Sage Journals: Discover world-class research

Abstract

This study investigates how general trust is measured and expressed across psychological and survey research, using diverse methods and levels of analysis. In Study 1, self- and informant-reports on the A1: Trust facet of the NEO PI-R/3 and the Single-Item General Trust (SIGT) scale were collected from a sample of 3232 adults. Study 2 compiled country-level aggregate scores for the same measures from various sources, covering 50 countries for self-reports and 49 for informant-reports. The two measures—SIGT and the A1: Trust facet—were significantly and positively correlated across both self- and informant-reports, and at both individual and country levels. They also exhibited broadly similar patterns across gender, age, and education. Taken together, the findings help bridge gaps in the literature by providing cross-disciplinary evidence for general trust as a construct and by identifying key similarities and differences in its measurement and demographic correlates across psychological and survey-based approaches.

Plain language summary

This study examines how two common measures of general trust compare across different fields, methods, and levels of analysis. General trust refers to the belief that most people—including strangers and large groups of unfamiliar individuals—can be trusted. We compared one trust scale commonly used in personality research with another often applied in survey research. Using both self- and informant-reports, we explored how these measures relate to each other at individual and country levels. We also examined whether they show similar or distinct patterns across demographic groups. Our findings highlight areas of agreement and difference between these measures, helping to unify different approaches to studying trust and improve its measurement across disciplines.

Keywords

general trust Single-Item General Trust NEO PI-R/3 A1: Trust self- and informant-reports Estonian Biobank World Values Survey

Trust is a concept that has attracted increasing interest over the past few decades from scholars across various disciplines, including sociology, psychology, political science, economics, philosophy, and law. Despite notable interdisciplinary differences, several common themes have emerged regarding how trust is viewed, often revolving around a person’s willingness to be vulnerable to another person or actor and the expectation of reciprocity in uncertain situations (Schilke et al., 2021). An often cited definition of trust by Rousseau and colleagues (1998) describes it as “a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behavior of another” (p. 395). Similarly, it has been argued that trust is simply “an expectation of beneficent reciprocity from others in uncertain or risky situations” (Foddy & Yamagishi, 2009, p. 17).

While a certain degree of consensus may have been reached on the core meaning of trust, research on trust remains fragmented and largely compartmentalized within specific disciplines, with few studies approaching the concept through the lenses of different academic fields (cf. Schilke et al., 2021). Furthermore, as different academic disciplines investigate trust using varying methodologies, an important question arises about to what extent do these different approaches measure the same construct?

Research on trust is further complicated by the existence of various types of trust. The first important distinction is between social and political (or institutional) trust (Delhey, 2014). Social trust refers to trust in people, while political/institutional trust pertains to trust or confidence in social and political institutions and their members. Social trust, which is the primary focus of this study, is further divided into two main forms—general (or generalized) trust and particular (or particularized) trust (Sztompka, 1999; Yamagishi, 2011). General trust refers to trust in people in general, it reflects a person’s belief that most people can be trusted, including large categories of unfamiliar people and strangers. In contrast, particular trust denotes trust in people we know or are familiar with (Delhey, 2014; Uslaner, 2002). This paper exclusively focuses on general trust—its conceptualization and measurement in psychological versus survey research—with the aim of assessing whether the two disciplines tap into the same underlying construct and of helping to defragment the existing body of research on general trust.

Two perspectives on general trust: Dispositional vs experiential

Social scientists recognized long ago that societies and economies thrive when people trust other members of their community and people in general (Fukuyama, 1995; Smith, 1982). A common understanding is that the entire fabric of our day-to-day lives, and our social order, rests on general trust—almost all of our decisions involve trusting someone else. The more complex the society, the greater the dependence on others and, consequently, the greater the need to trust others (Rotter, 1971). Remove this trust, as Smith (1982) argued more than 250 years ago, and society will collapse into a state of war of all against all. Therefore, general trust is often seen as the glue that holds societies together, builds bridges between people, and enhances cooperation.

There is growing empirical evidence that higher levels of general trust in a society are associated with many positive outcomes such as greater economic prosperity, more effective governance, reduced crime levels, lower corruption, and higher life satisfaction (see Beilmann et al., 2021; Schilke et al., 2021, for a review). Cross-country differences in general trust have been extensively documented, with Nordic countries exhibiting the highest levels of general trust globally (Bjørnskov, 2007; Delhey & Newton, 2005; Holmberg & Rothstein, 2020). This substantial variation in general trust levels across countries raises important questions about the causes and foundations of general trust. Broadly speaking, as with other human traits, the answer boils down to the nature versus nurture debate (Dinesen & Bekkers, 2017).

Dispositional perspective

In the dispositional perspective, general trust is viewed as a relatively stable personality trait influenced by genetic factors, early socialization experiences, or the combination of both. This perspective is predominant in psychological research on general trust but extends beyond that field as well (cf. Uslaner, 2017). Within the framework of the Five-Factor Model of personality (FFM; McCrae & John, 1992), which includes Neuroticism, Extraversion, Openness to Experience, Agreeableness, and Conscientiousness as the main factors of personality differences, general (or interpersonal) trust is considered an essential part of Agreeableness. Agreeable people are usually described as altruistic, sympathetic, cooperative, and trustful (Costa & McCrae, 1992). Behavioral genetics studies estimate that the average heritability estimate for Agreeableness is about 40% (Vukasović & Bratko, 2015), while the genetic influence on A1: Trust, a facet of Agreeableness is about 30% (Jang et al., 1996). The remaining variance in Agreeableness is believed to be due to nonshared environmental influences that make individuals within the same family more different from each other (Bleidorn et al., 2014). Although personality traits—including Agreeableness and its facet A1: Trust—can and do change across the lifespan due to intrinsic maturation, significant life events, and possibly other environmental influences (Bleidorn et al., 2018; Soto et al., 2011; Specht et al., 2011), these changes are relatively small (Bleidorn et al., 2022).

Experiential perspective

The second perspective, known as the experiential perspective, views general trust as an attitude that is shaped by people’s life experiences over the course of their lives (Dinesen & Bekkers, 2017). Central to this view is the belief that individuals learn to trust or distrust based on their actual experiences and interactions with others, that is, with people they know, strangers, or representatives of various institutions (Freitag & Traunmüller, 2009; Glanville & Paxton, 2007; Hardin, 2002; Paxton & Glanville, 2015). Empirical support for this perspective often comes from large-scale survey findings that demonstrate how general trust levels can and do change over time, sometimes quite rapidly (Beilmann et al., 2021; Sønderskov & Dinesen, 2014). Trustworthy state institutions and good governance appear to play a major role in generating general trust—when people perceive state officials as honest and trustworthy, this expectation often extends to others in their community and to people in general (Beilmann et al., 2021; Mewes et al., 2021; Seifert, 2018; Sønderskov & Dinesen, 2014). Heritability estimates of general trust range widely, from 5% to 50% (Hiraishi et al., 2008; Sturgis et al., 2010; Van Lange et al., 2014), such broad differences likely “owing to the definition of the trust phenotype, its underlying complex genetics, the assumptions of twin study designs, and the diversity of study populations” (Sequeros et al., 2024, p. 2).

In sum, as with most individual traits, it is reasonable to assume that general trust has a biological or genetic component—some people are naturally more trusting than others—but it is also malleable to environmental and social influences.

Measurement of general trust in psychological research

One of the most widely used personality inventories is the NEO Personality Inventory (NEO PI) and its later revisions, the NEO PI-R (Costa & McCrae, 1992) and the NEO PI-3 (McCrae et al., 2005). These latter versions consist of 240 items grouped into five factors: Neuroticism, Extraversion, Openness to Experience, Agreeableness, and Conscientiousness. Each of these domains, or global traits, is further represented by six facet scales of eight items. The selection and specification of facet scales are not uniquely determined, allowing for multiple valid approaches to their definition (Costa & McCrae, 1998; McCrae & Costa, 1992).

