The Personality Psychologists’ Perspective
Personality theorists are used to studying trust as a belief, an expectancy, or feeling that is deeply rooted in an individual’s personality (Cheung & Lee, 2006). The personality psychologists’ research interest is dominated by the fundamental question of why some people are just more likely to trust than others. According to several authors, the answer lies in substantial variations in the nature of individuals’ personalities - or more precisely in the peoples’ general willingness to trust others. In general, personality refers an individual’s general style of dealing with the world and in particular with other people - that is the social environment. More precisely, the concept considers individual differences in characteristic patterns of thinking, feeling, and behaving (Kazdin, 2000). Trust is here regarded as a constant personality trait or characteristic (such as sociability or open-mindedness) and hence element of an individual’s personality characteristic that influences a person’s patterns of interaction with the world at large in a variety of situations and across time. Within this context, trust is often defined as a generalized tendency of human beings to believe in the trustworthiness of others (e.g., Das & Teng, 2004), or in other words, trust is a person’s generalized expectancy that behaviours or words of other people can be relied on. Here, the object of trust is not a specific other but other people in general (i.e., the public). Hence, dispositional trust means primarily the general propensity to have the intention to depend on others (Mayer et al., 1995) and this propensity will influence how much trust one has in a trustee prior to data on that particular party being available. Bigley and Pearce (1998, p. 410) comment that “factors exist within individuals that predispose them to trust or distrust others, whom they do not know”. Because trust is regarded as a trait, it is assumed to be independent across situations and persons, being a stable within-party factor which encompasses individual characteristics of the truster that colours the interpretation of situations and actors in situations. It is assumed that people develop trust in varying degrees depending mainly on their personality types but also on personal experiences, prior socializations, and cultural backgrounds (Hofstede, 1980). An example of an extreme case of this kind of trust is what is commonly called “blind trust”. Some people can be observed to repeatedly trust in situations which the majority of people would agree do not warrant trust. Here, trust seems to be a synonym for pure faith. On the other hand, others are unwilling to trust in most situations, regardless of the circumstances that would support doing so (Mayer et al., 1995). These people hold a very sceptical world view.
Rotter (1967) was one of the first scholars to discuss trust as a form of personality. More specifically, he defined what he termed interpersonal trust as a generalized expectancy (see also Rosenberg, 1956). By following the trust typology proposed later by McKnight and Chevany (1996), this type of trust is commonly referred to as dispositional trust (see also Kramer, 1999). Several other authors have referred to this habitual pattern or personality-based form of trust by using different terms for more or less the same concept. In this tradition, trust has been termed generalized trust (Stack, 1978) or trust propensity (e.g., Bigley & Pearce, 1998; Cheung & Lee, 2001; Lee & Turban, 2001; Mayer et al., 1995; McKnight & Chervany, 2001).
Rotter’s (1967, 1971, 1980) work can be regarded as being one of the most representative and influential contributions in this area. Still, numerous researchers elaborate on his conceptual definition and his view widely influenced their conceptualizations of trust in various disciplines (e.g., Zaltman & Moorman, 1988). The scholar (Rotter, 1967, p. 651) himself defines trust as “an expectancy held by an individual or a group that the word, promise, verbal or written statement of another individual or group can be relied upon”. It has to be noted that, although this definition seems to suggest that the scholar is referring to trust for a specific referent, Rotter’s definition is aimed at and focuses on a generalized trust of others that constitutes a relatively stable personality characteristic that a person would presumably carry from one situation (or person) to another (Mayer et al., 1995). Rotter’s interpersonal trust definition therefore describes a general expectancy about the behaviour of generalized others and not a specific person. In his later work, Rotter describes trust in the same tradition as an expectation that the commitments undertaken by another person (or organization) will be fulfilled (Gefen, 2002b; Rotter, 1971).
