The empirical distinction between honesty-humility and big five-agreeableness: Initial evidence from German samples

Abstract

Since the advent of the HEXACO model, personality psychologists examined the distinctiveness of its honesty-humility factor from the already known agreeableness factor from the Big Five model. It has been argued that honesty-humility was redundant with Big Five-based agreeableness. However, an extensive evaluation of the separability of agreeableness and honesty-humility at a large-scale level is still pending. Using four German samples, consisting of students and individuals from the general population alike (n_total = 3976), honesty-humility turned out to be best understood as correlated with yet distinguishable from Big Five- and HEXACO-agreeableness in terms of factor structure and relations with focal criteria. Furthermore, honesty-humility and agreeableness scores augmented each other in accounting for variance in several criteria. These findings favor the theoretical and empirical separation between honesty-humility and agreeableness, irrespective of the underlying conceptualization of agreeableness. Cross-cultural replications and behaviorally oriented paradigms should confirm the present findings.

Keywords

big five five factor model construct validity jangle fallacy personality model

Introduction

In personality research, there is vigorous debate about whether five or six factors are needed for a comprehensive overview of broad human personality. The Big Five model posits that the constellation of Openness to Experience, Conscientiousness, Extraversion, Agreeableness, and Neuroticism was sufficient to summarize a broad array of individual differences, which, in turn, explain a host of inter- and intrapersonal outcomes. The HEXACO model is a rotated variant of the Big Five model. It also comprises dimensions entitled openness to experience, conscientiousness, and extraversion, which are mostly consistent with their Big Five pendants. Beyond that, the HEXACO model differs from the Big Five model with respect to the understanding of agreeableness (see below) and neuroticism, with the latter being labeled emotionality in the HEXACO model. Furthermore, the HEXACO model mentions honesty-humility as a unique factor (see Ashton & Lee, 2019, for an overview). Interestingly, the lexical approach according to which more important person descriptors are more prominent in spoken language guided the derivation of both models (Berezina & Gill, 2019). Early lexical studies on the Big Five model were conducted in English language and included different kinds of person descriptors (e.g., labels characterizing stable traits, social evaluations, physical appearance, and abilities), but not all these studies supported the Big Five structure (Saucier, 2009). In contrast, the HEXACO model rests on a larger item pool from different languages that focused on dispositional (rather than evaluative, body-, or ability-related) descriptors. The extension of the item pool also fostered the inclusion of indicators tapping features of honesty-humility that theorists devoted to the Big Five tradition have omitted — explaining why honesty-humility did not emerge in earlier factor analyses (De Raad & Mlačić, 2020; Lee & Ashton, 2020; Saucier, 2009). Opponents of the HEXACO model, however, criticize it for overrepresenting content related to social tendencies and the corresponding set of items for being pre-structured, both of which favor the extraction of specific factors (De Raad & Mlačić, 2020; DeYoung, 2020; McCrae, 2020).

In this vein, the introduction of the HEXACO model initiated a controversial debate about whether honesty-humility is really a unique dimension apart from agreeableness (see target paper by Ashton & Lee, 2020, and the accompanying commentaries in the corresponding special issue that, however, also address other issues, such as the validity of the lexical approach or whether the number of basic personality traits actually matters: https://doi.org/10.1002/per.2284). Most prominently, some critical commentors in the special issue posited honesty-humility as a facet of Big Five-agreeableness or argued that honesty-humility could be integrated into Big Five-agreeableness because contents (e.g., politeness, modesty, and straightforwardness) or themes (e.g., engagement in prosocial behavior) of honesty-humility were already covered by Big Five-agreeableness (DeYoung, 2020; Lynam et al., 2020; McCrae, 2020; Widiger et al., 2020). The debate about the redundancy of honesty-humility with Big Five-agreeableness has not yet been settled, and systematic evaluations in this regard are still pending. Important to note, Big Five-agreeableness is not a uniform, homogeneously defined construct, necessitating the investigation and contrast of honesty-humility vis-à-vis different conceptualizations of agreeableness.

The present study aims to test the alleged jangle fallacy (i.e., the erroneous assumption according to which two constructs/scales are different because they are named differently) affecting honesty-humility and different operationalizations of agreeableness, using German samples. Addressing this research gap is significant in the context of current machine learning-based endeavors to obtain alternative structures of personality and work devoted to a comprehensive map of distinguishable personality facets (e.g., Irwing et al., 2024; Speer et al., 2022). To ensure that these approaches yield an appropriate, complete portrayal of personality, clarity is needed with respect to whether agreeableness and honesty-humility are redundant or separable. Besides, considering cross-cultural differences and the role of culture for the expression of personality (e.g., Hofstede et al., 2010), it is essential to conduct further studies in other cultural backgrounds to underpin the generality of the claims made in the present research. For instance, Berezina and Gill (2019) applied a translation/back-translation approach to adjective markers of honesty-humility and Big Five-agreeableness in six Asian languages and found the markers to be sufficiently different from a semantical perspective.

Theoretical background

The jingle fallacy of agreeableness

In fact, the question of whether agreeableness comprises honesty-humility cannot be answered easily because agreeableness is affected by a jingle fallacy. That is, the false assumption that two measures/constructs tap the same concept because they possess the same name (Hodson, 2021). Agreeableness is not only differently conceptualized in the HEXACO and Big Five models; even different Big Five measures differ in the scope of content coverage of their agreeableness subscales (Ashton et al., 2014). Accordingly, the strongest disattenuated correlation between two Big Five-agreeableness subscales detected in a meta-analysis was ρ = .79, whereas same-named subscales of other Big Five dimensions were much more strongly correlated (ρs ≈ .90; Thielmann et al., 2022). Within the HEXACO model, the agreeableness-honesty-humility distinction has been established (Berezina & Gill, 2019; Thielmann et al., 2022; Zettler et al., 2020), but there is a paucity of systematic evidence to distinguish honesty-humility from different conceptualizations of Big Five-agreeableness. Combining the obvious jingle fallacy of agreeableness with the alleged jangle fallacy regarding honesty-humility and agreeableness, honesty-humility may be sufficiently distinct from one conceptualization of agreeableness but overlaps strongly with another one.

Definitional distinction of honesty-humility and agreeableness

Agreeableness is understood quite differently by various researchers and in different operationalizations (e.g., Thielmann et al., 2022). For instance, the operationalization of agreeableness through the NEO-Personality Inventory-Revised (Costa & McCrae, 1992) assesses the facets Altruism, Compliance, Modesty, Straightforwardness, Tendermindedness, and Trust. Straightforwardness and modesty thereby possess parallels with honesty-humility as these facets reflect sincerity (Lee et al., 2005). Similarly, the Big Five Aspects Scale (DeYoung et al., 2007) conceptualizes agreeableness via Compassion and Politeness, with the latter being close to modest features of honesty-humility (DeYoung, 2020). The Big Five Inventory-2 (Soto & John, 2017), in turn, assesses Compassion, Trust, and Respectfulness, which are not explicitly covered by honesty-humility. HEXACO-based agreeableness includes patience with others, forgiveness, gentle or lenient tendencies in dealing with others, and the flexibility to compromise and cooperate (Lee et al., 2005; Thielmann et al., 2022).

The HEXACO model is a variation or an extension of the Big Five model because essentially the same techniques (lexical approach) and statistical methods (exploratory factor analyses) were applied to derive both models, while contents not sufficiently mentioned in the Big Five model were added in the derivation of the HEXACO model. This re-rotation also accounted for better structural replicability across cultures (Ashton & Lee, 2008a, 2008b, 2019, 2020; Ashton et al., 2014). Furthermore, unlike the Big Five model, there is a standard measure of the HEXACO model (including some shortened versions), contributing to better construct and measurement clarity of the latter. That said, studies largely support the validity of the standard measures of the HEXACO model (Moshagen et al., 2019). The HEXACO model specifies that agreeableness and honesty-humility, along with emotionality, account for altruism (Ashton et al., 2014; Thielmann et al., 2022). Thereby, the motives and conditions differ under which each of the three predicts prosocial tendencies and concrete actions. Honesty-humility reflects truthful, sincere, moral, greed-avoiding, and modest properties. It is about unselfish behaviors, independence from material resources, and adherence to norms (Ashton & Lee, 2008a, 2008b; Zettler et al., 2020).