Agreeableness is typically conceptualized as a dimension of interpersonal tendencies, characterized by altruism, trust, sympathy, and compassion (Costa & McCrae, 1992). In the NEO-PI-R/3, the Agreeableness dimension is represented by the following six facets: A1: Trust, A2: Straightforwardness, A3: Altruism, A4: Compliance, A5: Modesty, and A6: Tender-Mindedness (Costa & McCrae, 1992). By contrast, the Big Five Inventory (BFI-2) measures Agreeableness with only three subscales: A1: Compassion, A2: Respectfulness, and A3: Trust (Soto & John, 2017). Typical items that measure dispositional or personality trust include statements like “My first reaction is to trust people,” “I believe that most people are basically well-intentioned,” and “I think most of the people I deal with are honest and trustworthy.” Across studies and countries, women usually score higher than men on Agreeableness (Costa et al., 2001; Murphy et al., 2021; Schmitt et al., 2008) and sometimes also on A1: Trust (McCrae et al., 2005b) though not consistently (Kajonius & Johnson, 2018). Older individuals tend to have higher scores of Agreeableness (Costa et al., 2000; McCrae et al., 1999) and A1: Trust (Mõttus et al., 2015) compared to younger individuals. We found no research explicitly examining the A1: Trust facet in relation to educational outcomes (e.g., academic performance, years of schooling), but prior studies have shown a positive relationship between Agreeableness and academic performance, including educational attainment (Poropat, 2009). This may be due to higher levels of university education fostering higher Agreeableness (Kassenboehmer et al., 2018) or because Agreeableness facilitates cooperation with learning processes (De Raad & Schouwenburg, 1996). However, the overall strength of these relationships is relatively modest (Poropat, 2009), and some studies have even found relationships in the opposite direction (Nießen et al., 2020).

Measurement of general trust in survey research

The classic single-item general trust (SIGT) measure often used in survey research is: “Generally speaking, would you say that most people can be trusted or that you need to be very careful in dealing with people?” According to Uslaner (2018), this question was first formulated by Elisabeth Noelle-Neumann in Germany in 1948 and was later used by Rosenberg (1956) in his study of misanthropy and political ideology. It has since been used in numerous large-scale surveys, including the American National Election Studies (since 1964), the American General Social Survey (since 1972), the World Values Survey (WVS) and the European Value Survey (EVS) (since 1980s). In most surveys, including the WVS, participants are asked to indicate their level of general trust on a dichotomous scale by choosing either “Most people can be trusted” or “Need to be very careful.” Research shows that older individuals (Li & Fung, 2013; Poulin & Haase, 2015) and men (IPSOS, 2022) tend to have higher levels of general trust than younger people and women, respectively, when measured using the SIGT scale. Education has been identified as the single most important predictor of general trust (Huang et al., 2009; Uslaner, 1999), with higher levels of education consistently associated with higher levels of general trust. However, some recent studies suggest that the positive relationship holds primarily at the individual level in stable, low-conflict societies (Wu, 2021) or when quality of public institutions is high (Charron & Rothstein, 2016).

Over the years, various concerns about the reliability and validity of the SIGT measure have been raised (e.g., Lundmark et al., 2016; Miller & Mitamura, 2003; Smith, 1997), but these concerns have been firmly refuted by its supporters (Uslaner, 2012, 2018). Some have suggested replacing the dichotomous scale with a 7 or 11-point Likert-type scale (Lundmark et al., 2016), as seen in the European Social Survey and the Swiss Household Survey, or using a multiple-item scale rather than a single item, due to the reputation of single-item measures having lower or uncertain reliability (cf. Allen et al., 2022) and concerns that the single item may not fully capture the concept of general trust (Robbins, 2022). Additionally, it has been noted that the meaning of “most people” might vary across countries, potentially biasing cross-country comparisons of general trust (Delhey et al., 2011; Reeskens, 2013; van Hoorn, 2014). This suggests that people in different countries may interpret “most people” differently, depending on their radius of trust—that is, which groups of people they include within that concept (Fukuyama, 1995; Realo et al., 2008; van Hoorn, 2014).

Aims of the present study

Based on the evidence presented, the main aim of the study is to assess the convergence and demographic consistency of two general trust measures used in psychological and survey research—the A1: Trust facet of the NEO PI-R/3 and the SIGT scale—across multiple methods (self- and informant-reports) and levels of analysis (individual and country).

To achieve this aim, in Study 1 we examine individual-level correlations between self- and informant-reports of the NEO PI-3 A1: Trust facet and the SIGT scale in a large-scale, population-based sample of adults from the Estonian Biobank cohort (Vaht et al., 2024), who completed both measures. If the two general trust measures assess the same underlying construct, they should be significantly correlated regardless of the method of assessment (i.e., self-vs informant-reports).

Study 1 also examines demographic consistency—specifically, gender, age, and education differences—in the two general trust measures, using both self- and informant-reports. This analysis assesses whether the measures exhibit similar or distinct patterns across demographic groups.

As noted earlier, when assessed with the SIGT scale in social surveys, men, older individuals, and those with higher educational attainment tend to report higher levels of general trust than women, younger individuals, and those with lower levels of education. Similarly, levels of Agreeableness and A1: Trust facet typically increase from adolescence to adulthood. However, women consistently score higher than men on Agreeableness, including the A1: Trust facet, while the association between Agreeableness and educational attainment is relatively modest. Taken together, these results suggest that general trust, as measured in psychological and survey research, may yield distinct relationship patterns across demographic factors such as gender and education, even if they capture similar underlying constructs.

In Study 2, we examine the correlations between country-level aggregate scores on the self-reported SIGT scale from the WVS Wave 7 and the EVS 2017 joint dataset (EVS/WVS, 2022) and aggregate self-reports (Allik et al., 2017) and informant-reports (McCrae & Terracciano, 2008, Appendix 10A) of the NEO-PI-R/3 A1: Trust facet scale, across samples of 50 and 49 countries, respectively. The substantial overlap of countries across the different datasets allows us to examine whether populations in countries with higher levels of aggregated general trust, as measured by the SIGT scale, also exhibit higher levels of personality-based trust, as measured by the A1: Trust facet, and vice versa. Even though it has been repeatedly demonstrated that relationships between variables at one level of analysis (e.g., the individual level) do not necessarily hold at another level (e.g., the aggregate country level), either for conceptual or statistical reasons (cf. Mõttus et al., 2010; Ostroff, 1993), one would still expect countries that score high on one measure of general trust (e.g., the SIGT scale) to also score high on other measures of the same construct, such as self- and informant-rated aggregate scores of A1: Trust.

Similarly to Study 1, we examine whether the two general trust measures show similar or distinct patterns across gender, age, and education. Due to data limitations, sample’s age and gender characteristics are only available on the self-reported SIGT scale from the WVS Wave 7 and the EVS 2017 joint dataset (EVS/WVS, 2022) and A1: Trust facet informant-reports (McCrae & Terracciano, 2008, Appendix 10A). As for education, we will test whether the established relationship between general trust and educational attainment also holds true for aggregate country-level self- and informant-rated A1: Trust facet scale scores. To this aim, a component of the Human Development Index (HDI) measuring the average number of years of education received by people aged 25 years and older in a given country will be used (UNDP, 2025).

We acknowledge that ideally, comparing the relationships between the two general trust measures—the A1: Trust facet of the NEO PI-R/3 and the SIGT scale—across individual and country levels would require individual-level data from multiple countries to enable multilevel structural equation modeling or similar analyses. However, our dataset includes individual-level data for both measures from only one country, while personality data for other countries consist solely of country-level aggregate scores compiled from various sources. As a result, we are unable to empirically test whether the same measurement model holds across levels, and instead, we explore patterns of trust at the individual and country levels descriptively.

Study 1

Method

Participants

Participants for this study were drawn from the Estonian Biobank cohort which is a large-scale sample of the Estonian adult population (Leitsalu et al., 2015; Milani et al., 2025). Of these, 3604 participants had available NEO PI-3 data, either through self-reports, informant-reports, or both (Vaht et al., 2024). The final sample for this study consists of 3232 individuals (1916 women, 59.3%) with a mean age of 46.5 years (SD = 17.0, ranging from 18 to 91 years), for whom both self- and informant-reports on the A1: Trust facet scale and on the SIGT scale were available. For subsequent analyses, participants were divided into five approximately equal-sized age groups: 18–30 years (n = 719, 22.2%); 31–40 years (n = 614, 19.0%); 41–50 years (n = 567, 17.5%); 51–64 years (n = 721, 22.3%); and 65 years and older (n = 611, 18.9%). Approximately 40% of the participants (n = 1297) had higher education, 913 (28.25%) had secondary vocational education, 782 (24.40%) had secondary education, and 240 (7.43%) had basic education. For later analyses, three educational levels were used for simplicity, with the secondary and secondary vocational education groups merged (52.44%).

Among the informants, 2252 (69.7%) were women and 916 (28.3%) were men, 64 did not report their gender. The mean age of informants was 41.7 years (SD = 15.9). About 42% of the informants had higher education. On average, informants had known the target participants for 23.1 years (SD = 15.0). The majority of informants were either spouses or partners (46.8%), parents (16.8%), or friends (15.6%).

Data collection took place between 2008 and 2014 (see Vaht et al., 2024, for more details). About 78% of participants and informants completed the questionnaire on paper, while approximately 22% completed it online (the online version was offered only from 2011 onwards).

