Keywords

1 Introduction

Trust, in general, is a multi-layered phenomenon as well as a topical subject. The significance of trust for modern societies and for social coexistence is emphasized in sociological approaches in particular (e.g., Luhmann 1979, p. 4; Kohring 2004, p. 80; Barbalet 2009). According to these approaches, a society without trust cannot exist (Barber 1983, p. 19; Deutsch 1962). For organizations, the trust of their stakeholders represents a valuable, intangible resource. This trust is significant for organizations because it promotes acceptance of their actions, leads to support for their activities, and, ultimately, expands their scope of action (Morgan and Hunt 1994). In addition, trust is seen as a key factor for an enduring, loyal relationship with their stakeholders (Ki and Hon 2007; Aurier and N’Goala 2010; Grayson et al. 2008). Despite the significance of stakeholder trust in organizations, this field of research has received hardly any systematic examination to date. The present contribution addresses this research gap.

In this contribution, we consider the questions of what stakeholder trust in organizations specifically looks like and how it can be measured accordingly. The focus of this examination is on the trust of external stakeholders. Yet, engaging with trust is a challenge in general because nothing resembling a common understanding of what the term means has so far been established, which is why the field of research is diffuse and heterogeneous (Barbalet 2009). The diverse individual results of the numerous theoretical approaches and empirical studies on trust each relate to a specific problem, which makes it all the more difficult to combine them to form a coherent image of the current state of research (Möllering 2006, pp. 128–129).

We will describe the function and significance of stakeholder trust for organizations primarily on the basis of the sociological assumptions by Luhmann (1979), Kohring (2004; Kohring and Matthes 2007), and Giddens (1991), which we shall apply to trust in organizations. We will also consider the approach to trust originating from organizational psychology that is taken by Mayer et al. (1995; see also Mayer and Davis 1999; Schoorman et al. 2007) because this approach describes trust as a process. This makes the elusive entity of trust into something concrete that can be operationalized. Although the approach of Mayer et al. was developed to research interpersonal trust within organizations, it can also be applied to trust that is placed in organizations (Schoorman et al. 2007). We will describe our experiences of applying the approach with reference to two studies in which we examined trust in non-profit organizations (NPOs). For the examination, we focused on the perceived trustworthiness of organizations because this factor is central to the process of trust. We will present the problems that we encountered in measuring the trustworthiness of organizations and critically discuss the approach. Despite the problems with measurement and the resultant limitations, our application of the approach so far has enabled us to reveal valuable implications both for further trust research and for the strategic communication of organizations. We will present these as part of our conclusion.

2 Trust in Organizations: Theoretical Derivation

When discussing trust in an organizational context, a distinction must be made between two different types of trust relationships: the trust of internal stakeholders and the trust of external stakeholders. In the first type of trust relationship, the focus is on the trust that members of an organization place in the organization or on intraorganizational trust, which describes the trust between members of an organization (Schweer and Thies 2003, p. 57). In our contribution, we will concentrate on the trust that external stakeholders place in NPOs. Before we depict this type of trust with reference to the example of political parties and non-governmental organizations (NGOs), we shall first define the term “trust”. Such a definition of the term is necessary because trust is a “[…] contested term […]” (Levi and Stoker 2000, p. 476) and there are different views concerning what is meant by the term (McKnight and Chervany 2001). We will theoretically demonstrate why trust in organizations is a necessity in modern, differentiated societies and what the function of trust consists of before finally describing trust as a process following Mayer et al. (1995).

2.1 The Necessity of Trust in Organizations

Organizations are often described as corporate actors with a specified structure that is formalized or schematized to an above-average degree. Organizations exhibit a specific purpose orientation and pursue defined objectives (Endruweit 1981, pp. 17–18). The actions of organizations consist of the actions of the relevant organization members, which are decisively influenced by the organizational mechanisms. These mechanisms cause the motives of the organization members to become generalized and lead to a specification of behavior. Consequently, it is possible for the diverse actions of the organization members to be substantively and temporally generalized (Luhmann 2005, p. 14)Footnote 1. Organizations are of vital significance to members of a modern society because expert knowledge is integrated in them. This knowledge is used to provide certain goods and services for certain target groups. Stakeholders are dependent on organizations because they are not able to produce these goods and services themselves in a society shaped by the division of labor and specialization. They lack the knowledge, skills, and resources required to do so. Moreover, they often do not have the opportunity or do not wish to acquire the requisite knowledge, skills, and resources (Giddens 1991, pp. 83–84).