Given the above characterizations of facets of agreeableness, I argue that their common conceptual core lies in directly and openly targeting positive relationships with others. Compared to this, honesty-humility rather emphasizes internalized aspects of social behavior such as a modest self-image, frugal attitudes, and unassuming views in interpersonal situations. The views and motives related to honesty-humility do have social implications, but I argue that they are more indirect than the cognitions, behaviors, and emotions related to different conceptualizations of agreeableness. To acknowledge this, I treated the common core of agreeableness as an externalized, direct form of prosociality and honesty-humility as an internalized, indirect form of prosociality. The general separability of the respective contents of both constructs is supported by Berezina and Gill (2019) who found that adjective markers are distinguishable in different Asian languages.

Lawson and Robins’ (2021) Framework for the examination of alleged jangle fallacies

This study provides an extensive evaluation of the ostensible jangle fallacy (i.e., false assumption that two measures/constructs tap different concepts because they are named differently [Hodson, 2021]) of honesty-humility and agreeableness, as claimed by opponents of the HEXACO model (e.g., DeYoung, 2020; Lynam et al., 2020; McCrae, 2020; Widiger et al., 2020). Additionally, and to directly compare the degree of distinctiveness across models, I extended evidence on the distinction between honesty-humility and agreeableness within the HEXACO model. I adopted the protocol presented by Lawson and Robins (2021) to evaluate jingle and jangle fallacies: Accordingly, constructs are affected by jangle fallacies if (1) they are conceptually similar, (2) they possess similar expected nomological networks, (3) their measures possess similar observed nomological networks, (4) their measures are themselves empirically strongly related, (5) the measures’ items tend to constitute a common rather than separate latent factors, (6) they exhibit little incremental validity beyond each other, (7) they have similar developmental routes, (8) they can be traced back to common environmental, genetic, or neural causes, (9) they are causally related, and/or (10) they are state or trait manifestations of the same entity. I focus on the conceptual and statistical Criteria 1 to 6 because these are most accessible from a psychometrical perspective. Note that conclusions about jingle or jangle fallacies are not necessarily dichotomous. Instead, there are spectra between complete absence and complete presence and evaluations of different criteria can exhibit different conclusions (Lawson & Robins, 2021). Importantly, I did not evaluate the Big Five and HEXACO models, but the raison d`être of honesty-humility on its own.

Differential propensities of agreeableness and honesty-humility towards externalized, direct versus internalized, indirect prosociality and the tendencies of the inherent features to contribute to the formation of social relationships suggest conceptual distinctiveness between the two. It stands to reason that agreeableness and honesty-humility complement rather than “clone” each other (see also Berezina & Gill, 2019). Thus, I consider Lawson and Robins’ (2021) Criterion 1 for the evaluation of jangle fallacies (i.e., conceptual redundancy) to be predominantly untenable. Consequently, I hypothesized respective measures to be more accurately represented as two correlated latent factors rather than a common one (Criteria 4 and 5; hereafter structural distinctiveness).

In the following, I derive hypotheses on differential correlations of agreeableness and honesty-humility with several criterion classes (i.e., assumed nomological networks; Criterion 2). Criterion 3 on distinct observed nomological networks reflects the empirical test of Criterion 2. In what follows, I derive hypotheses about other constructs with which I expected honesty-humility and agreeableness to be correlated at different levels. Simultaneously, I also hypothesized the construct expected to be more strongly related to a certain criterion to possess more incremental validity for this criterion beyond the other construct than the other way around (Criterion 6).

Distinctiveness of nomological networks and incremental validity

Aversive personality traits

The Dark Tetrad reflects the four aversive traits Machiavellianism (i.e., misanthropic, egotistic exploitation), narcissism (i.e., exaggerated self-views and beliefs of superiority), psychopathy (i.e., impulsive, violent harming of others), and sadism (i.e., enjoyment when others suffer). The model was sought to extend the personality space beyond the Big Five by aversive content not yet sufficiently covered by agreeableness. It turned out, however, that honesty-humility explains almost all the shared variance of the Dark Tetrad traits (Kowalski et al., 2021). This can be attributed to the fact that honesty-humility and the Dark Tetrad reflect diametral positions for the endorsement of deception, entitlement, exploitation, and unethical conduct (Lee & Ashton, 2014; Zettler et al., 2020). The latter behaviors are less prominent in extant conceptualizations of agreeableness. Thus, I hypothesized honesty-humility to be more strongly negatively related to the Dark Tetrad traits than any agreeableness conceptualization (Hypotheses 1–4).

Aggression, cognitions, and affects

Ashton et al. (2014) described honesty-humility and HEXACO-based agreeableness as tendencies to cooperate with another person even when “one could successfully exploit that individual” and “that individual appears to be somewhat exploitative”, respectively (p. 144). From this, they concluded that individuals scoring low on honesty-humility defect proactively and for opportunistic reasons, whereas those low in HEXACO-based agreeableness defect in order to take vengeance. Ashton and Lee (2008b) emphasized the inclination to criminality stemming from low honesty-humility and attributed this to low moral commitment. Consistent with this, Book et al. (2012) found that within the HEXACO model, agreeableness was more strongly negatively related to reactive aggression than honesty-humility, which, in turn, was more strongly negatively related to proactive aggression (see also Dinić & Smederevac, 2019). Consistent with extant evidence, I hypothesized honesty-humility to be more strongly negatively related to delinquency and proactive aggression than any concept of agreeableness (Hypotheses 5 and 6, respectively). Across models, agreeableness emphasizes positive other-related views, attitudes, forgiveness, harmony-seeking, and trusting aspects as well as refusal to evoke or escalate conflicts (Ashton et al., 2014; Ashton & Lee, 2008a, 2008b; Dinić & Smederevac, 2019; Thielmann et al., 2022). Thus, I expected agreeableness, irrespective of conceptualization, to be more strongly negatively related to reactive aggression, negative reciprocity, vengeance (all of which reflect revenge-driven interpersonal behaviors), anger (i.e., intense, negative emotional response to perceived or actual provocation, frustration, or threat), hostility (i.e., antagonistic cognitions toward others), mistrust (i.e., low faith that other people have good intentions; beliefs that others pursue aversive goals), and cynicism (i.e., skeptical attitude of people’s motives, such as beliefs of selfishness, opportunism, and dishonesty) as well as more strongly positively to forgiveness (i.e., getting over experienced resentment), trust propensity (i.e., belief that others pursue good intentions), and positive reciprocity (i.e., responding kindly to kindness) than honesty-humility (Hypotheses 7–16). Acknowledging shared prosocial tendencies of honesty-humility and agreeableness in the broadest sense (Ashton et al., 2014), both were expected to be positively related to cognitive (i.e., recognizing others’ affective states) and affective empathy (i.e., sharing others’ affective states [Jolliffe & Farrington, 2006]; Hypotheses 17 and 18, respectively). However, I did not specify hypotheses about whether either agreeableness or honesty-humility are more strongly related to cognitive and affective empathy.

Motives, values, and social preferences

As outlined above, low honesty-humility emphasizes proactive search for advantages, selfishness, immodesty, and entitlement more strongly than any concept of agreeableness. Thus, I expected honesty-humility to be more strongly negatively related to dominance, prestige, and leadership striving than any conception of agreeableness (Hypotheses 19–21) because all these criteria reflect proactive rather than reactive forms of searching for advantages (Suessenbach et al., 2019). Given the unique desire for positive relationships with others in agreeableness (Ashton & Lee, 2008b; Thielmann et al., 2022), I expected it to be more strongly related than honesty-humility to sociableness, hope for affiliation (both positive), detachment, and unsociability (both negative; Hypotheses 22–25).

Current research

In their meta-analysis, Zettler et al. (2020) investigated the nomological networks of the HEXACO traits. The nomological networks were spanned by broad classes of constructs. Their meta-analysis showcased that honesty-humility and agreeableness indeed exhibit unique relations with external criteria, but clustering validation criteria to relatively broad classes arguably exacerbated interpretations in terms of concrete constructs that are more strongly related to either agreeableness or honesty-humility. Second, unlike the present study, their meta-analysis was limited to the HEXACO model. To address the alleged jangle fallacy of the two, the current study thus augments extant research on the uniqueness of honesty-humility vis-à-vis HEXACO- and Big Five-agreeableness, employing pertinent recommendations for the evaluation of jangle fallacies (Lawson & Robins, 2021). That is, I addressed the separation of honesty-humility from agreeableness by testing their nomological networks, their factor structure, and mutual incremental validity. For the nomological networks and for incremental validity, based on prior evidence and theoretical reasoning alike, I tested associations with criteria I deemed focal to the distinction (especially with respect to moral, emotional, and cognitive correlates), but also links with criteria for which I did not derive specific hypotheses. In doing so, the scope can be extended (Lawson & Robins, 2021; Phan et al., 2020) and new areas can be identified for which a distinction between honesty-humility and agreeableness is reasonable. All endeavors combined serve the goal to examine similarities and differences of agreeableness concepts and honesty-humility at the theoretical and validity levels.