Measures

NEO PI-3 A1: Trust facet scale

The Estonian version of the NEO Personality Inventory-3 (NEO PI-3; McCrae et al., 2005) is a slightly modified version of the Estonian NEO PI-R (Kallasmaa et al., 2000). It consists of 240 items that measure five domains—Neuroticism, Extraversion, Openness to Experience, Agreeableness, and Conscientiousness—and 30 lower-level personality facets. The items are answered on a five-point scale ranging from 0 = “strongly disagree” to 4 = “strongly agree.” The A1: Trust facet is one of the six scales of Agreeableness and consists of eight items. High scorers on this facet have a disposition to believe that others are honest and well-intentioned, while low scorers tend to be cynical and skeptical, often assuming that others may be dishonest or dangerous (Costa & McCrae, 1992, p. 17). Cronbach alphas were .75 and .81 for self- and informant-rated A1: Trust facet scales, respectively.

Single-item general trust (SIGT) scale

Participants were asked to answer the question “Generally speaking, would you say that most people can be trusted” on an eleven-point response scale ranging from 0 = “most people cannot be trusted” to 10 = “most people can be trusted.” Informants were asked to respond to the following question about the target: “In your opinion, how would they respond to the following question: “Can most people be trusted?””. The response scale ranged from 0 = “most people cannot be trusted” to 10 = “most people can be trusted.” Higher scores indicate higher levels of general trust.

Results

The correlations between the SIGT scale and the A1: Trust facet scale scores were r = .57, 95% CI [.55, .59] and r = .62, 95% CI [.60, .64] for self- and informant-reports, respectively (see Table 1). Although both correlations were positive and statistically significant at p < .001, they differed significantly from each other, as indicated by Fisher’s z = −3.11, p (two-tailed) = .002. For comparison, the correlations between self- and informant-reports were much smaller, although still statistically significant for both the SIGT (r = .34, 95% CI [.31, .37]) and the A1: Trust facet scale (r = .40, 95% CI [.37, .43]), p < .001. Overall, these findings suggest that the SIGT and A1: Trust scales show a relatively strong degree of convergent validity, indicating they may capture overlapping aspects of general trust despite being derived from different research traditions.
Table 1.
Pearson product-moment correlations with 95% confidence intervals among self- and informant-reported SIGT scale and NEO PI-3 A1: Trust facet scale scores in the Estonian Biobank sample (Study 1).

A1: Trust (s) SIGT (s) A1: Trust (i) SIGT (i)

A1: Trust (s) -

SIGT (s) .57 [.55, .59] -

A1: Trust (i) .40 [.37, .43] .29 [.26, .32] -

SIGT (i) .32 [.29, .35] .34 [.31, .37] .62 [.60, .64] -

Note. N = 3232. A1: Trust = NEO PI-3 A1: Trust facet scale; SIGT = Single-item general trust scale (“Generally speaking, would you say that most people can be trusted”); (s) = self-reports; (i) = informant-reports. 95% confidence intervals are indicated in parentheses. All correlations are significant at p < .001. Correlations between the SIGT and A1: Trust scales for self- and informant-reports are shown in bold.

Next, we examined gender differences in all four sets of general trust ratings. As shown in Table S1 of the Supplemental Material, women had significantly higher scores on the A1: Trust facet scale than men, both in self- and informant ratings at p < .001, Cohen’s d = .24, 95% CI [.17, .32] and d = .13, 95% CI [.05, .20], respectively. There were no significant gender differences in self-reported SIGT scale scores (p = .656); however, interestingly, women scored significantly higher than men on this measure when informant-reports were used (p = .038), with Cohen’s d = .08, 95% CI [.00, .15].

As a general trend, age was positively correlated with various measures of general trust, indicating that older individuals tend to report higher levels of trust. The correlations of age with the SIGT scale were statistically significant at p < .001 for both self- and informant reports: r = .09, 95% CI [.05, .12] and r = .06, 95% CI [.02, .09], respectively. Age was also correlated with self-report scores on the A1: Trust facet scale (r = .05, 95% CI [.01, .08], p = .006) but not with informant-reported A1: Trust facet scale scores (r = .00, 95% CI [.00, .03], p = .880).

We also conducted two series of MANOVAs with gender and age group as independent and either A1: Trust subscales (self- and informant-reports) or the SIGT scales (self- and informant-reports) as dependent measures. The results are shown in Table S2 of the Supplemental Material. The interaction of gender and age group was significant only for the informant-rated scores of the A1: Trust facet scale, F (4) = 3.16, p = .011, η_p² = .004. As shown in Figure 1, the informant rated A1: Trust scores for men over 40 years of age were significantly lower than those for women of the same age.
Figure 1.
Estimated marginal means for NEO PI-3 A1: Trust and SIGT scores by gender and age group (self- and informant-reports; Estonian Biobank sample, Study 1). Note. A1: Trust = NEO PI-3 A1: Trust facet scale; SIGT = Single-item general trust scale (“Generally speaking, would you say that most people can be trusted”); (s) = self-reports; (i) = informant-reports; error bars indicate 95% confidence intervals.

Finally, we examined the relationship between the education level and four general trust measures. The results are presented in Table S2 of the Supplemental Material. For all four indicators, general trust increased with higher levels of education, with all group differences statistically significant at p < .001. The effect sizes (η²) ranged from .049, 95% CI [.035, .064], for self-reported A1: Trust, to .016, 95% CI [.008, .025], for informant-reported SIGT.

Overall, the results of Study 1 indicate that the two general trust measures from psychological and survey research are significantly correlated at the individual level and demonstrate roughly similar response patterns across gender, age, and education in both self- and informant-reports.

Study 2

Method

Measures and data

NEO PI-R/3 A1: Trust facet scale (self-reports)

The mean self-rated NEO-PI-R/3 scores for 30 subscales have been reported for 71,870 participants from 76 samples across different countries (Allik et al., 2017). These mean scores were converted to T-scores, standardized to a mean of 50 and a standard deviation of 10. For the conversion, the mean score across all samples was used (Allik et al., 2017, Appendix). Multiple samples were aggregated to compute single country scores for the current study: two samples for Burkina Faso, Russia, South Africa, and South Korea; three samples for Czechia, Estonia, India, Norway, the Philippines, Spain, Switzerland, and the United States; and four for Italy and Sweden. Data from Great Britain were used for Northern Ireland, and data from former Yugoslavia for Serbia. Sample sizes ranged from 71 (Zimbabwe) to 11,674 (Estonia), with an average of 1244.91 (SD = 1915.00). The sample characteristics in terms of age and gender were not provided by Allik et al. (2017). For 50 out of the 56 countries, both the self-reported NEO PI-R/3 A1: Trust facet scale scores and the WVS/EVS SIGT scale scores were available. The corresponding scores are provided in Table S4 of the Supplemental Material.¹

NEO PI-R A1: Trust facet scale (informant-reports)

In a large-scale cross-cultural study of personality traits (McCrae et al., 2005a; McCrae et al., 2005b), college students from 51 countries and territories were asked to identify someone they knew well—either an adult over the age of 40 or a college-aged individual (18–21 years)—and rate them using the informant version of the NEO PI-R. In total, 12,156 target participants were assessed. Overall mean factor T-scores were calculated using combined international norms. A few years later, mean country-level T-scores for all 30 facets—including A1: Trust—were reported (McCrae & Terracciano, 2008, Appendix 10A). Data from the German- and French-speaking Swiss samples were combined to produce a single aggregate score for Switzerland in the present study. Data from the WVS/EVS SIGT scale were available for all countries except Botswana, yielding an overlapping sample of 49 countries.

Sample sizes for all countries and territories were taken from Table 2 of McCrae, Terracciano, et al. (2005a). The sample sizes ranged from 106 (Northern Ireland) to 919 (United States), with a mean of 243.10 (SD = 142.25). Sample characteristics in terms of age and gender were obtained from Table 1 of McCrae, Terracciano, et al. (2005b). Across all samples, the mean age of targets was 35.27 years (SD = 2.32), ranging from 30 years (Canada) to 46 years (Czechia). The proportion of male participants ranged from 26% (Canada) to 57% (Morocco), with a mean of 48.65% (SD = 5.41). Age and gender data for Iranian participants are missing, as Iran was added to the analysis at a later stage.
Table 2.
Pearson product-moment correlations among self-reported SIGT scores (WVS7/EVS2017), NEO PI-R/3 A1: trust facet scores (self- and informant-reports), and the HDI mean years of schooling index at the country level (Study 2).