However, there are fundamental problems in the relationship between stakeholders and organizations that can make it risky for stakeholders to link an action of their own with the actions of organizations. Firstly, organizations, just like their stakeholders, have a plethora of possible courses of action available to them. Organizations can choose the courses of action expected by the stakeholders, but they can also make a different choice. The actions of corporate actors are therefore contingent; in other words, it is also always possible for everything to be done differently. The contingency of the actions makes it more difficult for stakeholders to anticipate the actions of organizations. It is therefore possible and likely that the expectations placed on organizations will be disappointed by the organizations. Consequently, stakeholders are able to perceive uncertainties and risks when they enter into relationships with the organizations. Secondly, the organizations usually have an edge in terms of knowledge that can be used to the advantage or to the disadvantage of the stakeholders. If stakeholders have insufficient information about the abilities and characteristics of the organizations and about their intended actions and objectives, it is difficult for them to anticipate or recognize what might be opportunistic actions that are being taken by the organizations (Schichtmann 2007; Sztompka 1999, p. 23; Möllering 2008). Finally, the intensity and pace of processes of social change are making it increasingly difficult for organizations to fulfill the expectations of the stakeholders. Processes of change such as globalization and internationalization create complexity, uncertainties, and incalculabilities—that is, life’s risks. These risks also have an effect on the actions of organizations and make it more difficult for them to perform their tasks (Sztompka 2006; Cook and Cook 2011; Giddens 1991, pp. 109–110).

If, owing to the aforementioned problems, stakeholders perceive a risk in the relationship with an organization, problems ensue. As laypeople, stakeholders are dependent on organizations and on their goods and services in a society shaped by the division of labor (Giddens 1991, pp. 83–92). At the same time, the purchasing of goods and services is vitally significant to organizations. So that relationships between stakeholders and organizations can exist under the condition of perceived risks, a suitable mechanism is required in order to be able to deal with these risks in the relationships. Trust is such a mechanism.

2.2 The Function of Trust in Organizations

Sociological approaches often primarily describe trust in systems, where these take the form of subareas or functional systems of society, such as politics or the economy. According to these approaches, personal trust is increasingly turning into system trust in modern, differentiated societies (Giddens 1991, pp. 100–111; Luhmann 1979, pp. 48–58). Individuals have to make themselves dependent on the functioning of highly complex, inscrutable systems together with their actors. Although trust in organizations or institutions—which are understood as being corporate actors—is indicated in these approaches, they do not clearly and compellingly represent and define it as such. Nevertheless, many of the assumptions can be applied to organizations.

Kohring (2004, pp. 110–111) explains that system trust is dependent on the social function of the relevant functional or expert system. This refers to the proper functioning of the system as seen from the subjective perspective of the layperson. According to this understanding, trust in systems develops on the basis of generalized specific expectations about the future performance of the system (Kohring 2004, p. 131). As well as being based on expectations, system trust, according to Giddens (1991, p. 83), is based on the assessment of the trustworthiness of established expert knowledge pertaining to these systemsFootnote 2. This perceived trustworthiness serves the assessment of benefits and risks in trust-related situationsFootnote 3.

According to Luhmann (1979, pp. 93–94), the function of trust in general is to reduce social complexity. Missing information is replaced with an internally guaranteed certainty. Yet, the use of trust to establish complete certainty must be seen in a critical light because the fundamental problems that make trust necessary, but also risky, continue to exist objectively. Because of the contingency of the actions, the information asymmetry that favors organizations, and the processes of social change, the risk to stakeholders that their expectations will be disappointed remains objectively present. Rather, trust has the function of making it possible to tolerate the risk perceived by a stakeholder in his or her relationship with the organization. Trust “[…] is the selective linking of the actions of another with one’s own actions under the condition of the toleration of the perceived risk in a way that cannot be legitimated by means of factual arguments” (Kohring 2004, p. 131, translated from the German). A pretense is made that the desirable outcomes of the act of trust are likely to occur:

[T]rust is an ongoing process of building on reason, routine, and reflexivity, suspending irreducible social vulnerability and uncertainty as if they were favorably resolved, and maintaining thereby a state of favorable expectation toward the actions and intentions of more or less specific others. (Möllering 2006, p. 111)

From the perspective of the organizations, toleration of the perceived risk leads to an increase in their legitimacy and effectiveness. Organizations that are trusted can maintain or expand their scope of action owing to the fact that stakeholder trust increases the acceptance of their activities. However, if organizations lack stakeholder trust, their legitimacy, stability, and scope of action are threatened (Hoffjann 2011, pp. 65–66).