Crowe et al. (2018) pursued a similar goal but a different methodology. First, they included a huge number of conceptually and empirically strongly correlated items on agreeableness (all derived from Big Five measures), but a comparatively small set of honesty-humility items. Second, they explored the common structure of agreeableness and honesty-humility items rather than contrasting plausible model alternatives. It stands to reason that employing a comparatively small set of honesty-humility items along with a pre-structured set of items on agreeableness in exploratory analyses artificially favored their conclusion that honesty-humility was a facet of agreeableness (see also DeYoung, 2020; Lee & Ashton, 2020). Third, Crowe et al.’s subsequent analyses of construct validity regarded the factors extracted from their exploratory approach, not scales supposed to measure honesty-humility and agreeableness. Thus, in contrast to the present study, they did not evaluate the structural or validity-related redundancy of agreeableness and honesty-humility itself.

Important to note, no single study can ultimately answer the question about redundancy versus distinctiveness of agreeableness and honesty-humility. First, cultural differences related to the particular samples used must be acknowledged such that the two traits may manifest differently in different cultures. Therefore, it is possible that agreeableness and honesty-humility can be better differentiated in one culture than in another one. For this reason, the present study aims to provide initial evidence about this issue utilizing data from German samples, and it should be understood as a call for cross-cultural replications.

Method

Study design

I used four independent samples, all of which were adopted from published studies. The distinction between honesty-humility and agreeableness, however, has not yet been examined in any of the samples. Blötner et al. (2024) administered the same cross-sectional online survey to two independent samples, with Sample 1 being a student sample (incentivized through partial course credit) and Sample 2 being a general population sample (incentivized monetarily). To reduce the number of items to be responded to by each participant while having a wide array of contents, a planned missingness design was employed. All participants responded to the HEXACO-60 and were then randomly assigned to one of three random conditions with different measures. In each random condition, similar constructs have been assessed (e.g., different Dark Tetrad measures in different conditions). Missing data were imputed at the level of scales or subscales of interest, using the R package mice (version 3.16.0; van Buuren & Groothuis-Oudshoorn, 2011).

I adopted the data from Sample 3 from Blötner et al.’s (2022) validation study of the Short Dark Tetrad (Paulhus et al., 2021). These authors employed a cross-sectional online survey with a planned missingness design similar to that used in Samples 1 and 2 of the current research. All participants responded to the Big Five Inventory-2 (Danner et al., 2019; Soto & John, 2017) and were then assigned to one of two random conditions with similar contents and numbers of items. Like in Samples 1 and 2, I imputed missing information at the level of (sub)scales of interest.

Sample 4 was adopted from Thielmann et al.’s (2020) Prosocial Personality Project. A longitudinal design was used with six measurement occasions, each containing different measures. All honesty-humility and agreeableness subscales were thereby administered at the first occasion. I adopted only those variables from the four datasets that were pertinent to the current investigation.

Samples

In all online surveys, participants provided informed consent. Data collections were approved by the Institutional Review Boards of the University of Hagen (Samples 1 and 2), the University of Münster (Sample 3), and the University of Koblenz-Landau (Sample 4). This research was not preregistered. The supplement provides the data, R scripts, and supplemental materials: https://osf.io/ja9fs/.

Sample 1

From 1256 participants, 1076 passed two attention checks (367, 349, and 360 participants in Random Conditions 1 to 3; 762 women, 301 men, 13 diverse or preferred not to answer; M_age = 30.8, SD_age = 10.8; 996 had a high school diploma or above; 823 were university students).

Sample 2

From 568 participants, 507 passed the attention checks (168, 175, and 164 in the three random conditions; 239 women, 266 men, 2 diverse; M_age = 47.9, SD_age = 15.5; 298 had a high school diploma or above; 45 were students).

Sample 3

The third sample entailed 600 participants. Participants who experienced technical problems during the data collection (n = 4) or who were under 18 years old (n = 2) were excluded, resulting in a final sample of 594 participants (271 and 323 participants; 77% women, M_age = 28.4, SD_age = 9.0).

Sample 4

From 4585 participants, 1799 provided complete data for the items relevant in the present study (867 women [48%], 930 men [52%], two diverse [<1%; M_age = 44.1, SD_age = 12.4, ranging from 18 to 69; 1123 held a high school diploma or above).

Measures

In the following, I list the measures administered in each sample. Table S1 in the Open Science Framework supplement comprises more specific information about the measures, number of items per (sub)scale, estimates of reliability, and response formats (https://osf.io/ja9fs/).

Honesty-humility and agreeableness

Participants in all samples responded to the honesty-humility subscale from the HEXACO-60 (Ashton & Lee, 2009; Moshagen et al., 2014). Participants in Samples 1, 2, and 4 also responded to the agreeableness subscale from the HEXACO-60. In Samples 3 and 4, the agreeableness subscale from the Big Five Inventory-2 (Danner et al., 2019; Soto & John, 2017) was presented. Only in Sample 4, the agreeableness subscales from the Big Five Aspects Scale (DeYoung et al., 2007; Mussel & Paelecke, 2018), from the NEO Five-Factor Inventory (Borkenau & Ostendorf, 2008; Costa & McCrae, 1992), and from the NEO representation of the International Personality Item Pool (Goldberg, 1992; Treiber et al., 2013) have been presented.

Measures in samples 1 and 2

In Samples 1 and 2, participants also responded to the remaining subscales from the HEXACO-60. Beyond that, they responded to the Machiavellian Approach and Avoidance Questionnaire, the Self-Report Psychopathy Scale—Fourth Edition, the Levenson Self-Report Psychopathy Scale, the Triarchic Psychopathy Measure, the Varieties of Sadistic Tendencies, the Comprehensive Assessment of Sadistic Tendencies, the Assessment of Sadistic Personality, the Short Sadistic Impulse Scale, the Reactive and Proactive Aggression Questionnaire, the Buss-Perry Aggression Questionnaire, the Global Assessment of Internet Trolling, the Checklist for Antisocial Behaviors and Criminality, Self-Reported Delinquency in Adolescents Scale, a brief form of the Urgency Premeditation Perseverance Sensation Seeking Impulse Behavior Scale, a brief form of the Barratt Impulsiveness Scale, the Dominance subscale from the Dominance Prestige and Leadership Scale, the People Pleasing Scale, the Fear Enjoyment Questionnaire, the Social Cynicism Scale, the Basic Empathy Scale, the Affective and Cognitive Measure of Empathy, and a brief form of the Reading the Mind in the Eyes Test.

Measures in sample 3

Participants in Sample 3 were provided with the remaining subscales from the Big Five Inventory-2, a brief form of the Personality Inventory for DSM-5, the Short Dark Tetrad, a short form of the Narcissistic Admiration and Rivalry Questionnaire, a short form of the Narcissistic Personality Inventory, the Mach-IV scale, the M7 and P7 scales, a short form of the Self-Report Psychopathy Scale—Third Edition, the Varieties of Sadistic Tendencies, a short form of the Buss-Perry Aggression Questionnaire, a brief form of the Urgency Premeditation Perseverance Sensation Seeking Impulse Behavior Scale, the Unified Motives Scales, the Dominance Prestige and Leadership Scales, the Personal Values Scale, the Rosenberg Self-Esteem Scale, the Revised Sociosexual Orientation Inventory, and a short form of the Interpersonal Adjective List.

Measures in sample 4

Last, in Sample 4, participants responded to the Short Dark Triad, the Short Sadistic Impulse Scale, the Buss-Perry Aggression Questionnaire, the Propensity to Trust Scale, the Personal Norm of Reciprocity Questionnaire, the Barratt Impulsiveness Scale, the Social Dominance Orientation Scale, the Trait Forgiveness Scale, the Brief Vengeance Scale, the Cynicism Scale, and the Interpersonal Reactivity Index.