SIGT (s) A1: Trust (s) A1: Trust (i) Mean years of schooling

SIGT (s) -

A1: Trust (s) .60^* [.38, .75] -

N = 50

A1: Trust (i) .56^* [.34, .73] .55^* [.26, .75] -

N = 49 N = 34

Mean years of schooling .49^* [.32, .63] .56^* [.35, .72] .42^ [.16, .63] -

N = 96 N = 55 N = 49

Note. WVS7 = World Values Survey seventh data collection wave; EVS2017 = European Values Survey 2017 data collection wave; NEO PI-R/3 = The NEO Personality Inventory-R/3; A1: Trust = NEO PI-3 A1: Trust facet scale; SIGT = Single-item general trust scale (“Generally speaking, would you say that most people can be trusted or that you need be very careful in dealing with people”); (s) = self-reports; (i) = informant-reports; Mean years of schooling = Average number of years of education received by people aged 25 years and older, one of the two education indicators used in calculating the Education Index component of the Human Development Index (HDI) 2023. 95% confidence intervals are indicated in parentheses.

^p < .001.

^p < .01.

Single-item general trust (SIGT) scale

We used the WVS Wave 7 (WVS7) and the EVS 2017 (EVS2017) joint integrated dataset, which includes data for 90 countries collected in 2017–2022 (EVS/WVS, 2022), and was downloaded from https://www.worldvaluessurvey.org/WVSEVSjoint2017.jsp. Our primary focus was the SIGT measure “Generally speaking, would you say that most people can be trusted or that you need be very careful in dealing with people” (A165). Participants were asked to choose between two alternatives “Most people can be trusted” (1) and “Need to be very careful” (0).

For nine countries (i.e., Algeria, Belgium, Burkina Faso, India, Kuwait, Mali, Malta, South Africa, and Uganda) where either self- or informant-reported NEO PI-R/3 A1: Trust facet scores were available (please see the description above) but which did not participate in the WVS7/EVS2017 data collection, the SIGT scale scores were taken from the most recent data collection wave of the WVS/EVS in which they had participated, respectively (https://www.worldvaluessurvey.org/WVSEVStrend.jsp). For further details, please see Table S4 of the Supplemental Material. In total, country-level aggregate SIGT scale scores were available for 99 countries (N = 170,902), indicating the percentage of respondents who agreed with the statement “Most people can be trusted.” The mean score of the SIGT scale was .25 (SD = .18) ranging from .02 in Zimbabwe to .77 in Denmark. The sample sizes ranged from 447 (Northern Ireland) to 4549 (Netherlands), with a mean of 1726.28 (SD = 858.86). Across all samples, the mean age of targets was 44.55 years (SD = 6.26), ranging from 30.74 years (Kenya) to 57.68 years (New Zealand). The proportion of male participants ranged from 32% (Uruguay) to 64% (Kuwait), with a mean of 46.70% (SD = 0.05).

HDI mean years of schooling

Mean years of schooling is one of the two education indicators used in calculating the Education Index component of the Human Development Index (HDI), developed by the United Nations Development Programme (UNDP), the other component being expected years of schooling. Mean years of schooling indicates the “average number of years of education received by people ages 25 and older, converted from education attainment levels using official durations of each level” (UNDP, 2025, p. 278). The latest HDI data from 2023 were used (UNDP, 2025). The United Kingdom’s score was applied to both Great Britain and Northern Ireland.

All country-level scores are shown in Table S4 of the Supplemental Material.

Results

The correlations among different general trust measures at the country level are shown in Table 2. The correlations between the self-reported SIGT scale and the self- and informant-rated A1: Trust facet scale scores were r(50) = .60, 95% CI [.38, .75] and r(49) = .56, 95% CI [.34, .73], respectively, both significant at p < .001 (see Figure S1 in the Supplemental Material). The correlations did not differ significantly from each other, Fisher’s z = .29, p (two-tailed) = .771. In comparison, the aggregate country-level self- and informant-rated A1: Trust facet scale scores were correlated at r(34) = .55, 95% CI [.26, .75], p < .001.

Next, we examined the relationships between the self-reported SIGT scores and informant-rated A1: Trust scores with the mean age and percentage of males in each national sample. As noted above, we were unable to perform the same analysis using the self-rated A1: Trust scores because Allik et al. (2017) did not report the age and gender characteristics of the samples in their study.

Consistent with the individual-level findings, SIGT scores were positively correlated with the mean age of samples, r(99) = .55, 95% CI [.39, .67], p < .001. In other words, general trust levels tended to be higher in samples with older average age.² In contrast, the correlation between SIGT scores and the proportion of males in the sample was not significant, r(99) = −.06, 95% CI [−.26, .14], p = .542. This suggests that gender composition in national samples does not systematically influence general trust levels as measured by the SIGT scale.

Informant-rated A1: Trust scores were not significantly associated with either the average age of the rated targets in each sample or the percentage of male targets, r(49) = .09, 95% CI [−.19, .37], p = .524, and r(49) = −.09, 95% CI [−.36, .20], p = .534, respectively.

Finally, we examined the relationship between three aggregate country-level indicators of general trust—self-reported SIGT scores, and self- and informant-reported A1: Trust scores—and one component of the HDI: mean years of schooling, which reflects the average number of years of education received by people aged 25 and older. This indicator served as a proxy for a country’s overall level of education. All three trust indicators were significantly associated with mean years of schooling, ps < .001, with Pearson correlation coefficients ranging from r = .49, 95% CI [.32, .63], to r = .56, 95% CI [.35, .72] (see Table 2). To assess whether these correlations differed significantly from one another, we conducted pairwise comparisons using Fisher’s r-to-z transformation. None of the resulting z statistics were statistically significant, ps > .05, indicating that the strength of the associations with average years of education did not differ meaningfully across the three trust indicators.

However, upon inspecting scatterplots of these relationships, it became evident that the associations between the country-level aggregate general trust measures and the mean years of schooling indicator were nonlinear (see Figure S2 in the Supplemental Materials). The non-linear, U-shaped relationship was most pronounced for general self-reported trust as measured by the SIGT scale. The best-fitting equation was: SIGT(s) = 0.00654(Mean Years of Schooling)²−0.09(Mean Years of Schooling)+0.38, which explained 38% of the variance. As shown in Figure 2, SIGT scores were lowest in countries where the average number of years of education among individuals aged 25 and older was approximately 8–10 years. However, beginning at approximately 12 years of schooling, each additional year is associated with an accelerating increase in general trust.
Figure 2.
Self-reported general trust as measured by the SIGT scale (WVS7/EVS2017) increases exponentially with the average years of education received by people aged 25 and older (Study 2). Note.* WVS7 = World Values Survey seventh data collection wave (2017–2022); EVS2017 = European Values Survey 2017 data collection wave; SIGT (s) = Self-reported single-item general trust scale (“Generally speaking, would you say that most people can be trusted or that you need be very careful in dealing with people”); Mean Years of Schooling (HDI 2023) = Average number of years of education received by people aged 25 years and older, one of the two education indicators used in calculating the Education Index component of the Human Development Index (HDI) 2023.

General discussion

In recent decades, there has been a proliferation of interest in research on social trust and related concepts such as social capital from academics, policymakers, and practitioners alike. Social trust is recognized as one of the essential facilitators of cooperation in nearly any kind of social interaction. As such, trust is vital for people’s social lives as it “helps to convert the Hobbesian state of nature from something that is nasty, brutish, and short, into something that is more pleasant, more efficient, and altogether more peaceful” (Newton, 2011, p. 202). Social trust manifests in various forms and can be understood both as an individual characteristic and as a societal property (Kim & Kim, 2021), while being susceptible to both genetic and environmental influences (Sequeros et al., 2024). However, the growing popularity of social trust as a research topic across various academic disciplines has led to challenges, such as the concept being defined, operationalized, and measured in numerous ways. Consequently, research on social trust is fragmented across fields like philosophy, psychology, sociology, political science, and economics, often with limited cross-disciplinary interaction. This study aimed to address this issue by focusing on one specific type of social trust: general trust.

Main findings of the study

Although general trust measured in social surveys and the propensity to trust measured by personality questionnaires appear similar at first glance, we are not aware of any studies examining the relationships of the two measures. The findings of the study strongly support the agreement between two general trust measures from psychological and survey research at both the individual and country levels, using both self-reports and informant-reports. Furthermore, the NEO PI-R/3 A1: Trust facet and the SIGT scale scores exhibited broadly similar response patterns across gender, age, and education in both self- and informant-reports, at both the individual and country levels. However, a few notable variations in how the general trust measures related to demographic characteristics did emerge, which we discuss in the following sections.

Individual-level data (Study 1)

Previous research has shown that general trust levels tend to increase with age when assessed using the SIGT measure commonly employed in survey research (IPSOS, 2022; Li & Fung, 2013). Older individuals have also been found to score higher on Agreeableness (Bleidorn et al., 2022; McCrae, Costa, Hrebickova, et al., 2004; McCrae et al., 1999) and on the A1: Trust facet scale (Mõttus et al., 2015) than younger individuals. Therefore, we did not anticipate differences between the A1: Trust facet scale and the SIGT in their relationships with age. However, we predicted that these measures might exhibit distinct response signatures for men and women, as men have shown to score lower on Agreeableness (Costa et al., 2001; Mac Giolla & Kajonius, 2019; Murphy et al., 2021; Schmitt et al., 2008) and on A1: Trust facet scale (McCrae et al., 2005b) than women across studies and countries, but higher on general trust when assessed using the SIGT scale (IPSOS, 2022).