2.3 Trust in Organizations as a Process

Following our description of the function of trust for organizations, our next step will be to theoretically derive an understanding of what trust specifically is. For this purpose, the elaborated approach to trust taken by Mayer et al. (1995; see also Schoorman et al. 2007; Mayer and Davis 1999), which originates from organizational psychology, will be taken into consideration. This approach describes trust as a process and thereby makes the elusive entity of trust into something concrete that can be operationalized. The sociological explanations considered previously lack this kind of specific process description. Although the approach was developed to research interpersonal trust within organizations, it can also be applied to trust that is placed in organizations (Schoorman et al. 2007).

Applied to organizations, trust, following Mayer et al. (1995), is the willingness to make oneself vulnerable to an organization in a certain situation in a relationship with that organization. Trust is

[…] the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party. (1995, p. 712)

In addition, Mayer et al. describe trust as a state of willingness to take a risk in a relationship. Whether or not stakeholders enter such a state depends, firstly, on their propensity to trust—that is, their “[…] general willingness to trust others” (Mayer et al. 1995, p. 715). Secondly, trust is significantly determined by the trustworthiness of the organizations as perceived by the stakeholders. The assessment of an organization’s trustworthiness is made on the basis of factors that comprise the organization’s ability, benevolence, and integrity (Mayer et al. 1995). These antecedents of perceived trustworthiness are also explicitly or implicitly present in many other approaches and studies that describe, among other things, trust in organizations (e.g., Grayson et al. 2008; McKnight and Chervany 2001; Schichtmann 2007; Hon and Grunig 1999, p. 3; Renn and Levine 1991, pp. 179–180). With respect to organizations, the ability factor describes skills, competencies, and characteristics that enable a specific task to be carried out in a specific situation. Consequently, ability is always domain-specific. Benevolence relates to the stakeholders’ perception of whether organizations have a benevolent disposition toward them and act in their interests or whether they will instead pursue egocentric motives. Finally, the assessment of integrity takes into account the principles, values, and beliefs of the organization. The consistency of the actions and reliability are also assigned to this factor. The three antecedents of perceived trustworthiness vary independently of each other and are able to influence each other (Mayer et al. 1995).

Yet, trust, understood as a state, does not automatically lead to the behavioral manifestation of trust. Nevertheless, trust does increase the likelihood that stakeholders will take the risk in the relationship with the organization and make themselves vulnerable to it. Within the model, context factors can have a moderating effect on the connection between trust and the taking of a risk in a relationship. The risk is only taken if trust manifests itself in an action (Mayer et al. 1995).

The construct of perceived trustworthiness, with the factors “ability”, “benevolence”, and “integrity”, is central to trust in organizations, which is why our following analyses will concentrate on this construct. Moreover, we do not restrict the understanding of trust to that of a state. When defining trust, we also include the associated action and refer to it as the act of trust. The process is only complete when the act of trust is performed. We also take the view that the risk does not just have an effect after there has been a willingness to make oneself vulnerable. Rather, the perceived risk in the relationship with the organization is the precondition for the process of trust because it is only in that case that a mechanism for dealing with this risk is required. However, we do agree with the assumption that context factors are able to influence the process of trust (Luhmann 1979, p. 33). Accordingly, whether and how trust is able to arise and exist also depends on the situation and on the context.

It is also necessary to establish that, although the stakeholder expectations primarily relate to the outcome of the organization’s actions, other aspects of the organization also have to be considered in regard to the formation of expectations. Lepsius (1997, pp. 286–288) provides a theoretical description of three dimensions that shape the expectations placed on institutions. In our opinion, these insights can also be applied to organizations. According to this idea, a distinction must be made between an organization’s central idea, its internal order, and its material results. Each of these dimensions can shape stakeholder expectations, and trust can relate to each of these dimensions. If, for example, the results of an organization are not satisfactory, the central idea as well as the internal order or structure of the organization can be called into question. If the central idea is no longer recognized, it becomes difficult to gain trust with the material results of the organization.