Analytic strategy

Structural distinctiveness

To test for the common factor structure of honesty-humility and agreeableness, I employed confirmatory factor analyses with the R package lavaan (version 0.6–16; Rosseel, 2012). The honesty-humility subscale as well as three of the four employed agreeableness subscales propose hierarchical structures with lower-level facets. Thus, I first assigned the items to the facets they intend to measure. Then, I specified (a) a model in which all facets were collapsed to the same higher-order factor and (b) a two-factor model in which facets were assigned to their corresponding agreeableness and honesty-humility factors, which, in turn, were allowed to be correlated. To identify potential redundancies at a lower level, I also conducted confirmatory factor analyses at the level of facets. That is, I examined the correlations among the facets of honesty-humility and agreeableness as proposed by the respective scale authors without specifying higher-order agreeableness and honesty-humility factors. Since all measures contain inverted items, fit indices can erroneously suggest poor model fit. This, however, can be attributed to a method artifact due to item polarity. To account for this, I added method factors for inverted items to all confirmatory factor analyses (Weijters & Baumgartner, 2022). Besides, no residual correlations or cross-loadings were specified.

Considering the Likert-type response format of the items, I used the mean- and variance-adjusted weighted least squares estimator. I regarded Comparative Fit Indices > .90, Root Mean Square Error of Approximation < .06, and Square Root Mean Residuals < .08 as indicators of sufficient model fit (Hu & Bentler, 1999). Considering criticism against the strict reliance on fit measures (e.g., Heene et al., 2011), however, and the rationale of the present study (i.e., structural separability of honesty-humility and agreeableness), my primary interest rested on the correlations between the latent honesty-humility and agreeableness factors.

Distinctiveness of nomological networks

I computed bivariate correlations between honesty-humility and agreeableness scores on the one hand and the stated criteria on the other. For the sake of interpretability, I computed the correlation analyses based on overall mean scores of the respective agreeableness and honesty-humility items instead of the corresponding lower-order facet scores. Using the R package diffcor (version 0.8.3; Blötner, 2024), I tested whether the honesty-humility and agreeableness scores yielded different correlations with the criteria employed, α = .001. Last, the agreement of the observed correlation profiles of honesty-humility and agreeableness were quantified by means of the Double-Entry Intraclass Correlation (ICC_DE), using the R package iccde (version 0.3.6). The ICC_DE can be understood as a vector correlation and is robust against differential profile elevation, scatter, and shape (Blötner & Grosz, 2024). Extreme agreements leave little opportunity for uniqueness of the compared correlation profiles.

Incremental validity

To test incremental validity of honesty-humility beyond a given agreeableness measure (and vice versa) for any criterion, I used hierarchical multiple regression. First, I specified a regression model in which either agreeableness or honesty-humility was the sole predictor of a given criterion and extracted the variance accounted for by this predictor (i.e., R²; Block I). Then, I added the respectively other score as a second predictor in Block II (i.e., honesty-humility was added to the equation if agreeableness was entered first in Block I; agreeableness was added to the equation if honesty-humility was entered first in Block I) and evaluated the change of explained variance due to the added predictor (i.e., ΔR²; Block II). To acknowledge multiple testing, I applied a Type I error level of α = .001.

Results

Structural distinctiveness

Irrespective of whether agreeableness was conceptualized in accordance with the HEXACO (Samples 1, 2, and 4) or the Big Five model (Samples 3 and 4), the model alternatives in which agreeableness and honesty-humility were specified as correlated factors yielded better fit properties than corresponding models in which a common latent factor was proposed (see Table 1). That is, when compared to the corresponding model in which honesty-humility and agreeableness items were forced to load onto the same latent factor, better fit was observed for models in which items were assigned to the factors they are supposed to measure. Heywood cases occurred in the analyses in Sample 2, and in Sample 3, the algorithm of the single-factor model did not converge. In Sample 4, the algorithms of the single- and two-factor models did not converge when testing honesty-humility against the agreeableness subscale from the International Personality Item Pool Representation of the Revised NEO-Personality Inventory (Goldberg, 1992). Note that Heywood cases and non-convergence do not necessarily invalidate measurement models as they can also be attributed to small sample size or non-normality of the data (Fan & Lance, 2017; Nasser & Wisenbaker, 2003). Nevertheless, the respective analyses should not be overinterpreted. In the two-factor models, honesty-humility and agreeableness factors (as constituted by items of different parent measures) were moderately correlated in Samples 1, 2, and 3, .33 (Sample 1; HEXACO-agreeableness) ≤ ρ ≤ .51 (Sample 3; Big Five-agreeableness). When I tested the common factor structure of honesty-humility and the agreeableness subscales from the NEO Five-Factor Inventory (Costa & McCrae, 1992) and the Big Five Aspects Scale (DeYoung et al., 2007) in Sample 4, however, the two specified latent factors were comparatively more strongly correlated than in Samples 1 to 3, ρs = .62 and .78, respectively. To further explore the structures, I intended to specify pertinent bifactor models for the combination of honesty-humility with any agreeableness scale, but the reported single- and two-factor models were already very complex (i.e., method factors for inverted items and hierarchical structure with lower-order facets). Consequently, the analyses of corresponding bifactor models frequently revealed non-convergence.

Table 1.

Single Factor and Two Factor Confirmatory Factor Analyses for all Samples.

Models	χ²(df)	CFI	RMSEA	SRMR	ρ
Sample 1
Single factor	1589.68 (152)	.82	.08	.08	—
Two correlated factors	576.25 (151)	.92	.06	.05	.33
Sample 2
Single factor*	838.85 (152)	.83	.08	.08	—
Two correlated factors*	580.72 (151)	.86	.07	.06	.45
Sample 3
Single factor	(no convergence)
Two correlated factors	344.51 (189)	.96	.05	.06	.51
Sample 4 – HEXACO
Single factor	2862.04 (152)	.81	.08	.08	—
Two correlated factors	1267.22 (151)	.89	.06	.05	.41
Sample 4 – NEO-FFI
Single factor	2525.77 (190)	.82	.08	.07	—
Two correlated factors	2525.77 (190)	.82	.08	.07	.62
Sample 4 – BFI-2
Single factor	3334.26 (190)	.82	.09	.07	—
Two correlated factors	1601.22 (189)	.88	.08	.05	.50
Sample 4 – BFAS
Single factor	4002.02 (382)	.84	.08	.06	—
Two correlated factors	3752.13 (381)	.84	.08	.06	.78
Sample 4 – IPIP-NEO
Single factor	(no convergence)
Two correlated factors	(no convergence)				.45

Note. df = Degrees of freedom. CFI = Comparative fit index. RMSEA = Root mean square error of approximation. SRMR = Square root mean residual. ρ = Latent factor correlation. Labels HEXACO, NEO-FFI, BFI-2, BFAS, and IPIP-NEO indicate the utilized agreeableness subscale. * Heywood cases occurred.

Table S2 in the supplement provides the same analyses without method factors. In line with expectations, ostensibly worse fit was indicated, but fewer Heywood cases occurred (Fan & Lance, 2017; Weijters & Baumgartner, 2022). Besides, the same conclusions arose about the structures and factor correlations. Table S3 furthermore summarizes the confirmatory factor analyses at the level of facets. The highest correlation between facets intended to measure different constructs was observed between the Modesty facet from the honesty-humility subscale of the HEXACO-60-honesty-humility and the Politeness facet of the Big Five Aspects Scale, ρs = .64 and .68 (with and without method factor, respectively).

To delve even more deeply into the structure and to test whether clear-cut factors separating agreeableness from honesty-humility items emerge when examining honesty-humility items along with those of any agreeableness scale, I further conducted exploratory factor analyses with the R package psych (version 2.4.6.26; Revelle, 2024), using Oblimin rotation. The number of extracted factors was based on the number of facets proposed in the parent measures of the analyzed scales. For instance, the subscales from the HEXACO-60 have four facets each, and the subscales from the Big Five Inventory-2 have three subscales each. Thus, I extracted eight factors for the items in Samples 1 and 2 (i.e., HEXACO-based agreeableness and honesty-humility) and seven factors for the items in Sample 3. As can be seen in the RMarkdown in the Open Science Framework supplement, the extracted factors in all samples clearly emphasized either agreeableness or honesty-humility contents. Only few cross-loadings occurred, most of which, however, were very small. The only noticeable exception regarded the items of the Politeness facet of the Big Five Aspects Scale that exhibited substantial loadings onto the same factor as honesty-humility-related items.