We were also uncertain about the relationship between general trust and educational attainment, as we found no prior research explicitly examining the A1: Trust facet in relation to educational outcomes (e.g., academic performance or years of schooling). Existing studies on Agreeableness and academic performance have reported small and often conflicting associations (Nießen et al., 2020; Poropat, 2009). However, findings from survey-based research suggested a strong positive relationship between general trust and educational attainment, particularly in societies characterized by stability, low levels of conflict (Wu, 2021), and high-quality public institutions (Charron & Rothstein, 2016)—such as the context of our sample.

The results of Study 1 partially supported our expectations. Generally, age was positively and significantly correlated with both general trust measures across different types of reports, except for the informant-reported A1: Trust facet scale scores. Regarding gender differences, women indeed scored significantly higher on A1: Trust facet scale than men both according to self- and informant ratings. Interestingly, women also scored higher on the SIGT score than men, but this was only true for informant-reports; no significant gender differences were found in self-reported SIGT scale scores.

When examining the interaction effects of gender and age on the different general trust scores, a significant gender gap emerged in the informant-ratings of the A1: Trust facet scale among individuals over the age of 40. Middle-aged and older men were described by their partners, friends, and family members as considerably less honest, trusting, and well-intentioned than women of the same age. Curiously, there were no gender differences in the informant-ratings of A1: Trust facet scale among men under 40 years of age. Whether this is a true age effect or a generational/cohort effect driven by changing gender stereotypes and gender role behaviors needs to be determined by future research.

As for educational attainment, the relationship was consistently strong and positive across both measures of general trust and rating sources: individuals with higher levels of education tended to report higher general trust scores.

Country-level data (Study 2)

As noted above, self-reported aggregate SIGT scores and both self- and informant-reported A1: Trust scores were significantly positively correlated at the country level (Study 2), further supporting their construct convergence. Given that indicators of the same construct do not always correlate strongly at the country level (Allik & Realo, 2017), this finding reinforces the conclusion that in countries where individuals report higher levels of general trust as measured by the SIGT scale, they also tend to view others as honest and well-intentioned, as reflected by the NEO PI-R/3 A1: Trust facet scale (across both self- and informant-reports).

As in Study 1, we examined whether the two general trust measures exhibited similar or distinct patterns across demographic characteristics such as age, gender, and education. However, due to data limitations, these analyses could only be conducted for the self-reported SIGT and informant-reported A1: Trust scales. Neither measure showed significant associations with the gender composition of national samples, suggesting no substantial country-level gender differences in general trust.

General trust levels, as measured by SIGT, tended to be higher in samples with an older average age, whereas informant-rated A1: Trust scores were not significantly associated with the average age of the rated individuals. This lack of association may be partly attributable to the study design, which imposed age-related criteria that restricted variability; participants were asked to rate someone they knew well who was either 18–21 years old or over the age of 40 (McCrae et al., 2005b). However, when controlling for countries’ levels of development, the correlation between mean age and SIGT scores was also no longer significant (p = .132). This suggests that the initially observed relationship between SIGT and sample age may have been confounded by national development levels, as higher-income, more developed countries tend to have aging populations (Chen et al., 2025) as well as higher levels of general trust (Özcan & Bjørnskov, 2011).

With regard to education, we found that the positive relationship with educational attainment—measured by the HDI mean years of schooling indicator (UNDP, 2025)—applied consistently across all three country-level aggregate indicators of general trust: self-reported SIGT and both self- and informant-reported A1: Trust scores. In other words, countries with higher average levels of education also tend to exhibit higher levels of general trust, regardless of the measure. However, closer inspection revealed that these associations—particularly between the SIGT scale and the HDI mean years of schooling indicator—were best characterized as non-linear in nature. As illustrated in Figure 2, countries with very low average levels of schooling—such as Burkina Faso, Ethiopia, and Mali (approximately 2 years)—exhibit levels of general trust comparable to those in countries with much higher educational attainment, like Croatia and Serbia (approximately 12 years). Interestingly, the lowest levels of general trust are observed in countries with average schooling of around 8–9 years. Moreover, starting at around 12 years of schooling, each additional year is associated with an accelerating increase in general trust.

As with the observed relationship between country-level age and general trust, these findings should be interpreted with caution. First, it is possible that a third variable may underlie this association. Second, the country-level pattern may be obscured by heterogeneous within-country relationships between education and general trust, as suggested by prior research (Charron & Rothstein, 2016; Wu, 2021).

Strengths and limitations

To the best of our knowledge, this is the first study to examine how general trust is measured and expressed across psychological and survey research using different methods and multiple levels of analysis. Our findings thus help bridge gaps in the literature by providing cross-disciplinary evidence on general trust and identifying key similarities and differences in its measurement and demographic correlates across psychological and survey-based approaches.

Our study also contributes to the discussion surrounding the validity and reliability of SIGT measures in psychological and social survey research. The SIGT scale, like many other single-item measures, has faced criticism for potentially low or uncertain reliability, and for failing to capture the various dimensions and meanings associated with the concept of general trust (Allen et al., 2022; Robbins, 2022). It has also been criticized for potential cultural biases, as the phrase “most people” can hold different meanings for individuals across various cultures, depending on their radius of trust (Delhey et al., 2011; Fukuyama, 1995; Realo et al., 2008; Reeskens, 2013; van Hoorn, 2014). Our results demonstrated that across both self- and informant-reports, as well as various levels of analysis, the SIGT measure was strongly correlated with the 8-item A1: Trust facet scale. While this does not resolve the issue of “most people” meaning different things to different people in different countries, it does support the reliability and validity of the SIGT scale in capturing people’s faith in, and belief in, others’ benevolence, intentions, and goodwill.

Another strength and novelty of our study is the use of both self- and informant-ratings of general trust, which is a standard practice in personality research (Allik et al., 2016; Dobewall et al., 2014; Mõttus et al., 2020) but, to our knowledge, has not previously been applied to the measurement of general trust with the SIGT scale. The use of different modes of measurement has been widely advocated to mitigate the effects of various self-report and response biases such as socially desirable, extreme, or neutral responding, and acquiescence (Mõttus et al., 2012), as well as to avoid or minimize potential common method variance biases (Podsakoff et al., 2003). The correlation between the self- and informant-ratings of the SIGT scale (r = .34 [.31, .37]) in Study 1 was slightly lower than that of the A1: Trust facet scale (r = .40 [.37, .43]), the latter being very similar to what has typically been reported for Agreeableness in previous research (Allik et al., 2010; McCrae, Costa, Martin, et al., 2004).

Our study does not come without limitations. First, the sample in Study 1 was cross-sectional and therefore, we were unable to disentangle age, generational, and period effects on levels of general trust (cf. Realo & Dobewall, 2011). Second, the Study 1 sample was drawn from a highly developed country (UNDP, 2025) and was relatively homogenous in terms of ethnic and other sociodemographic characteristics. Future studies should examine whether our findings can be generalized to other countries and ethnic groups with varying degrees of human development and economic prosperity. Third, due to data limitations (e.g., missing informant-rated country-level SIGT scores and missing sample characteristics for self-reported aggregate country-level A1: Trust scores), we were unable to replicate all individual-level analyses in Study 2. Last but not least, as acknowledged in the Introduction, we could not examine the relationships between psychological and survey measures of general trust across both individual and country levels simultaneously, as this would require access to individual-level data from multiple countries. Consequently, we were unable to empirically test whether the same measurement model of general trust holds across levels. Instead, our analyses explore patterns of trust at the individual and country levels descriptively. Future research using individual-level data from multiple countries would be valuable for formally testing measurement equivalence across levels of analysis.

Conclusions

In conclusion, our findings help unify the fragmented research on general trust. By demonstrating strong agreement between two widely used measures—the A1: Trust facet of the NEO PI-R/3, common in personality research, and the SIGT scale, frequently used in survey research—across methods and levels of analysis, this study contributes to the integration of concepts and methodologies across disciplinary boundaries.