Finally, the representatives of the organizations must be considered in the process of trust (Lepsius 1997, p. 289). Giddens (1991, pp. 83–92) describes representatives of abstract systems as access points because it is through them that a connection is established between stakeholders as laypeople and the expert system to which these stakeholders do not belong. If this idea is applied to organizations, it follows that stakeholders have experiences in encounters with representatives of organizations that can form, maintain, or strengthen trust in organizations. However, trust can also be undermined through incorrect conduct by the representatives.Footnote 4 For this reason, this is a place where the organization is vulnerable (Grayson et al. 2008).

3 Measuring Trust in Organizations

Having provided a general outline of stakeholder trust in organizations, we will describe our experiences of measuring trust in organizations in this next part of the contribution. To reduce complexity and because of practical research considerations, we placed two delimitations on the research. Firstly, we are analyzing NPOs, specifically political parties and non-governmental organizations. Secondly, we are concentrating on perceived trustworthiness because this construct is the basis for the process of trust. On the basis of perceived trustworthiness, stakeholders assess whether or not organizations are able to fulfill their expectations. This assessment plays a decisive role in determining whether stakeholders will link their actions with those of the organizations. As soon as they link their actions, the results of the stakeholders’ own acts of trust are dependent on the results of the actions taken by the organizations. Therefore, the assessment of trustworthiness is crucial in determining whether or not the stakeholders will make themselves vulnerable to organizations. In addition to our analysis concerning the question of whether the construct of perceived trustworthiness posited by Mayer et al. (1995) can be applied to organizations, we will also summarize the central findings of the studies.

3.1 Trust in Political Parties

The object of the first study was citizens’ trust in political parties (Wiencierz 2016). For the parties, as political corporate actors, this trust is significant because it leads to the acceptance of the party activities and increases the parties’ legitimacy, ability to act, and effectiveness (Hetherington 1998; Braithwaite and Levi 1998; Benz 2002). Moreover, trust in parties is essential to the functioning of a party democracy such as that of Germany: Without a minimum degree of trust in parties—that is to say, trust placed in intermediary actors or links between citizen and state—a party democracy cannot function smoothly (Benz 2002; Höhne 2006, p. 42; Strøm 2009). There is, therefore, a risk that a lack of trust in parties will diminish not only the ability to act and the effectiveness of the parties, but also, ultimately, the ability to act and the effectiveness of the entire political system.

Because of the significance of interpersonal influence for political opinion formation, we examined whether eligible German voters talked about election advertising by the political parties and about the content of such advertising; we also examined what effect these conversations had on trust in parties. Such an analysis had so far been largely neglected in communication science. For the examination, a three-stage panel survey was used in which N = 496 (18–84 years, M = 47.21, SD = 15.98) for the first survey wave, N = 322 (18–84 years, M = 47.93, SD = 15.83) for the second survey wave, and N = 264 (18–84 years, M = 48.85, SD = 15.49) for the third survey wave. The samples obtained were comparable across all three survey times.

We phrased the items for measuring the trustworthiness of political parties following those of Mayer and Davis (1999). We took these items, which were designed for measuring interpersonal trust in organizations, into consideration as a basis for creating four items in each case for the factors of perceived trustworthiness, namely ability, benevolence, and integrity. The items had to be evaluated by using a five-point Likert scale (1 = ”Strongly disagree” through 5 = ”Strongly agree”). Because voting by citizens is understood as being a cumulative action, which takes into consideration different facts and circumstances, such as financial and economic policy, family policy, environmental policy, etc., we phrased the items in a general way (see Table 1). In some cases, German translations of the original items of Mayer and Davis (1999) were used; in other cases, items had to be adapted. For example, the original item “X is very capable of performing its job.”, which measures ability, was—as per the example of the CDU, the German liberal-conservative Christian democratic party—rephrased as follows: “The CDU demonstrates competence in solving problems relevant to society.” We removed other items, such as “Top management is well qualified.”, because they did not seem plausible for the party context. Instead, we added items, such as “I consider the CDU to be credible.” We assigned the credibility of political parties to the dimension of integrity.

Table 1 Items (translated from the German) and intercorrelations concerning the measurement of the CDU’s trustworthiness (all correlations: p < .001)
Full size table

To test whether the derived items actually did relate to citizens’ existing expectations, we first conducted an exploratory survey to determine the expectations by using five items that were phrased on the basis of the factors of perceived trustworthiness (α = .93). It was apparent that citizens had clear expectations in relation to ability, according to which the parties were supposed to have suitable solutions for the most important problems, for example. In a similar way, it was possible to assign the expectation that the focus should be on citizens’ needs, for example, to the dimension of the benevolence of the political parties; likewise, it was possible to assign the expectation that the party will keep its word to the dimension of integrity.