Distinctiveness of nomological networks and incremental validity

Aversive personality traits

Tables 2 and 3 provide the correlation coefficients observed in the four samples. Except for the Short Sadistic Impulse Scale (p_diff = .07), honesty-humility was more strongly negatively related to all Dark Tetrad measures than HEXACO-agreeableness in Sample 1, all p_diffs < .001, unless stated otherwise. Honesty-humility was more strongly negatively related to three of seven Dark Tetrad measures than HEXACO-agreeableness in Sample 2 (all p_diffs ≥ .007 in the remaining cases). In Sample 3, honesty-humility was more strongly negatively correlated with the Machiavellianism subscale from the Short Dark Tetrad than agreeableness as measured by the Big Five Inventory-2, all other p_diffs ≥ .001. Honesty-humility was more strongly negatively related to Machiavellianism than any agreeableness scale in Sample 4. Results for the other Dark Tetrad traits were mixed: Either no significant correlation differences emerged, or an agreeableness scale was more strongly negatively related to a particular Dark Tetrad score than honesty-humility. In Samples 1 to 3, honesty-humility predominantly explained more variance of Dark Tetrad traits beyond agreeableness scales than the other way around (i.e., ΔR²; see Table 2). In Sample 4, the opposite pattern occurred in most cases (Table 3). The findings yielded mixed support for the Hypotheses 1 to 4.

Table 2.

Correlations and Incremental Validity of Honesty-Humility and Agreeableness with Criteria for Samples 1 to 3.

	Sample 1^H (n = 1076)				Sample 2^H (n = 507)				Sample 3^B (n = 594)
Correlates	HH	A	ΔR²_A+HH	ΔR²_HH+A	HH	A	ΔR²_A+HH	ΔR²_HH+A	HH	A	ΔR²_A+HH	ΔR²_HH+A
Broad traits
Emotionality/Neuroticism	.09	−.04	.011	.004	−.02	−.10	.000	.009	−.08	−.36	.000	.121
Extraversion	−.01	.09	.001	.009	.12	.19	.003	.023	.06	.24	.000	.056
Conscientiousness	.19	.03	.036	.000	.30	.09	.079	.001	.12	.24	.003	.047
Openness	.09	.06	.006	.001	.05	.07	.001	.003	.18	.24	.013	.040
Maladaptive personality traits
Negative affectivity	—	—	—	—	—	—	—	—	−.14	−.21	.007	.033
Detachment	—	—	—	—	—	—	—	—	−.03	−.39	.006	.159
Antagonism	—	—	—	—	—	—	—	—	−.28	−.40	.033	.117
Disinhibition	—	—	—	—	—	—	—	—	.09	−.15	.017	.031
Psychoticism	—	—	—	—	—	—	—	—	.00	−.17	.002	.030
Dark Tetrad
Machiavellianism (MaaQ)	−.45	−.29	.148	.034	−.27	−.26	.035	.028	—	—	—	—
Machiavellianism (SD4)	—	—	—	—	—	—	—	—	−.40	−.21	.127	.000
Machiavellianism (M7)	—	—	—	—	—	—	—	—	−.32	−.30	.065	.047
Machiavellianism (Mach IV)	—	—	—	—	—	—	—	—	−.49	−.49	.136	.136
Narcissism (SD4)	—	—	—	—	—	—	—	—	−.22	−.06	.044	.000
Narcissism (NARQ)	—	—	—	—	—	—	—	—	−.30	−.24	.058	.028
Narcissism (NPI)	—	—	—	—	—	—	—	—	−.30	−.27	.054	.041
Psychopathy (SD4)	—	—	—	—	—	—	—	—	−.20	−.34	.012	.087
Psychopathy (P7)	—	—	—	—	—	—	—	—	−.17	−.28	.011	.060
Psychopathy (SRP-3-SF)	—	—	—	—	—	—	—	—	−.30	−.35	.046	.077
Psychopathy (SRP-4)	−.50	−.23	.209	.011	−.51	−.43	.139	.066	—	—	—	—
Psychopathy (LSRP)	−.65	−.39	.317	.056	−.69	−.40	.335	.022	—	—	—	—
Psychopathy (TriPM)	−.49	−.34	.172	.048	^†	^†	^†	^†	—	—	—	—
Sadism (SD4)	—	—	—	—	—	—	—	—	−.28	−.42	.032	.126
Sadism (VAST)	−.40	−.26	.116	.027	−.35	−.23	.082	.010	−.29	−.36	.039	.086
Sadism (CAST)	−.44	−.28	.143	.031	−.38	−.20	.109	.004	—	—	—	—
Sadism (ASP)	−.46	−.34	.152	.053	−.56	−.36	.212	.024	—	—	—	—
Sadism (SSIS)	−.34	−.27	.076	.037	−.24	−.26	.024	.031	—	—	—	—
Aggression and delinquency
Reactive aggression	−.31	−.39	.046	.104	−.40	−.41	.068	.078	—	—	—	—
Proactive aggression	−.36	−.23	.096	.021	−.42	−.24	.124	.007	—	—	—	—
Physical aggression	−.33	−.35	.063	.076	−.29	−.28	.040	.032	−.21	−.36	.015	.096
Verbal aggression	−.17	−.43	.004	.159	.00	−.11	.002	.015	−.13	−.41	.000	.150
Anger	−.28	−.58	.017	.282	−.24	−.49	.003	.185	−.11	−.36	.000	.114
Mistrust	−.14	−.23	.007	.039	−.23	−.23	.024	.024	−.13	−.28	.004	.064
Internet trolling	−.34	−.34	.071	.066	−.32	−.18	.075	.003	—	—	—	—
Antisocial and criminal tendencies	−.38	−.25	.109	.024	−.28	−.21	.047	.013	—	—	—	—
Juvenile delinquency	−.26	−.26	.038	.040	−.25	−.22	.033	.017	—	—	—	—
Impulsivity
Negative urgency	−.34	−.34	.067	.068	−.30	−.23	.050	.016	−.19	−.32	.011	.078
Lack of premeditation	−.20	−.16	.028	.012	−.31	−.08	.088	.002	−.07	−.22	.000	.043
Lack of perseverance	−.21	−.11	.035	.003	−.22	−.04	.047	.003	−.04	−.23	.001	.052
Sensation Seeking	−.08	.11	.011	.017	−.20	.00	.048	.007	.06	.25	.000	.059
Positive Urgency	—	—	—	—	—	—	—	—	−.12	−.13	.008	.011
Non-planning impulsivity	−.06	−.05	.003	.001	.03	.00	.001	.000	—	—	—	—
Motor impulsivity	−.23	−.10	.044	.002	−.23	−.22	.024	.021	—	—	—	—
Attentional impulsivity	−.21	−.11	.037	.003	−.24	−.20	.030	.013	—	—	—	—
Motives and values
Hope for achievement	—	—	—	—	—	—	—	—	−.04	.12	.005	.018
Hope for power	—	—	—	—	—	—	—	—	−.23	−.14	.040	.007
Hope for affiliation	—	—	—	—	—	—	—	—	.16	.39	.003	.133
Fear of loss	—	—	—	—	—	—	—	—	−.02	−.12	.000	.014
Dominance seeking	−.54	−.38	.209	.060	−.58	−.37	.222	.027	−.44	−.47	.108	.134
Prestige seeking	—	—	—	—	—	—	—	—	−.33	−.05	.106	.002
Leadership seeking	—	—	—	—	—	—	—	—	−.17	−.08	.022	.001
Self-transcendence	—	—	—	—	—	—	—	—	.23	.37	.018	.106
Self-enhancement	—	—	—	—	—	—	—	—	−.27	−.17	.054	.011
Openness to change	—	—	—	—	—	—	—	—	.19	.27	.015	.050
Conservation	—	—	—	—	—	—	—	—	.19	−.07	.049	.017
Miscellaneous
Self-esteem	—	—	—	—	—	—	—	—	.04	.17	.000	.027
People pleasing	.09	.32	.000	.096	.00	.15	.004	.025	—	—	—	—
Fear enjoyment	−.31	.03	.109	.014	−.23	−.20	.026	.016	—	—	—	—
Sociosexual orientation	—	—	—	—	—	—	—	—	−.23	−.03	.054	.001
Assertiveness	—	—	—	—	—	—	—	—	.11	.11	.008	.006
Cynicism (IPC)	—	—	—	—	—	—	—	—	.00	−.43	.013	.195
Cynicism (SCS)	−.30	−.18	.068	.012	−.24	−.21	.030	.015	—	—	—	—
Hostility	—	—	—	—	—	—	—	—	−.12	−.49	.000	.222
Unsociability	—	—	—	—	—	—	—	—	−.03	−.30	.003	.093
Shyness	—	—	—	—	—	—	—	—	−.07	−.11	.002	.009
Obedience	—	—	—	—	—	—	—	—	−.10	.13	.020	.025
Empathy (IPC)	—	—	—	—	—	—	—	—	.15	.59	.000	.320
Cognitive empathy (BES)	.05	.16	.000	.024	.29	.09	.076	.001	—	—	—	—
Cognitive empathy (ACME)	.16	.11	.018	.005	.24	.08	.051	.000	—	—	—	—
Affective empathy (BES)	.22	.18	.034	.015	.24	.18	.035	.009	—	—	—	—
Affective empathy (ACME)	.44	.29	.140	.035	.47	.24	.166	.005	—	—	—	—
RMET-10	.09	.08	.005	.003	^†	^†	^†	^†	—	—	—	—
Sociableness	—	—	—	—	—	—	—	—	.14	.38	.001	.127
ICC_DE	.76		—	—	.79		—	—	.63		—	—