Supplemental Material

Supplemental Material - Investigating the patterns and measurement of general trust across disciplines, methods, and levels of analysis

Supplemental Material for Investigating the patterns and measurement of general trust across disciplines, methods, and levels of analysis by Anu Realo and Jüri Allik in European Journal of Personality

	A1: Trust (s)	SIGT (s)	A1: Trust (i)	SIGT (i)
A1: Trust (s)	-
SIGT (s)	.57 [.55, .59]	-
A1: Trust (i)	.40 [.37, .43]	.29 [.26, .32]	-
SIGT (i)	.32 [.29, .35]	.34 [.31, .37]	.62 [.60, .64]	-

	SIGT (s)	A1: Trust (s)	A1: Trust (i)	Mean years of schooling
SIGT (s)	-
A1: Trust (s)	.60^*** [.38, .75]	-
N = 50
A1: Trust (i)	.56^*** [.34, .73]	.55^*** [.26, .75]	-
N = 49	N = 34
Mean years of schooling	.49^*** [.32, .63]	.56^*** [.35, .72]	.42^** [.16, .63]	-
N = 96	N = 55	N = 49

Footnotes

Acknowledgments

Anu Realo was a KONE Foundation Fellow at the Helsinki Collegium for Advanced Studies during the preparation of the first draft of the manuscript.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Open science statement

The studies were exploratory and, therefore, not pre-registered. The data analyzed in Study 1 cannot be shared openly due to access restrictions. Procedures for requesting access to the Estonian Biobank data are outlined at https://genomics.ut.ee/en/content/estonian-biobank. The data analyzed in Study 2 were compiled from previously published sources and are fully presented in .

Ethical approval

Data collection for Study 1 was approved by the Research Ethics Committee of the University of Tartu (approvals: 236/M-29, 14 May 2014; 206/T-4, 22 Aug 2011; 170/T-38, 28 April 2008; 166/T-21, 17 Dec 2007).

ORCID iDs

Anu Realo

Jüri Allik

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Allen

M. S.

Iliescu

Greiff

(2022). Single item measures in psychological science. European Journal of Psychological Assessment, 38(1), 1–5. https://doi.org/10.1027/1015-5759/a000699

Allik

Church

A. T.

Ortiz

F. A.

Rossier

Hřebíčková

de Fruyt

Realo

McCrae

R. R.

(2017). Mean profiles of the NEO personality inventory. Journal of Cross-Cultural Psychology, 48(3), 457–478. https://doi.org/10.1177/0022022116628671

Allik

de Vries

R. E.

Realo

(2016). Why are moderators of self-other agreement difficult to establish? Journal of Research in Personality, 63, 72–83. https://doi.org/10.1016/j.jrp.2016.05.013

Allik

Realo

(2017). How valid are culture-level mean personality scores? In Church

A. T.

(Ed.), The praeger handbook of personality across cultures: Vol. 1 - Trait psychology across cultures (pp. 193–224). Praeger Publishers.

Allik

Realo

Mõttus

Esko

Pullat

Metspalu

(2010). Variance determines self-observer agreement on the big five personality traits. Journal of Research in Personality, 44(4), 421–426. https://doi.org/10.1016/j.jrp.2010.04.005

Beilmann

Lilleoja

Realo

(2021). Learning to trust: Trends in generalized social trust in the three Baltic countries from 1990 to 2018. In Almakaeva

A. M. A.

Wilkes

(Eds.), Social capital and subjective well-being: Insights from cross-cultural studies (pp. 19–43). Springer.

Bjørnskov

(2007). Determinants of generalized trust: A cross-country comparison. Public Choice, 130(1), 1–21. https://doi.org/10.1007/s11127-006-9069-1

Bleidorn

Hopwood

C. J.

Lucas

R. E.

(2018). Life events and personality trait change. Journal of Personality, 86(1), 83–96. https://doi.org/10.1111/jopy.12286

Bleidorn

Kandler

Caspi

(2014). The behavioural genetics of personality development in adulthood—Classic, contemporary, and future trends. European Journal of Personality, 28(3), 244–255. https://doi.org/10.1002/per.1957

10.

Bleidorn

Schwaba

Zheng

A. Q.

Hopwood

C. J.

Sosa

S. S.

Roberts

B. W.

Briley

D. A.

(2022). Personality stability and change: A meta-analysis of longitudinal studies. Psychological Bulletin, 148(7-8), 588–619. https://doi.org/10.1037/bul0000365

11.

Charron

Rothstein

(2016). Does education lead to higher generalized trust? The importance of quality of government. International Journal of Educational Development, 50, 59–73. https://doi.org/10.1016/j.ijedudev.2016.05.009

12.

Chen

Lim

Koh

Beard

Rowe

J. W.

Rowe

Antonucci

Berkman

Börsch-Supan

Carstensen

Goldman

Fried

Furstenberg

Jackson

Kohli

Olshansky

Rehkopf

Rother

Zissimopoulos

for the Research Network on an Aging

(2025). A global analysis of adaptation to societal aging across low-middle- and high-income countries using the global aging society index. Nature Aging, 5(1), 113–121. https://doi.org/10.1038/s43587-024-00772-3

13.

Costa

P. T.

Jr. McCrae

R. R.

(1992). Revised NEO Personality Inventory (NEO PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Psychological Assessment Resources.

14.

Costa

P. T.

Jr. McCrae

R. R.

(1998). Six approaches to the explication of facet-level traits: Examples from conscientiousness. European Journal of Personality, 12(2), 117–134. https://doi.org/10.1002/(SICI)1099-0984(199803/04)12:2<117::AID-PER295>3.0.CO;2-C

15.

Costa

P. T.

Jr. McCrae

R. R.

Martin

T. A.

Oryol

V. E.

Senin

I. G.

Rukavishnikov

A. A.

Shimonaka

Nakazato

Gondo

Takayama

Allik

Kallasmaa

Realo

Molfese

V. J.

Molfese

D. L.

(2000). Personality development from adoloscence through adulthood: Further cross-cultural comparisons of age differences. In Molfese

V. J.

Molfese

D. L.

(Eds.), Temperament and personality development across life span (pp. 235–252). Lawrence Erlbaum Associates.

16.

Costa

P. T.

Jr. Terracciano

McCrae

R. R.

(2001). Gender differences in personality traits across cultures: Robust and surprising findings. Journal of Personality and Social Psychology, 81(2), 322–331. https://doi.org/10.1037/0022-3514.81.2.322

17.

Delhey

(2014). Trust. In Michalos

A. C.

(Ed.), Encyclopedia of quality of life and well-being research (pp. 6743–6747). Springer. https://doi.org/10.1007/978-94-007-0753-5_3067

18.

Delhey

Newton

(2005). Predicting cross-national levels of social trust: Global pattern or Nordic exceptionalism? European Sociological Review, 21(4), 311–327. https://doi.org/10.1093/esr/jci022

19.

Delhey

Newton

Welzel

(2011). How general is trust in “most people”? Solving the radius of trust problem. American Sociological Review, 76(5), 786–807. https://doi.org/10.1177/0003122411420817

20.

De Raad

Schouwenburg

H. C.

(1996). Personality in learning and education: A review. European Journal of Personality, 10(5), 303–336. https://doi.org/10.1002/(SICI)1099-0984(199612)10:5<303::AID-PER262>3.0.CO;2-2

21.

Dinesen

P. T.

Bekkers

(2017). The foundations of individuals’ generalized social trust: A review. In Van Lange

P. A. M.

Rockenbach

Yamagishi

(Eds.), Trust in social dilemmas (pp. 77–99). Oxford University Press.

22.

Dobewall

Aavik

Konstabel

Schwartz

Realo

(2014). A comparison of self-other agreement in personal values versus the big five personality traits. Journal of Research in Personality, 50, 1–10. https://doi.org/10.1016/j.jrp.2014.01.004

23.

EVS/WVS . (2022). European values study and world values survey: joint EVS/WVS 2017-2022 dataset (joint EVS/WVS). JD systems institute & WVSA (Version 4.0.0). https://doi.org/10.14281/18241.21

24.

Foddy

Yamagishi

(2009). Group-based trust. In Cook

K. S.

Levi

Hardin

(Eds.), Whom can we trust? How groups, networks, and institutions make trust possible (pp. 17–41). Russell Sage Foundation. https://books.google.co.uk/books?id=YZzH7ftmBscC

25.

Freitag

Traunmüller

(2009). Spheres of trust: An empirical analysis of the foundations of particularised and generalised trust. European Journal of Political Research, 48(6), 782–803. https://doi.org/10.1111/j.1475-6765.2009.00849.x

26.

Fukuyama

(1995). Trust: The social virtues and the creation of prosperity. Free Press.

27.

Glanville

J. L.

Paxton

(2007). How do we learn to trust? A confirmatory tetrad analysis of the sources of generalized trust. Social Psychology Quarterly, 70(3), 230–242. https://doi.org/10.1177/019027250707000303

28.

Hardin

(2002). Trust and trustworthiness. Russell Sage Foundation.

29.

Hiraishi

Yamagata

Shikishima

Ando

(2008). Maintenance of genetic variation in personality through control of mental mechanisms: A test of trust, extraversion, and agreeableness. Evolution and Human Behavior, 29(2), 79–85. https://doi.org/10.1016/j.evolhumbehav.2007.07.004

30.