In a similar vein, Farris and his colleagues (Farris, Senner & Butterfield, 1973, p. 145) define trust as “a personality trait of people interacting with peripheral environment of an organization”. Here, trust is again conceptualized as a trait that leads to a generalized expectation about the trustworthiness of others. In their work, this trait is referred to as the propensity to trust. The same notion has been used by Mayer, Davis, and Schoorman (1995, p. 715), as they define propensity to trust as “a stable within party factor that will affect the likelihood the party will trust”. In line with other definitions of propensity to trust, Bigley and Pearche (1998, p. 410) state that “factors exist within individuals that predispose them to trust or distrust others, whom they do not know”. This leads to the notion that people can be positioned on a spectrum ranging from high to low. Mayer, Davis, and Schoorman (1995) note that at one extreme people grant total blind trust, while individuals at the other end of the continuum will not trust anybody, regardless of the circumstances that make it relevant to trust.
A slightly different perspective has been taken by McKnight and Chervany (2001). They define disposition to trust as “the extent to which one displays a consistent tendency to be willing to depend on others in general across a broad spectrum of situations and persons” (McKnight & Chervany, 2001, p. 45). Within this definition, trust does not literally refer to a personality trait; rather it means that a person has a general disposition to be willing to depend on others. A person’s disposition to trust does not automatically mean that one generally thinks that other individuals are trustworthy or believable, but for whatever the reason, one tends to be willing to depend on others. Dispositional trust hence has nothing to do with having good reasons to trust, but merely the concept represents a general approach towards interacting with the social environment. People scoring high on dispositional trust would trust another person under any circumstances.
Personality psychologists recognize that this individual characteristic is a product of peoples’ lives, where they develop “generalized expectations about the trustworthiness of other people” (Rotter, 1967). Here, the tendency to trust is not based upon experience with or knowledge of a specific trusted party (McKnight et al., 1998), but rather the result of an ongoing lifelong experience (Rotter, 1971), socialization (Fukuyama, 1995) but also learning. To some degree, the concept seems to be linked to naivete. Trust as a personality trait is for the most part developed during childhood as an infant asks for and receives help from benevolent caregivers/parents (Bowlby, 1982; Erikson, 1968; Rotter, 1971; 1980). Erikson (1968) advances that people grow up with a disposition to trust or may develop it later in life. However, as this trait is developed, it is acted out as a generalized reaction to life’s experiences with other people and a mental strategy which causes habitual patterns of behaviour, thought, and emotion. According to the model of human nature advanced by the psychoanalyst Erik Erikson, trust represents the first stage of human development. In developmental child psychology, trust is regarded as the fundamental relationship between a mother and her infant which is based on consistency and regularities. Erikson (1963) suggests that a necessary precondition of a “healthy personality” is what he labels “basic trust”. Due to these early-life experiences, the young child develops a constant tendency to produce trust in a broad range of situations (Rotter, 1967). However, the readiness to trust is subsequently formed and potentially altered by the socialization process. By drawing from the social learning theory, Rotter (1967) argues that individuals typically have different experiences with other people’s fulfilment of their promises and varying reinforcements of previous social interactions (i.e., reinforcement history). Therefore, people develop different expectancies that specific (negative/positive) reinforcements will occur when something is promised by other people. The scholar furthermore states that an individual’s expectations that the promise of a social agent will be kept will be generalized from one social agent (trust object) to another. As a consequence, people differ in generalized expectations that the oral or written statements of other people can be relied upon. This is also certainly true online.
The importance of social interaction for the development of trust is also advocated by Worchel (1979), who views trust as a function of an individual’s environment. He emphasizes that trust is a personality trait or response pattern which is the result of early socialization. Hence, “a person exhibits a disposition to trust to the extent that she/he demonstrates a consistent tendency to be willing to depend on others across a broad spectrum of situations and [persons]” (McKnight et al., 1998). Analogously, Tyler and Kramer (1996) note that an individual’s willingness to trust is based on the person’s estimation of the probability that those trusted will reciprocate the trust.
The above argumentation outlines the potential of personality characteristics to impact a person’s trust response and to influence peoples’ amount and level of trust in their interaction partners. However, results on the influence of generalized trust are controversial, as some empirical studies suggest that in a particular situation the influence of trust as a context-free generalized trait is limited compared to the attributes of the trustee (e.g., Schlenker et al. 1973). In general, research has shown that the impact is particularly relevant in cases when the former has limited knowledge about the latter or in the absence of available or experiential information on which to base a judgment (Hofstede, 1980). Thus, as an determinant of (situational) trust, disposition to trust is most influential in the early phases of a relational exchange when the trusting parties are still widely unfamiliar with each other, and before extensive ongoing relationships provide the necessary background information for the formation of other trustbuilding beliefs (Gefen, 2000). Rotter (1971, p. 445) comments that “the more novel the situation, the greater weight generalized expectancies have”. Further, he notes, “the situation partially determines the response, and the theory predicts that situations of considerable familiarity are less predictable from a generalized tendency than those involving more novelty”.
