a generic benefit for the morally ambivalent organization when compared with the organization with competence ambivalence in Experiment 1.

It is interesting to note the divergent results between evaluation and behavioral intention, which were only observed in Experiment 1 in the present study, meaning that they were not observed in Experiment 2 or in the study by Chung and Park (2017). The difference in the overall psychological distance seen in Experiment 1, Experiment 2, and Chung and Park’s (2017) study might be a critical factor in this regard. In Experiment 1, we presented a hospital in Taiwan as spatially near and a hospital in Europe as spatially far. In Experiment 2, both Apple and Coconut might have been perceived as spatially far because for our respondents, whose native language was Mandarin, the company names printed in English might have triggered a feeling of remoteness. The increase in the spatial distance itself might have induced a general preference for the moral domain, leading to higher overall ratings for the competence ambivalent company than for the morally ambivalent company in terms of both evaluation and behavioral intention in Experiment 2. Chung and Park (2017) used a fictitious company referred to a retailer X as the research material, which did not provoke any thought concerning spatial distance. However, the broad term “retailer” might have triggered a feeling of abstractness, meaning that the psychological distance in their study was likely to have been generally higher than in Experiment 1, which used a “hospital” as the research material. This concrete concept of a hospital, together with the low-level decision-making process associated with behavioral intention, might explain higher behavioral intention regarding the morally ambivalent organization when compared with the competence ambivalent organization.

Theoretical and practical contributions

One significant contribution of the present study is that it extends applications of fundamental psychological theories on real-world problems to the corporate level. The construal level theory has been applied to studies on attitudes and behavioral changes (e.g., Eyal et al., 2009), but previous studies focused on the individual level. Chung and Park (2017) used the theory to study corporate behavior, but they only focused on the temporal and social domains of psychological distance. Given that the spatial distance is a major dimension that determines the psychological distance (e.g., Henderson et al., 2006; Zhang & Wang, 2009; Trope & Liberman, 2010), a study that probes the effect of spatial distance is required, and our study serves to fulfill this goal. Moreover, we also examined the effect of the type of organization, and the lack of a moderating effect of the type of organization suggested that the psychological mechanisms underlying consumers’ evaluation and behavioral intention are equivalent for non-profit and for-profit organizations.

In terms of practical contributions, our study provides empirical basis for the need to take spatial distance into account when developing effective promotion strategies. Before the Internet became a major platform for human interactions, most of the companies just had to focus on building positive images for potential consumers and stockholders in the vicinities of their geographical locations. The development of modern telecommunication technologies and the widespread usage of social media have enabled commercial activities to take place among parties that are geographically far from each other. Given that the consumers’ priorities shift towards the moral domain as the spatial distance increases, companies need to realize that corporate social responsibility is not just an obligation, but also a strategy to attract potential consumers.

One interesting issue that worth investigating is how people perceive the spatial distance of online or virtual businesses. They could be perceived to be far, because they are physically unreachable, and this should entail a prioritization of morality over competence. They could also be perceived to be near, because we can always interact with them anytime and anywhere, and this should entail a prioritization of competence over morality. Future studies are needed to tackle this issue.

Limitations

It must be acknowledged that the present study has certain limitations. First, it is quite challenging to draw a strong conclusion regarding hypothetical distance based on the lack of a significant effect. Although our manipulation of the hypothetical distance in Experiment 2 proved successful, the difference between the hypothetically near and hypothetically far conditions might still have been too small. Indeed, they might both have been perceived as hypothetically far, as the respondents knew that they were taking part in an online experiment and that all the stories were presumably fabricated. An ideal way to create a “hypothetically near” scenario would have entailed presenting the respondents with real news stories; however, it may have been difficult to find true stories about two companies with equivalent popularity, with one exhibiting moral ambivalence and the other exhibiting competence ambivalence.

Second, we used only texts to manipulate the psychological distance in both experiments. Other studies have shown that words trigger a high level of construal due to their abstract and symbolic nature, whereas pictures are perceived more concretely (Amit et al., 2009, 2012) and can probe a low level of construal. Hence, future research should manipulate the form of the information (text or pictures) to examine how it interacts with both behavioral ambivalence and psychological distance.

Third, our sample of organizations was very limited. In Experiment 1, we used only hospitals as the target organizations, and an interaction effect between spatial distance and behavioral ambivalence was observed. Although we used a different type of organization in Experiment 2, the spatial distance was not manipulated in this experiment. To determine whether the interaction between spatial distance and behavioral ambivalence can be generalized to other types of organizations, more studies need to be conducted.