Holmberg

Rothstein

(2020). Social trust: The Nordic gold? QoG Working Paper Series 2020:1. https://gupea.ub.gu.se/handle/2077/63131.

31.

Huang

Maassen van den Brink

Groot

(2009). A meta-analysis of the effect of education on social capital. Economics of Education Review, 28(4), 454–464. https://doi.org/10.1016/j.econedurev.2008.03.004

32.

IPSOS . (2022). Interpersonal trust across the world. IPSOS. https://www.ipsos.com/en/interpersonal-trust-across-the-world

33.

Jang

K. L.

Livesley

W. J.

Vernon

P. A.

(1996). Heritability of the big five personality dimensions and their facets: A twin study. Journal of Personality, 64(3), 577–591. https://doi.org/10.1111/j.1467-6494.1996.tb00522.x

34.

Kajonius

P. J.

Johnson

(2018). Sex differences in 30 facets of the five factor model of personality in the large public (N = 320,128). Personality and Individual Differences, 129, 126–130. https://doi.org/10.1016/j.paid.2018.03.026

35.

Kallasmaa

Allik

Realo

McCrae

R. R.

(2000). The Estonian version of the NEO-PI-R: An examination of universal and culture-specific aspects of the five-factor model. European Journal of Personality, 14(3), 265–278. https://doi.org/10.1002/1099-0984(200005/06)14:3<265::aid-per376>3.3.co;2-2

36.

Kassenboehmer

S. C.

Leung

Schurer

(2018). University education and non-cognitive skill development. Oxford Economic Papers, 70(2), 538–562. https://doi.org/10.1093/oep/gpy002

37.

Kim

S. H.

Kim

(2021). Social trust as an individual characteristic or societal property? International Review of Psycho-Analysis, 26(1), 1–17. https://doi.org/10.1080/12294659.2020.1834677

38.

Leitsalu

Haller

Esko

Tammesoo

M. L.

Alavere

Snieder

Perola

P. C.

Mägi

Milani

Fischer

Metspalu

(2015). Cohort profile: Estonian biobank of the Estonian genome center, University of Tartu. International Journal of Epidemiology, 44(4), 1137–1147. https://doi.org/10.1093/ije/dyt268

39.

Fung

H. H.

(2013). Age differences in trust: An investigation across 38 countries. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 68(3), 347–355. https://doi.org/10.1093/geronb/gbs072

40.

Lundmark

Gilljam

Dahlberg

(2016). Measuring generalized trust: An examination of question wording and the number of scale points. Public Opinion Quarterly, 80(1), 26–43. https://doi.org/10.1093/poq/nfv042

41.

Mac Giolla

Kajonius

P. J.

(2019). Sex differences in personality are larger in gender equal countries: Replicating and extending a surprising finding. International Journal of Psychology: Journal International de Psychologie, 54(6), 705–711. https://doi.org/10.1002/ijop.12529

42.

McCrae

R. R.

Costa

P. T.

Jr. (1992). Discriminant validity of NEO-PI R facet scales. Educational and Psychological Measurement, 52(1), 229–237. https://doi.org/10.1177/001316449205200128

43.

McCrae

R. R.

Costa

P. T.

Jr. Hrebickova

Urbanek

Martin

T. A.

Oryol

V. E.

Rukavishnikov

A. A.

Senin

I. G.

(2004). Age differences in personality traits across cultures: Self-report and observer perspectives. European Journal of Personality, 18(2), 143–157. https://doi.org/10.1002/per.510

44.

McCrae

R. R.

Costa

P. T.

Jr. Martin

T. A.

Oryol

V. E.

Rukavishnikov

A. A.

Senin

I. G.

Hřebı́čková

Urbánek

(2004). Consensual validation of personality traits across cultures. Journal of Research in Personality, 38(2), 179–201. https://doi.org/10.1016/S0092-6566(03)00056-4

45.

McCrae

R. R.

Costa

P. T.

Jr. Pedroso de Lima

Simoes

Ostendorf

Angleitner

Marusic

Bratko

Caprara

G. V.

Barbaranelli

Chae

J. H.

Piedmont

R. L.

(1999). Age differences in personality across the adult life span: Parallels in five cultures. Developmental Psychology, 35(2), 466–477. https://doi.org/10.1037/0012-1649.35.2.466

46.

McCrae

R. R.

Costa

P. T.

Martin

T. A.

(2005). The NEO-PI-3: A more readable revised neo personality inventory. Journal of Personality Assessment, 84(3), 261–270. https://doi.org/10.1207/s15327752jpa8403_05

47.

McCrae

R. R.

John

O. P.

(1992). An introduction to the five-factor model and its applications. Journal of Personality, 60(2), 175–215. https://doi.org/10.1111/j.1467-6494.1992.tb00970.x

48.

McCrae

R. R.

Terracciano

78 Members of the Personality Profiles of Cultures Project . (2005a). Personality profiles of cultures: Aggregate personality traits. Journal of Personality and Social Psychology, 89(3), 407–425. https://doi.org/10.1037/0022-3514.89.3.407

49.

McCrae

R. R.

Terracciano

78 Members of the Personality Profiles of Cultures Project . (2005b). Universal features of personality traits from the observer’s perspective: Data from 50 cultures. Journal of Personality and Social Psychology, 88(3), 547–561. https://doi.org/10.1037/0022-3514.88.3.547

50.

McCrae

R. R.

Terracciano

(2008). The five-Factor model and its correlates in individuals and cultures. In van de Vijver

van Hemert

D. A.

Poortinga

Y. H.

(Eds.), Multilevel analysis of individuals and cultures (pp. 249–283). Lawrence Erlbaum Associates.

51.

Mewes

Fairbrother

Giordano

G. N.

Wilkes

(2021). Experiences matter: A longitudinal study of individual-level sources of declining social trust in the United States. Social Science Research, 95, 102537. https://doi.org/10.1016/j.ssresearch.2021.102537

52.

Milani

Alver

Laur

Reisberg

Haller

Aasmets

Abner

Alavere

Allik

Annilo

Fischer

Hofmeister

Hudjashov

Jõeloo

Kals

Karo-Astover

Kasela

Kolde

Krebs

Metspalu

(2025). The Estonian Biobank’s journey from biobanking to personalized medicine. Nature Communications, 16(1), 3270. https://doi.org/10.1038/s41467-025-58465-3

53.

Miller

A. S.

Mitamura

(2003). Are surveys on trust trustworthy? Social Psychology Quarterly, 66(1), 62–70. https://doi.org/10.2307/3090141

54.

Mõttus

Allik

Realo

(2010). An attempt to validate national mean scores of conscientiousness: No necessarily paradoxical findings. Journal of Research in Personality, 44(5), 630–640. https://doi.org/10.1016/j.jrp.2010.08.005

55.

Mõttus

Allik

Realo

(2020). Do self-reports and informant-ratings measure the same personality constructs? European Journal of Psychological Assessment, 36(2), 289–295. https://doi.org/10.1027/1015-5759/a000516

56.

Mõttus

Allik

Realo

Rossier

Zecca

Ah-Kion

Amoussou-Yeye

Backstrom

Barkauskiene

Barry

Bhowon

Bjorklund

Bochaver

de Bruin

Cabrera

H. F.

Chen

S. X.

Church

A. T.

Cisse

D. D.

Johnson

(2012). The effect of response style on self-reported conscientiousness across 20 countries. Personality and Social Psychology Bulletin, 38(11), 1423–1436. https://doi.org/10.1177/0146167212451275

57.

Mõttus

Realo

Allik

Esko

Metspalu

Johnson

(2015). Within-trait heterogeneity in age group differences in personality domains and facets: Implications for the development and coherence of personality traits. PLoS One, 10(3), Article e0119667. https://doi.org/10.1371/journal.pone.0119667

58.

Murphy

S. A.

Fisher

P. A.

Robie

(2021). International comparison of gender differences in the five-factor model of personality: An investigation across 105 countries. Journal of Research in Personality, 90, 104047. https://doi.org/10.1016/j.jrp.2020.104047

59.

Newton

(2011). Trust, social capital, civil society, and democracy. International Political Science Review, 22(2), 201–214. https://doi.org/10.1177/0192512101222004

60.

Nießen

Danner

Spengler

Lechner

C. M.

(2020). Big five personality traits predict successful transitions from school to vocational education and training: A large-scale study. Frontiers in Psychology, 11, 1827. https://doi.org/10.3389/fpsyg.2020.01827

61.

Ostroff

(1993). Comparing correlations based on individual-level and aggregated data. Journal of Applied Psychology, 78(4), 569–582. https://doi.org/10.1037/0021-9010.78.4.569

62.

Özcan

Bjørnskov

(2011). Social trust and human development. The Journal of Socio-Economics, 40(6), 753–762. https://doi.org/10.1016/j.socec.2011.08.007

63.