The importance of dispositional trust is likely to diminish when the person interacts more with the trusted party (McKnight et al., 1998; Rotter, 1971). Or, as McKnight and Chervany (2001, p. 45) put it, “disposition to trust will affect trust in a specific other (interpersonal trust), but only when novel situations arise in which the other and the situation are unfamiliar”. Therefore, the kind of relationship and its development seem to determine the impact of generalized trust. Accordingly, it can be assumed that, for online review communication, dispositional trust is a key variable. However, it is also reasonable to argue that people develop generalized trust that is specific to the context (e.g., using online reviews) and which coexists with the more abstract context-free personal trust disposition. Research in the field of personality trait also tells us that people regularly develop trust that is independent from a specific situation and predisposes them to think, feel, and act in a certain stable way.
For a long time, personality psychologists have refrained from developing a psychometric measure of trust as a personality characteristic. A notable exception has been made by Rotter (1967), who applied his definition of trust to the development of the Interpersonal trust scale (ITS) - one of the most prominent psychometric scaling approaches for trust measurement still. By conducting a series of experiments, the researcher was able to compose and test a trust scale which enables researchers to measure individuals’ general beliefs about the expected reliability of others, or as Rotter (1967; p. 653) puts it, the instrument can be applied to measure the individual’s “general optimism regarding the society”. Hence, solely the aspect of constant positive expectation is emphasized. The ITS is a self-report inventory that consists of 25 Likert- format items that essentially deal with the perceived credibility of social agents. Rotter was striving to deduce a person’s tendency to trust society by asking people to characterize certain agents critical for society. More specifically, the ITS measures a person’s general tendency to trust a wide variety of social entities, including but not limited to teachers, politicians, physicians, classmates, journalists, salespersons and strangers. In order to develop an instrument that is able to capture a generalized expectation towards society, Rotter particularly focused on selecting questions on social objects that do not represent close social relationships, such as parents, friends, and life partners (i.e., primary group relationships), but rather constitute “generalized” others. The scale includes 12 positive and 13 negative formulated expectations. Many of the items deal - as noted - with the credibility of social agents and include items such as “most elected officials are really sincere in their campaign promises” or “in dealing with strangers one is better off to be cautious until they have provided evidence that they are trustworthy”. Other phrases reflect an individual’s general expectancy on others’ motives and reliability (e.g., “it is safe to believe that in spite of what people say, most people are primarily interested in their own welfare” or “parents usually can be relied upon to keep their promises”), and general scepticism about the future of society (e.g., “hypocrisy is on the increase in our society”). Meanwhile, the scale is frequently applied to differentiate “high trusters” (those who are mostly trusting others) from “low trusters”. In his subsequent research, Rotter (1980) found significant differences between these two groups, as he provides some evidence that people
who trust more are less likely to lie and are additionally less likely to cheat or steal. Moreover, these people are less likely to be unhappy, conflicted or maladjusted. In general, it appears that high trusters have many social and psychological advantages - an insight that has been very recently explored by evolutionary psychologists.
Besides Rotter’s (1967) approach, several other researchers have discussed trust in similar ways and have attempted to measure a person’s general tendency to trust others. For instance, Dasgupta (1988) proposes a treatment of trust that includes generalized expectations of others; for example, “can I trust people to come to my rescue if I am about to drown?” (p. 53). However, the most commonly used measures of trust include Rosenberg’s (1956) Faith-in-people scale, the trust facet of the NEO PI-R Agreeableness scale (Costa & McCrae, 1992a), Mayer and Davis’s (1999) Trust propensity scale, and several other self-report personality inventories incorporating constructs similar to trust. For example, the trust-defensiveness subscale of the Comprey personality scale (Comrey, 1970), the trust in humanity subscale of personal orientation dimensions (Shorstrom, 1975), and the trustworthiness subscale of philosophies of human nature (Wrightsman, 1974). While these personality researchers do not offer an explicit definition of the meaning of trust, their contributions to the measurement of trust commonly imply that they regard trust as a faith in humanity. These measures contribute to an understanding of the nature of trust in that they broaden the construct by recognizing the necessity to introduce general optimism and confidence for the concept of trustworthiness and honesty of people in general.