To test the construct of perceived trustworthiness adopted from Mayer et al., with its three antecedents, we first checked the intercorrelations of the 12 trustworthiness items with the data from the first survey wave with reference to the example of the CDU. The analysis of the correlation (n = 387) according to Pearson, with a listwise deletion, revealed high intercorrelations from r = .60 to r = .82 (see Table 1). Contrary to expectations, intercorrelations did not only occur between the relevant four items of the assumed subdimension, namely ability, benevolence, and integrity, in each case. The results also showed intercorrelations between all 12 items. For a more detailed examination, we carried out a principal component analysis with the 12 items for measuring the perceived trustworthiness of the CDU with an oblique rotation (oblimin) and with a listwise deletion owing to some missing values. The Kaiser-Meyer-Olkin measure indicated very good suitability for a principal component analysis, with a KMO value of .97. The individual values of the KMO statistic for the individual variables, which were taken from the anti-image matrix, were > .95. This exceeded the acceptable value of .5. Bartlett’s test of sphericity χ 2(66) = 4886.47, p < .001, indicated that the variables in the survey population were correlated. Taking into account Kaiser’s criterion (eigenvalue > 1) and the analysis of the scree plot, the analysis indicated a one-factor solution. This factor explained 74.7 % of the variance. Therefore, the results clearly contradicted the assumption of the distinct subdimensions of perceived trustworthiness, namely ability, benevolence, and integrity. The factor loadings ranged from .91 to .82. The Cronbach’s alpha value of .97 indicated very high reliability of the trustworthiness scale with all 12 items taken into account.

We carried out the test of the construct of the three factors of perceived trustworthiness with the data of the CDU and the SPD, the social democrats, from all three survey waves. The results of the six principal component analyses carried out in total also supported a general factor of trustworthiness in each case. The results therefore refuted the assumption of the three distinct factors of perceived trustworthiness, which is why the application of the trustworthiness construct posited by Mayer et al. was unsuccessful in this study.

Nevertheless, it was possible to derive important insights about the significance of trustworthiness for parties from the results that were calculated with the general factorFootnote 5: The results showed a clear connection between the perceived trustworthiness of the parties and the willingness to vote for them, which we interpreted in this study as being the willingness to trust. There was also a clear connection between perceived trustworthiness and the actual vote, cast as a postal vote, which we interpreted as being an act of trust.

In addition, the study made it clear that the subjects talked about the election advertising by the CDU and by the SPD—especially election posters and TV advertisements—and that these conversations had an influence on perceived trustworthiness. The more positive the conversations about election advertising that took place and were investigated as part of the survey, the more positive the evaluation of the trustworthiness of the parties. The more negative the evaluation of these conversations, the more negative the evaluation of trustworthiness. The way in which the parties were talked about depended considerably on party affiliation

3.2 Trust in Non-Governmental Organizations (NGOs)

In the second study, which is presented below, we examined the trust placed by potential donors in NGOs. For NGOs, trust is significant because it generally promotes a willingness to donate and leads to support for an NGO’s activities (e.g., Bekkers 2003; Sargeant and Lee 2004; Beldad et al. 2014). Therefore, trust is a critical variable in the success of NGOs in general and those soliciting donations in particular, and it has a decisive influence on whether reference groups support the relevant organization (Bryce 2007; Lambright et al. 2010). At the same time, NGOs put further donations at risk if stakeholders lose trust in the relevant organization and feel deceived by it (Sisco 2012).

Because interpersonal influence can also be extensive in relation to the effect of donation campaigns, we examined the effect of user comments about campaign motifs published on the social networking site Facebook and of like counts on the perceived trustworthiness of NGOs in our experimental study. The object of investigation was a campaign motif of the World Wide Fund for Nature (WWF). The study was conducted as an online experiment with a 3 (connotation: positive vs. negative vs. positive and negative comments) x 2 (likes: high amount of likes vs. no likes) between-subject design and a control group without comments and likes. The experiment was implemented as an online survey of the German population with interlocked quotas according to age and sex. The sample contained 369 people, who were randomly assigned to the experimental conditions (18–69 years, M = 44.36; SD = 14.06).