Paxton

Glanville

J. L.

(2015). Is trust rigid or malleable? A laboratory experiment. Social Psychology Quarterly, 78(2), 194–204. https://doi.org/10.1177/0190272515582177

64.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

J. Y.

Podsakoff

N. P.

(2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903. https://doi.org/10.1037/0021-9010.88.5.879

65.

Poropat

A. E.

(2009). A meta-analysis of the five-factor model of personality and academic performance. Psychological Bulletin, 135(2), 322–338. https://doi.org/10.1037/a0014996

66.

Poulin

M. J.

Haase

C. M.

(2015). Growing to trust: Evidence that trust increases and sustains well-being across the life span. Social Psychological and Personality Science, 6(6), 614–621. https://doi.org/10.1177/1948550615574301

67.

Realo

Allik

Greenfield

(2008). Radius of trust: Social capital in relation to familism and institutional collectivism. Journal of Cross-Cultural Psychology, 39(4), 447–462. https://doi.org/10.1177/0022022108318096

68.

Realo

Dobewall

(2011). Does life satisfaction change with age? A comparison of Estonia, Finland, Latvia, and Sweden. Journal of Research in Personality, 45(3), 297–308. https://doi.org/10.1016/j.jrp.2011.03.004

69.

Reeskens

(2013). But who are those “most people” that can be trusted? Evaluating the radius of trust across 29 European societies. Social Indicators Research, 114(2), 703–722. https://doi.org/10.1007/s11205-012-0169-7

70.

Robbins

B. G.

(2022). Measuring generalized trust: Two new approaches. Sociological Methods & Research, 51(1), 305–356. https://doi.org/10.1177/0049124119852371

71.

Rosenberg

(1956). Misanthropy and political ideology. American Sociological Review, 21(6), 690–695. https://doi.org/10.2307/2088419

72.

Rotter

J. B.

(1971). Generalized expectancies for interpersonal trust. American Psychologist, 26(5), 443–452. https://doi.org/10.1037/h0031464

73.

Rousseau

D. M.

Sitkin

S. B.

Burt

R. S.

Camerer

(1998). Not so different after all: A cross-discipline view of trust. Academy of Management Review, 23(3), 393–404. https://doi.org/10.5465/amr.1998.926617

74.

Schilke

Reimann

Cook

K. S.

(2021). Trust in social relations. Annual Review of Sociology, 47(1), 239–259. https://doi.org/10.1146/annurev-soc-082120-082850

75.

Schmitt

D. P.

Realo

Voracek

Allik

(2008). Why can’t a man be more like a woman? Sex differences in big five personality traits across 55 cultures. Journal of Personality and Social Psychology, 94(1), 168–182. https://doi.org/10.1037/0022-3514.94.1.168

76.

Seifert

(2018). Yet another case of nordic exceptionalism? Extending existing evidence for a causal relationship between institutional and social trust to the Netherlands and Switzerland. Social Indicators Research, 136(2), 539–555. https://doi.org/10.1007/s11205-017-1564-x

77.

Sequeros

C. B.

Hansen

T. F.

Westergaard

Louloudis

Kalamajski

Röder

Rohde

P. D.

Schwinn

Clemmensen

L. H.

Didriksen

Nyegaard

Hjalgrim

Nielsen

K. R.

Bruun

M. T.

Ostrowski

S. R.

Erikstrup

Mikkelsen

Sørensen

Banasik

Consortium

D. G.

(2024). A genome-wide association study of social trust in 33,882 Danish blood donors. Scientific Reports, 14(1), 1402. https://doi.org/10.1038/s41598-024-51636-0

78.

Smith

(1982). In Skinner

(Ed.), The wealth of nations: Books I-III. Penguin Classics (1776).

79.

Smith

(1997). Factors relating to misanthropy in contemporary American society. Social Science Research, 26(2), 70–96. https://doi.org/10.1006/ssre.1997.0592

80.

Sønderskov

K. M.

Dinesen

P. T.

(2014). Danish exceptionalism: Explaining the unique increase in social trust over the past 30 years. European Sociological Review, 30(6), 782–795. https://doi.org/10.1093/esr/jcu073

81.

Soto

C. J.

John

O. P.

(2017). The next Big Five Inventory (BFI-2): Developing and assessing a hierarchical model with 15 facets to enhance bandwidth, fidelity, and predictive power. Journal of Personality and Social Psychology, 113(1), 117–143. https://doi.org/10.1037/pspp0000096

82.

Soto

C. J.

John

O. P.

Gosling

S. D.

Potter

(2011). Age differences in personality traits from 10 to 65: Big five domains and facets in a large cross-sectional sample. Journal of Personality and Social Psychology, 100(2), 330–348. https://doi.org/10.1037/a0021717

83.

Specht

Egloff

Schmukle

S. C.

(2011). Stability and change of personality across the life course: The impact of age and major life events on mean-level and rank-order stability of the big five. Journal of Personality and Social Psychology, 101(4), 862–882. https://doi.org/10.1037/a0024950

84.

Sturgis

Read

Hatemi

P. K.

Zhu

Trull

Wright

M. J.

Martin

N. G.

(2010). A genetic basis for social trust? Political Behavior, 32(2), 205–230. https://doi.org/10.1007/s11109-009-9101-5

85.

Sztompka

(1999). Trust: A sociological theory. Cambridge University Press.

86.

UNDP . (2025). Human development report 2025: A matter of choice: People and possibilities in the age of AI. United Nations Development Programme. https://report.hdr.undp.org

87.

Uslaner

E. M.

(1999). Trust but verify: Social capital and moral behavior. Social Science Information, 38(1), 29–55. https://doi.org/10.1177/053901899038001002

88.

Uslaner

E. M.

(2002). The moral foundations of trust. Cambridge University Press.

89.

Uslaner

E. M.

(2012). Measuring generalized trust: In defense of the ‘standard’ question. In Lyon

Möllering

Saunders

M. N. K.

(Eds.), Handbook of research methods on trust (pp. 72–82). Edward Elgar.

90.

Uslaner

E. M.

(2017). The roots of trust. In Li

(Ed.), Handbook of research methods and applications in social capital (pp. 60–75). Edward Elgar Publishing.

91.

Uslaner

E. M.

(2018). The study of trust. In Uslaner

E. M.

(Ed.), The Oxford handbook of social and political trust (pp. 3–13). Oxford University Press. https://doi.org/10.1093/oxfordhb/9780190274801.013.39

92.

Vaht

Arumäe

Realo

Ausmees

Allik

Henry

Metspalu

Esko

Mõttus

Vainik

(2024). Cohort profile: Personality measurements at the Estonian biobank of the Estonian genome center, University of Tartu. PsyArXiv. Retrieved 20 July 2025, from. https://osf.io/preprints/psyarxiv/2aey6

93.

van Hoorn

(2014). Trust radius versus trust level: Radius of trust as a distinct trust construct. American Sociological Review, 79(6), 1256–1259. https://doi.org/10.1177/0003122414555398

94.

Van Lange

P. A. M.

Vinkhuyzen

A. A. E.

Posthuma

(2014). Genetic influences are virtually absent for trust. PLoS One, 9(4), Article e93880. https://doi.org/10.1371/journal.pone.0093880

95.

Vukasović

Bratko

(2015). Heritability of personality: A meta-analysis of behavior genetic studies. Psychological Bulletin, 141(4), 769–785. https://doi.org/10.1037/bul0000017

96.

(2021). Education and social trust in global perspective. Sociological Perspectives, 64(6), 1166–1186. https://doi.org/10.1177/0731121421990045

97.

Yamagishi

(2011). Trust: The evolutionary game of mind and society. Springer.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.47 MB

Investigating the patterns and measurement of general trust across disciplines,methods,and levels of analysis

Abstract

Plain language summary

Keywords

Two perspectives on general trust: Dispositional vs experiential

Dispositional perspective

Experiential perspective

Measurement of general trust in psychological research

Measurement of general trust in survey research

Aims of the present study

Study 1

Method

Participants

Measures

NEO PI-3 A1: Trust facet scale

Single-item general trust (SIGT) scale

Results

Study 2

Method

Measures and data

NEO PI-R/3 A1: Trust facet scale (self-reports)

NEO PI-R A1: Trust facet scale (informant-reports)

Single-item general trust (SIGT) scale

HDI mean years of schooling

Results

General discussion

Main findings of the study

Individual-level data (Study 1)

Country-level data (Study 2)

Strengths and limitations

Conclusions

Supplemental Material

Supplemental Material - Investigating the patterns and measurement of general trust across disciplines, methods, and levels of analysis

Footnotes

Acknowledgments

Declaration of conflicting interests

Funding

Open science statement

Ethical approval

ORCID iDs

Supplemental Material

Notes

References

Supplementary Material