Note. ΔR²_A+HH (ΔR²_HH+A) indicate incremental validity of honesty-humility beyond agreeableness (agreeableness beyond honesty-humility). HH = Honesty-humility. A = Agreeableness. Superscripted H and B indicate that broad traits were assessed in accordance with the HEXACO or Big Five model. MaaQ = Machiavellian Approach and Avoidance Questionnaire. SD4 = Short Dark Tetrad. NARQ = Narcissistic Admiration and Rivalry Questionnaire. NPI = Narcissistic Personality Inventory. SRP-3-SF and SRP-4 = Self-Report Psychopathy Scale-Third Edition Short Form, and Fourth Edition, respectively. LSRP = Levenson Self-Report Psychopathy Scale. Self-Report Psychopathy Scale-Fourth Edition. TriPM = Triarchic Psychopathy Measure. VAST = Varieties of Sadistic Tendencies. CAST = Comprehensive Assessment of Sadistic Tendencies. ASP = Assessment of Sadistic Personality. SSIS = Short Sadistic Impulse Scale. IPC = Interpersonal circumplex. SCS = Social Cynicism Scale. BES = Basic Empathy Scale. ACME = Affective and Cognitive Measure of Empathy. RMET-10 = 10-item version of the Reading the Mind in the Eyes Test. ICC_DE = Double-Entry intraclass correlation. Bolded parameters indicate significant correlation differences or incremental validity (p < .001). Italicized bivariate correlation coefficients were significantly different from zero in the samples, p < .001 (two-sided). ^† Scores could not be imputed in Sample 2 and were, therefore, omitted from analyses.

Table 3.

Correlations and Incremental Validity of Honesty-Humility and Agreeableness with Criteria for Sample 4.

Correlates	HH	HEXACO	ΔR²	NEO	ΔR²	BFAS	ΔR²	BFI-2	ΔR²	IPIP	ΔR²
Agreeableness
HEXACO	.30	—	−/−	. 60	.001/.275	.45	.010/.126	.64	.005/.327	.37	.036/.084
NEO	.45	.60	.076/.243	—	−/−	.73	.013/.354	.79	.028/.459	.66	.058/.301
BFAS	.47	.45	.122/.107	.73	.026/.344	—	−/−	.77	.042/.408	.84	.044/.525
BFI	.37	.64	.032/.311	.79	.000/.496	.77	.000/.455	—	−/−	.75	.016/.451
IPIP	.33	.37	.050/.083	.66	.001/.336	.84	.006/.603	.75	.003/.465	—	−/−
Machiavellianism	−.34	−.16	.095/.003	−.24	.069/.009	−. 25	.064/.011	−.11	.103/.000	−.04	.120/.005
Narcissism	−.51	−.28	.197/.018	−.51	.098/.102	−.45	.114/.056	−.41	.149/.056	−.36	.169/.044
Psychopathy	−.46	−.42	.126/.083	−.60	.048/.193	−.57	.049/.161	−.51	.089/.132	−.44	.114/.096
Sadism	−.33	−.30	.061/.046	−.45	.020/.115	−.43	.019/.101	−.40	.037/.092	−.35	.051/.064
Positive reciprocity	.17	.17	.015/.015	.30	.002/.062	. 37	.000/.109	.32	.003/.078	. 35	.003/.100
Negative reciprocity	−.41	−.41	.089/.089	−.51	.041/.134	−.46	.047/.090	−.48	.061/.130	−.42	.082/.090
Forgiveness	.34	.59	.029/.263	.51	.016/.160	.38	.033/.064	.55	.023/.207	.36	.056/.069
Vengeance	−.45	−.43	.113/.097	−.54	.057/.140	−.49	.062/.101	−.49	.085/.125	−.43	.108/.091
Physical aggression	−.34	−.38	.057/.082	−.50	.017/.152	−.46	.020/.116	−.46	.035/.128	−.39	.052/.084
Verbal aggression	−.15	−.41	.001/.143	−.44	.003/.172	−.33	.000/.090	−.37	.000/.112	−.21	.007/.030
Anger	−.30	−.57	.017/.256	−.48	.009/.150	−.34	.024/.051	−.46	.019/.141	−.24	.054/.023
Mistrust	−.30	−.32	.044/.059	−.49	.008/.158	−.30	.031/.033	−.40	.027/.095	−.27	.048/.035
Trust propensity	.18	.28	.009/.059	.37	.000/.104	.27	.003/.046	.45	.000/.171	.34	.005/.086
Affective empathy	.31	.31	.052/.051	.51	.008/.176	.64	.000/.310	.59	.011/.256	.67	.010/.357
Cognitive empathy	.28	.33	.038/.063	.41	.012/.103	.50	.003/.173	.46	.016/.145	.51	.016/.191
Dominance	−.31	−.13	.083/.001	−.28	.043/.026	−.33	.032/.042	−.28	.052/.030	−.29	.053/.039
Cynicism	−.24	−.25	.029/.033	−.35	.008/.077	−.23	.021/.018	−.31	.018/.055	−.23	.029/.027
Impulsivity
Attention	−.23	−.20	.031/.020	−.28	.014/.038	−.24	.018/.021	−.27	.019/.041	−.20	.030/.018
Non-planning	−.03	−.04	.000/.001	−.08	.000/.005	−.12	.001/.014	−.10	.000/.009	−.14	.000/.017
Motor	−.16	−.06	.021/.000	−.08	.017/.000	−.07	.019/.000	−.02	.025/.002	.04	.032/.009
ICC_DE	—	.90	—	.84	—	.81	—	.82	—	.83	—

Note. n = 1,799. HH = Honesty-humility. NEO = NEO Five-Factor Inventory. BFAS = Big Five Aspects Scale. BFI-2 = Big Five Inventory-2. IPIP = International Personality Item Pool Representation of the Revised NEO-Personality Inventory. ΔR² = Incremental validity of honesty-humility over agreeableness/agreeableness over honesty-humility. ICC_DE = Double-entry intraclass correlation. Bolded parameters indicate significant correlation differences or incremental validity (p < .001). Italicized bivariate correlation coefficients were significantly different from zero in the samples, p < .001 (two-sided).

Aggression, cognitions, and Affects

Partly in line with Hypothesis 5, honesty-humility was more strongly negatively related to antisocial and criminal tendencies than HEXACO-agreeableness, but no significant correlation differences occurred for juvenile delinquency, p_diffs ≥ .45 (Samples 1 and 2). Consistent with Hypothesis 6, however, honesty-humility was more strongly negatively related to proactive aggression; findings on incremental validity of honesty-humility over HEXACO-agreeableness — as compared to the reversed pattern — support this. In Samples 1 and 2, HEXACO-agreeableness provided substantial incremental validity beyond honesty-humility for reactive aggression, but only in Sample 1, incremental validity of agreeableness was considerably higher than incremental validity of honesty-humility (Hypothesis 7).