We operationalized the expectations placed on the WWF in the donation context examined in the study by postulating that the expectations were that the NGO would use the donations efficiently in accordance with the declared objectives. In contrast to the first study described in this contribution, we incorporated the expectations directly into the trustworthiness items for the experiment. In other words, expectations were formulated simultaneously with the trustworthiness items because the assessment of trustworthiness referred to the expectations (for example, “I expect the WWF to have the competence to save the jaguar.” was an item used to measure the perceived ability of the WWF) (see Table 2).

Table 2 Items (translated from the German) and intercorrelations concerning the measurement of the WWF’s trustworthiness (all correlations: p < .001)
Full size table

The items for measuring the perceived trustworthiness of the WWF were also phrased following Mayer and Davis (1999). We analyzed the trustworthiness scale analogously to the study about the political parties. The item intercorrelations showed a range from r = .29 to r = .86 (see Table 2). A PCA (oblique rotation: oblimin) on the 12 items of the trustworthiness scale (KMO = .95; Bartlett’s test of sphericity: χ 2(66) = 4463.91, p < .001) permitted either a one-factor solution (explaining 68 % of total variance) considering the scree plot, or a two-factor solution according to Kaiser’s criterion with just two items loading clearly on the second factor (explaining 79 % of total variance). Therefore, the application of the trustworthiness construct posited by Mayer et al., with ability, benevolence, and integrity as distinct factors, also did not work in this study. Because of the results, we also decided in this study to consider the general factor solution with factor loadings ranging from .57 to .92. The reliability test revealed a Cronbach’s α of .95.

Despite this limitation, this study also made it clear how important it is for organizations to be seen as trustworthy. Our analyses showed that the willingness to trust exists if the NGO is perceived as trustworthy: The more trustworthy the WWF was perceived to be, the more the participants were willing to donate to the relevant campaign and the more they could envisage inviting others to support this campaign. Nearly half of the participants indicated that they would comment on the WWF donation campaign; the higher they rated the trustworthiness of the WWF, the more positive this comment would be. We did not collect data on the actual act of trust, such as the actual donation, in this study.

Finally, in regard to the study’s research question, the results showed that user comments on social media influence the perceived trustworthiness of NGOs. Negative comments result in a lower level of perceived trustworthiness. No additional effect of the positive comments could be demonstrated when compared with the control group, which showed the pure campaign effect without manipulations. According to these findings, negative comments have a negative bearing on the campaign effects, whereas positive comments have no additional effect. Similarly, with regard to the like counts, no significant effect was identified.

3.3 Critical Discussion

In the studies presented, we tested our approach to trust empirically with reference to the example of NPOs, specifically political parties and an NGO. We gave particular attention to the construct of the perceived trustworthiness of organizations because stakeholders base their trust on this assessment. We tested whether the construct of perceived trustworthiness posited by Mayer et al. (1995), with subdimensions consisting of ability, benevolence, and integrity, could also be applied to organizations. However, the results clearly do not support the assumption of distinct subdimensions comprised of ability, benevolence, and integrity. Moreover, they allow the inference that the items of Mayer and Davis (1999), which were developed to examine interpersonal trust in an organizational context, are not suitable for measuring the trustworthiness of organizations. Admittedly, it must be made clear in this context that the items of Mayer and Davis were not adopted on a one-to-one basis. For instance, some items were rephrased so that they could be applied in a political context to political parties or in a donation context to conservation organizations. Nevertheless, the items used were primarily based on those of Mayer and Davis, which is why doubts remain regarding the suitability of these items for analyzing the perceived trustworthiness of organizations. There is also the question of whether the construct of perceived trustworthiness posited by Mayer et al. can be applied to organizations at all.

As studies already show, organizations can be perceived as being able, benevolent and/or of integrity, or, alternatively, organizations are associated with attributes that can be assigned to these factors. Accordingly, it must also be fundamentally possible to examine trust in organizations following Mayer et al.’s approach. A lack of information on the part of the subjects might be a reason for the failure to identify the three factors of the perceived trustworthiness of organizations on the basis of the items of Mayer and Davis in both studies presented. It is possible that the subjects were not able to make a clear assessment of the ability, benevolence, and integrity of the organization because, in most cases, they did not have sufficiently good knowledge about the organization. Whereas relationships with work colleagues are much clearer, more specific, and more concrete, the experience of dealing with an organization often merely consists of the purchase of products or the use of services. Direct contact with organizations is frequently limited to the use of their services and offerings; in some cases, it also includes interpersonal exchanges with individual representatives of the organizations. Ultimately, information about organizations is mostly perceived indirectly via the mass media. Consequently, it might have been the case that the subjects had difficulty in judging the relevant organizations, as abstract entities, and in anticipating their actions, which meant that they had corresponding difficulty in assessing the presented trustworthiness items.