While Rotter’s (1967) ITS represents one of the classical works in the field of dispositional trust, the scale itself has also been criticized. For example, Heretick (1981) researched on the constructs of suspicion and trust by relating them to gender differences. She states that the ITS scale generally suffers from a lack of discriminant validity and being confounded with another construct referred to an individual’s expectation about the social environment, namely locus of control. In particular, Heretick’s criticism refers to the circumstance that the ITS, while being supposed to measure interpersonal trust, shows relatively high correlations with Rotter’s (1966) earlier developed Internal-External locus of control (I-E) scale. Motivated by the need for a scale that exhibits psychometrically independent measurement characteristics, Heretick’s (1981) 6-item Trust-Suspiciousness scale (T-SS) was developed to measure “expectancies concerning the motives of other individuals” (p. 269). The T-SS comprises mostly items from the Minnesota Multiphasic Personality Inventory (MMPI), including items striving to measure others’ honesty and selfishness, as well as the individual’s general caution towards others’ trustworthiness (e.g., “it is safer to trust nobody”). The author herself states that the scale exhibits reasonable reliability and only weak correlations with I-E. However, additional evidence concerning its validity has not been reported.
Rosenberg (1957) introduced a 5-item scale that measures a person’s general level of confidence in the trustworthiness, honesty, goodness, generosity, and brotherliness of people in general. The scale includes items such as “if you don’t watch yourself, people will take advantage of you”. Kanter and Mirvis (1989) developed for their survey on cynicism a 7-item measure of mistrust, which contains items such as “most people will tell a lie if they can get away with it”. Subsequent research reports (Gurtman, 1992) that the measure is reliable and highly correlated with scores on Rotter’s (1967) ITS. Another stream of literature implies that trust and mistrust are two distinct concepts rather than simply a bipolar construct. For example, Wrightsman (1974) constructed the 84-item Philosophies of human nature measure, which incorporates a trustworthiness subscale (Omodei & McLennan, 2000).
More recently, Omodei and McLennan (2000) introduced the scale of Global interpersonal mistrust-trust (Global IMTM), which strives to measure an individual’s global interpersonal mistrust. The construct is conceptualized by the authors as a general mistrust of the motives of others in situations related to one’s well-being: “A general tendency to view others as mean, selfish, malevolent, or unreliable people who are, thus, not to be depended on to treat one well” (p. 283).
The above-mentioned scales have in common that they all treat trust as a uni-dimensional construct without underlying factors. However, the individual items typically used seem to reflect several distinct elements of trust. For example, even though Rotter (1967) does not argue for multi-dimensionality of the ITS, it is apparent that the scale consists of three underlying dimensions, or factors: (i) credibility of social institutions; (ii) belief in others’ sincerity; and (iii) caution (Kaplan, 1973). This parallels the often-applied method by researchers to operationalize trust in the context of the individual scales. The concept of trust has often been broadened to include items that are usually associated with the concepts of fearfulness, cynicism, caution, optimism, or confidence in social institutions (e.g., Omodei & McLennan, 2000). Only recently, the multi-dimensional character of trust as a personality characteristic has been recognized. For example, McKnight et al. (1998) identify two sub-constructs of disposition to trust: First, faith in humanity, which refers to one’s belief that others are usually upright, well-meaning and reliable. Second, trusting stance, which means that “people believe that they will obtain better interpersonal outcomes by dealing with others as though they are well-meaning and reliable, regardless of whether those others are reliable or not” (Beldad et al., 2010). While faith in humanity deals with the attributes of general others, a trusting stance is a more personal approach to dealing with others (McKnight & Cherany, 2002; McKnight et al., 2002b). Besides telling us that trust can be regarded as a stable trait (be it towards the generalized others or online reviewers in general), these approaches also give us the insight that this trait is likely to be multifaceted and more complex than at first glance.