Delving more deeply into phenomena tapping reciprocity, a respective correlation difference for negative reciprocity could only be observed for the agreeableness subscale of the NEO Five-Factor Inventory. Besides, the agreeableness subscales from the Big Five Aspects Scale and the Big Five Inventory-2 exhibited higher incremental validity over honesty-humility than did honesty-humility over the respective agreeableness subscales (Hypothesis 8). Similarly, the agreeableness subscale from the NEO Five-Factor Inventory was more strongly negatively correlated with vengeance than was honesty-humility and respective incremental validities differ accordingly. The correlation of honesty-humility with vengeance did not differ significantly from those of the agreeableness subscales from the Big Five Aspects Scale and the Big Five Inventory-2, but the latter two yielded higher incremental validities beyond honesty-humility than did honesty-humility beyond the two agreeableness scales (Hypothesis 9).

In Samples 1 to 3, agreeableness was predominantly more strongly negatively related to anger and exhibited higher incremental validity beyond honesty-humility than the other way around (Hypothesis 10). In Sample 4, the correlations of the agreeableness subscale of the Big Five Aspects Scale and honesty-humility with anger did not differ significantly, but the agreeableness subscale yielded higher incremental validity beyond honesty-humility than honesty-humility over agreeableness. In Sample 3, and consistent with Hypotheses 11 and 13, Big Five-agreeableness was more strongly negatively related to hostility and cynicism (both measured in accordance with the Interpersonal Circumplex model) than honesty-humility and exhibited comparatively higher incremental validity. If agreeableness was measured in accordance with the HEXACO model and if cynicism was assessed with a standalone measure (Social Cynicism Scale [Sample 1]), the opposite patterns occurred for correlation differences and incremental validity. In Sample 4, the agreeableness subscale from the NEO Five-Factor Inventory was more strongly negatively related to cynicism than honesty-humility and yielded comparatively higher incremental validity over honesty-humility than in the reversed case. The respective correlation difference with the agreeableness subscale from the Big Five Inventory-2 was not significant, but incremental validity of this agreeableness subscale was comparatively higher than the incremental validity of honesty-humility.

In Samples 1 to 3, correlations of honesty-humility and agreeableness as measured with the HEXACO-60 and the Big Five Inventory-2 with mistrust did not differ (all p_diffs ≥ .003). Both conceptualizations of agreeableness had incremental validity over honesty-humility for mistrust (Samples 1 and 3). In Sample 1, honesty-humility and HEXACO-agreeableness augmented each other to almost the same extent in terms of incremental validity (see Table 2). In Sample 4, however, the agreeableness subscales of the NEO Five-Factor Inventory and the Big Five Inventory-2 were more strongly negatively related to mistrust than honesty-humility and yielded substantial incremental validity beyond honesty-humility. Irrespective of the parent measure, in turn, agreeableness was more strongly positively correlated than honesty-humility with trust propensity (Hypotheses 12 and 15).

The agreeableness subscales from the HEXACO-60, the NEO Five-Factor Inventory, and the Big Five-Inventory-2 were more strongly positively related to forgiveness than honesty-humility and incremental variance of these agreeableness subscales over honesty-humility were higher than incremental validity of honesty-humility over the agreeableness scales (Hypothesis 14). Considering positive reciprocity, all Big Five-agreeableness scales were more strongly positively related to positive reciprocity than was honesty-humility (Hypothesis 16).

Motives, values, and social preferences

In Sample 2, the cognitive empathy subscale from the Basic Empathy Scale was more strongly positively related to honesty-humility, whereas higher incremental validity for all empathy subscales was observed for honesty-humility (Hypothesis 17). In Samples 1 and 2, the affective empathy subscale of the Affective and Cognitive Measure of Empathy was more strongly positively correlated with honesty-humility than with HEXACO-agreeableness (Hypothesis 18). However, honesty-humility had comparatively higher incremental validity for both affective empathy scales as well as for the cognitive empathy subscale from the Affective and Cognitive Measure of Empathy. Overall empathy was more strongly positively correlated with the agreeableness subscale from the Big Five Inventory-2 (Sample 3), which is supplemented by higher incremental validity of this agreeableness subscale (see Table 2). In Sample 4, all Big Five-based agreeableness subscales were more strongly positively correlated with both cognitive and affective empathy, which was also supplemented by comparatively higher incremental validity (see Table 3).

Unlike Big Five-based agreeableness subscales (Samples 3 and 4), HEXACO-based agreeableness was less strongly negatively related to dominance seeking than was honesty-humility. Furthermore, honesty-humility exhibited higher incremental validity over HEXACO-agreeableness than the other way around (Samples 1, 2, and 4). Correlations of Big Five-based agreeableness subscales with dominance seeking did not differ significantly from those obtained for honesty-humility (p_diffs ≥ .14; Samples 3 and 4; Hypothesis 19). Prestige seeking was more strongly negatively related to honesty-humility than to agreeableness as measured with the Big Five Inventory-2 (Sample 3; Hypothesis 20). The correlations with leadership motivation did not differ significantly (p_diff = .08), but incremental validity of honesty-humility over this agreeableness subscale was higher than was incremental validity of agreeableness as measured with the Big Five Inventory-2 over honesty-humility (Sample 3; Hypothesis 21). Sociableness, hope for affiliation, detachment, and unsociability (all Sample 3) were more strongly related to agreeableness as measured by the Big Five Inventory-2 than to honesty-humility and the agreeableness subscale also exhibited higher incremental validity (see Tables 2 and 3; Hypotheses 22–25). Table 4 provides an overview of which hypotheses were supported.

Table 4.

Summary of Supported Hypotheses on Construct and Incremental Validity per Sample.

	Sample 1	Sample 2	Sample 3	Sample 4
Criterion (No. Of hypothesis)	HEX	HEX	BFI	HEX	NEO	BFAS	BFI	IPIP
Machiavellianism (1)	+/+	-/-	±/±	+/+	+/+	+/+	+/+	+/+
Narcissism (2)	x/x	x/x	-/±	+/+	-/-	-/-	+/+	+/+
Psychopathy (3)	+/+	+/+	-/-	-/ +	-/-	-/-	-/-	-/-
Sadism (4)	±/±	±/±	-/-	-/-	-/-	-/-	-/-	-/-
Delinquency (5)	±/±	±/ +	x/x	x/x	x/x	x/x	x/x	x/x
Proactive aggression (6)	+/+	+/+	x/x	x/x	x/x	x/x	x/x	x/x
Reactive aggression (7)	-/+	-/-	x/x	x/x	x / x	x/x	x/x	x/x
Negative reciprocity (8)	x/x	x/x	x/x	-/-	+/ +	-/+	-/ +	-/-
Vengeance (9)	x/x	x/x	x/x	-/-	+/+	-/+	- / +	- /-
Anger (10)	+/+	+/+	+/+	+/+	+/+	-/-	+ / +	- /-
Hostility (11)	x/x	x/x	+/+	x/x	x/x	x / x	x / x	x / x
Mistrust (12)	-/+	-/-	-/+	-/-	+/+	-/-	+/+	-/-
Cynicism (13)	-/-	-/-	+/+	-/+	+/+	-/-	-/+	-/-
Forgiveness (14)	x/x	x/x	x/x	+/+	+/+	-/+	+ / +	-/+
Trust propensity (15)	x/x	x/x	x/x	+/+	+/+	+/+	+/+	+/+
Positive reciprocity (16)	x/x	x/x	x/x	-/-	+/+	+/+	+/+	+/+
Cognitive empathy (17)	±/x	±/x	x/x	+/x	+ / x	+/x	+/x	+/x
Affective empathy (18)	+/x	+/x	x/x	+/x	+ / x	+ / x	+/x	+/x
Dominance motive (19)	+/+	+/+	-/-	+/+	-/+	-/-	-/+	-/+
Prestige motive (20)	x/x	x/x	+/+	x/x	x/x	x/x	x/x	x/x
Leadership motive (21)	x/x	x/x	-/+	x/x	x/x	x/x	x/x	x/x
Sociableness (22)	x/x	x/x	+/+	x/x	x/x	x/x	x/x	x/x
Hope for affiliation (23)	x/x	x/x	+/+	x/x	x/x	x/x	x/x	x/x
Detachment (24)	x/x	x/x	+/+	x/x	x/x	x/x	x/x	x/x
Unsociability (25)	x/x	x/x	+/+	x/x	x/x	x/x	x/x	x/x

Note. Evaluations refer to support for hypotheses concerning construct validity / incremental validity. + = Hypothesis supported. - = Hypothesis not supported. ± = Hypothesis partly supported. x = Not tested. HEX, BFI, NEO, BFAS, and IPIP indicate that the utilized agreeableness subscale was adopted from the HEXACO-60, Big Five Inventory-2, NEO Five-Factor Inventory, Big Five Aspects Scale, and the NEO representation of the International Personality Item Pool, respectively.