If it is true, then, that the three subdimensions of perceived trustworthiness often cannot be judged by stakeholders owing to a lack of information, the following questions arise: In the case of incomplete information about an organization, is a generalized, undifferentiated form of perceived trustworthiness the basis of the process of trust? If so, how can this type of trustworthiness be measured? These are key questions, which remain unanswered at this point and will need to be examined in future research on organizational trust. With the general factor that we formed, we did describe a type of undifferentiated trustworthiness pertaining to the relevant organization. However, it is questionable whether this general factor actually includes all of the factors that lead to an undifferentiated assessment of the trustworthiness of an organization. This is because in the case of trust in organizations, in contrast to that of interpersonal trust, factors such as the central idea of an organization, its internal order, and its material results can also play a role and shape the impression of an organization. Therefore, it is necessary to achieve a still much more precise, differentiated observation of the functional relationship between the stakeholder and the organization than the one made in the studies presented in this contribution. It can be useful to inquire about knowledge of the central idea, internal order, and material results of an organization in conjunction with ability, benevolence, and integrity. For instance, an organization can be assessed as being able if stakeholders evaluate the internal order as being efficient and the material results correspond to the expectations held by the stakeholders themselves. The central ideas of an organization and its internal order can play a crucial role in determining benevolence and integrity. If, as already suggested, there is a lack of the necessary knowledge about the organization, it is necessary to examine, by way of a next step, which of these factors are crucial and what a generalized, undifferentiated form of trustworthiness looks like.

Apart from the problems associated with measuring the factors of perceived trustworthiness, there is also the question that arises from a communication science perspective with regard to whether the antecedents of perceived trustworthiness described by Mayer et al. actually do measure trustworthiness or whether they instead measure phenomena such as reputation or image. In the literature, the relationship between these phenomena is consistently described as being very close but with considerable variation between them in terms of detail (e.g., Meijer 2009; Reputation Institute 2015; Einwiller et al. 2005; Koch et al. 2000). From a communication science perspective, Eisenegger’s (2005) approach to reputation is the most theoretically substantial and comprehensive one that can be applied to various organization types. Following Jürgen Habermas’s three-world concept, Eisenegger distinguishes between three types of reputation: functional, social, and expressive reputation (Eisenegger and Imhof 2008). Functional reputation relates to dominant performance goals and success criteria in the relevant social sphere of activity. The central question is whether an organization is perceived as successful within its scope of action. Social reputation describes the organization’s ability to act in accordance with moral and normative requirements and social values. Finally, expressive reputation describes an organization’s emotional power to fascinate and its perceived authenticity. The focus is on the emotional sensation experienced by the stakeholders, which is based in particular on the attractiveness, sympathy, and fascination associated with a person or organization (Eisenegger and Imhof 2008). A comparison of Eisenegger’s concept of reputation (2005; Eisenegger and Imhof 2008) and the model of trust by Mayer et al. (1995) makes it clear that Eisenegger, particularly with his reputation types of functional and social reputation, is essentially describing the factors of perceived trustworthiness put forward by Mayer et al.

In conducting our WWF study, we made an initial attempt to address this research gap and examine the relationship between trustworthiness and reputation. We hypothesized that an NPO’s reputation influences its trustworthiness decisively (Ingenhoff and Sommer 2010). We understand reputation to mean

[…] the generalized, collective assessment of an object (e.g., a company) by its stakeholders […], which they cumulatively obtain through direct and indirect experiences of it and also perceive via media and multipliers (Liehr et al. 2009, p. 4, translated from the German).

A comparison of Eisenegger’s concept of reputation (2005; Eisenegger and Imhof 2008) and the model of trust by Mayer et al. (1995) indicates the similarity of functional reputation to the antecedent ability, whereas social reputation includes the components of the antecedents benevolence and integrity. The decisive difference between these reputation and trust dimensions is seen in the time factor. Functional reputation is the subjective assessment of the generalized, collective perception of ability; social reputation is the assessment of the generalized, collective perception of benevolence and integrity. These assessments relate to public information from the past. In the case of trustworthiness, the assessment of ability, benevolence, and integrity relates to the task of solving a certain problem in the future. Trust is the future-oriented assessment of the extent to which the organization will act in the interests of the stakeholder. In addition, the assessment of emotional influences, such as the evaluation of attractiveness, sympathy, and fascination, is summarized under the category of expressive reputation. The outputs of acts of trust are ultimately experiences that, in turn, are able to influence reputation.

Our analyses showed that reputation can be seen as a predictor of trustworthiness. We conducted a simple regression to specify how much the perceived trustworthiness of the WWF was affected by its reputation. The perceived trustworthiness of the WWF was predicted by the evaluation of reputation. Thus, a prediction of the level of trustworthiness is significantly improved by taking into consideration perceived reputation; however, the proportion of the explained variance was lower than expected (27 %). With regard to the link between the constructs of reputation and trustworthiness, the results indicated that reputation and trustworthiness are two distinct, correlating constructs.

4 Outlook

Further research is still required to clarify how exactly trust in organizations, or, specifically, the trustworthiness of organizations, can be measured. In the studies presented in this contribution, we gave particular consideration to the approach to trust taken by Mayer et al. (1995). In so doing, we encountered problems in measuring the factors of perceived trustworthiness. However, the critical discussion of the operationalization of trustworthiness undertaken in this case should not create the impression that we are arguing for the approach to trust taken by Mayer et al. to be fundamentally rejected for the examination of trust in organizations. This approach represents one of the most well-elaborated approaches to trust to date (Fulmer and Gelfand 2012; Nienaber et al. 2015). Moreover, other approaches to trust also describe factors that can be interpreted as antecedents of the perceived trustworthiness of organizations. These factors include the ability or competence of organizations and their benevolence or fairness as well as their integrity or principles and their values or other equivalents. Despite the diffuse field of research, therefore, there are at least indications that suggest that the factors described in relation to the assessment of perceived trustworthiness play a role.

Instead, it is necessary to reconsider the operationalization of trustworthiness. In this context, moreover, we argue for the relationship between reputation, image, and trustworthiness to be clarified. These constructs are very similar to each other. An engagement with these concepts seems useful in order to achieve a better understanding of the construct of the trustworthiness of organizations and in order to refine the concept of trustworthiness.

Despite the problems encountered in measuring stakeholder trust in organizations, the studies that we have described are able to make a contribution to the discussion about the significance and measurement of trust in organizations and thereby provide implications for practice. Firstly, we have been able to make trust in organizations, such as that placed in political parties and NGOs, at least theoretically concrete. Secondly, we have been able to clearly demonstrate the connection between the trustworthiness of organizations and the willingness to use their goods and services.

Because of the significance of trust, organizations ought to take trust into consideration when planning, implementing, and evaluating their communication efforts, especially in light of the fact that trust primarily arises as a result of the way in which the object of trust presents itself (Luhmann 1979, pp. 39–46; Giddens 1991, pp. 85–86). The strategic communication of organizations, such as that which takes the form of campaigns, therefore plays a central role in building the trust of their stakeholders (Hon and Grunig 1999, pp. 10, 19; Ledingham and Bruning 1998; Ball et al. 2004; Morgan and Hunt 1994). Accordingly, strategic organizational communication can be an important link in building trust relationships between organizations and their stakeholders (Schweer and Thies 2003, p. 128). If the antecedents of perceived trustworthiness presented in this contribution are taken into account, organizations, as potential objects of trust, ought to emphasize their ability, benevolence, and integrity accordingly (Beckert 2002). Although we were unable to analyze these subdimensions in a way that clearly distinguishes them from each other, they are nevertheless included in the general factor of trustworthiness that was created. Moreover, numerous other studies have also identified factors of trust to which the subdimensions ability, benevolence, and integrity can be assigned. Perhaps organizations merely have to convey greater quantities of better targeted information in which they present themselves as being able, benevolent, and of integrity. If that were the case, the possibility of making the relevant assessment would also be greater and trustworthiness could increase.

When carrying out communication activities aimed at appearing trustworthy, organizations should, however, pay attention to interpersonal influence on the effect of their communication activities. In the two studies, we were able to provide a clear illustration of this influence. Likewise, organizations always ought to be aware that trust cannot be demanded. In addition to public image, the most effective method of appearing trustworthy is, ultimately, to act consistently in the interests of the trustor (Kuhlen 2008). If the expectations that have been formed are repeatedly confirmed by the object of trust, trust and the object of trust’s perceived trustworthiness increase: “He who stands by what he has allowed to be known about himself, whether consciously or unconsciously, is worthy of trust.” (Luhmann 1979, p. 39)