Profile agreements of nomological networks

The agreements of the nomological networks of honesty-humility and agreeableness scales ranged from ICC_DE = .63 to .90 (see also Figures S1 to S4 in the supplement for graphical illustrations of the trajectories of correlations across criteria). Intriguingly, the nomological network of honesty-humility overlapped most strongly with that of HEXACO-agreeableness (Sample 4), that is, the conception of agreeableness for which the distinction from honesty-humility has already been established (Berezina & Gill, 2019; Thielmann et al., 2022; Zettler et al., 2020). However, across samples, many hypothesized correlation differences between honesty-humility and HEXACO-agreeableness emerged as expected and, in many cases, failure to detect correlation differences affected criteria for which I did not derive hypotheses.

Discussion

This research extends previous studies on the distinctiveness versus redundancy of honesty-humility and agreeableness by providing an extensive evaluation of different aspects pertaining to their separation (Phan et al., 2020; Thielmann & Zettler, 2020) and by examining both Big Five- and HEXACO-based agreeableness. I tested (a) whether established honesty-humility and agreeableness items are better represented by a common factor or by two correlated factors, (b) differential construct validity of honesty-humility and agreeableness regarding other constructs, (c) incremental validity of honesty-humility and different conceptualizations of agreeableness beyond each other concerning these constructs, and (d) the overall similarity of the nomological networks. Across four samples, analyses of factor structure, construct validity, incremental validity, and similarities of nomological networks suggest that honesty-humility is distinct from prominent conceptualizations of agreeableness.

Structural distinctiveness

Although the fit properties of some two-factor models were not entirely consistent with Hu and Bentler’s (1999) guidelines according to which Comparative Fit Indices > .90, Root Mean Square Error of Approximation < .06, and Square Root Mean Residuals < .08 indicate sufficient model fit, coefficients by and large corroborated the separation of honesty-humility and agreeableness. In most cases, the honesty-humility and agreeableness factors were far from being perfectly correlated in the analyses in which higher-order honesty-humility and agreeableness factors were specified (see Tables 1 and S2), and the correlations among their corresponding facets were mostly at moderate levels (see Table S3). If the two constructs were in fact redundant, much higher latent correlations would have emerged, and model fits of single-factor and two-factor models would have been more similar (however, see measurement-related limitations below). Note in this respect the criticism voiced against the reliance on fixed cutoffs (Heene et al., 2011). Arguably, the approaches employed to test structural separability were very simple because only a few model alternatives were probed. Neither model should be regarded the “true” model, but the assumption according to which honesty-humility was a facet of agreeableness becomes at least less plausible. These findings also align with Saucier (2009, p. 1579) according to whom the HEXACO model “is neither reducible to nor entirely reproducible from the Big Five”. Thus, Lawson and Robins’s (2021) Criteria 4 and 5 for the evaluation of jangle fallacies considering extreme correlations of measures and the emergence of a common rather than two factors, respectively, are mostly deemed untenable for honesty-humility and agreeableness.

That said, slight differences exist in the distinguishability of honesty-humility from specific agreeableness scales, especially from the Big Five Aspects Scale (DeYoung et al., 2007; see the latent factor correlation in Table 1). This can at least in part be attributed to the fact that the agreeableness subscale of the Big Five Aspects Scale entails a politeness facet that is frequently said to capture important contents of honesty-humility. Other agreeableness subscales do not entail such a facet. The fact, however, that honesty-humility can be distinguished more clearly from the remaining conceptualizations of agreeableness in factor analyses again underscores the jingle fallacy affecting agreeableness measures (Thielmann et al., 2022; Zettler et al., 2020).

Distinctiveness of nomological networks and incremental validity

Considering refraining from demanding advantages in the broadest sense, correlations of agreeableness and honesty-humility with the Dark Tetrad traits and dominance seeking are worth noting (Hypotheses 1 to 4, 19). Specific agreeableness scales outperformed honesty-humility or were at least correlated with these criteria at similar levels, again underlining distinct conceptual scopes of agreeableness scales. Several aspects related to the Dark Tetrad do not map exclusively onto exploitative cognitions and seeking advantages (i.e., low honesty-humility) but also onto antipathy toward others, denial of help, and callousness (i.e., low agreeableness). Similarly, dominance orientation strongly aligns with honesty-humility and agreeableness as it reflects demanding advantages for oneself (i.e., low honesty-humility) at the expense of others (i.e., low agreeableness; Duckitt, 2020). Thus, it is reasonable to observe comparable relations of honesty-humility and agreeableness with Dark Tetrad traits and dominance striving in some cases. Furthermore, the reasoning about the role of honesty-humility referred to the shared variance of the Dark Tetrad traits rather than to trait-specific or unique variance (Kowalski et al., 2021). According to Bader et al. (2023), aversive traits are connected by striving for advantages at the expense of others, combining low honesty-humility with of low agreeableness, but they also cover non-aversive features (e.g., motivation to be liked, planning, and invulnerability). Thus, agreeableness and honesty-humility might play different roles in the striving for resources and influence (narcissism and Machiavellianism), the desire to harm others (sadism), or adopting an erratic and antisocial lifestyle (psychopathy). Importantly, different Dark Tetrad measures were used in different samples of the present research, with Dark Tetrad measures differing in their conceptual scope. That said, different measures on different aversive traits overlap to different extents — suggesting jingle and jangle fallacies in this research area (Kay & Arrow, 2022).

Despite conceptual discrepancies among agreeableness scales, I found HEXACO- and Big Five-based agreeableness to be more strongly correlated with criteria closely resembling core features of any conceptualization of agreeableness (e.g., refusal of negative reciprocity, vengeance, anger, hostility; forgiveness, positive and negative reciprocity, sociableness, hope for affiliation; detachment, and unsociability [Hypotheses 8 to 11, 14, 16, 22 to 25]). These findings align with earlier studies that, however, focused on the HEXACO model (e.g., Sokolovska et al., 2018). Similarly, Book et al. (2012) found HEXACO-agreeableness to be more strongly negatively related to reactive aggression than honesty-humility, which, in turn, was more strongly negatively related to proactive aggression. Different correlations with proactive aggression, incremental validity in terms of reactive aggression, and relations with criteria reflecting reciprocity — such as positive and negative reciprocity, vengeance, and forgiveness — in the present study partly align with this (Hypotheses 6 and 7).

It is also worth noting differences between the present research and studies that concluded redundancy or a trait-facet relationship between agreeableness and honesty-humility. For instance, Denissen et al. (2022) found highly similar correlation profiles for the agreeableness subscale from the Big Five Inventory-2 and a self-developed honesty-humility scale. At least three points need to be considered: First, across studies, they used only a small, specific set of validation criteria (other agreeableness scales, aversive personality traits, and a HEXACO-based honesty-humility subscale). Thus, the scope of criteria for which relations are deemed equivalent might be too narrow to support their claim (Lawson & Robins, 2021). Second, unlike their self-developed honesty-humility scale, the correlations of the original, HEXACO-based honesty-humility scale with the criteria were sufficiently distinct from those of the agreeableness subscales. Third, their self-developed honesty-humility scale and the agreeableness subscale possessed substantial incremental validity beyond each other in predicting a social value orientation (which was posited as a shared correlate). If honesty-humility and agreeableness were indeed redundant, they should not augment each other, or only to a negligible extent (see also the present findings on incremental validity). Unlike Denissen et al. (2022), I employed a larger set of validation criteria and found smaller correlations between honesty-humility and different agreeableness scores (except for the Big Five Aspects subscale). In keeping with Denissen et al., I found HEXACO-honesty-humility and agreeableness scores to augment each other in predictions of crucial criteria, regardless of the parent measure of the respective agreeableness subscale. Thus, Lawson and Robins’ (2021) criteria dealing with similar expected and observed nomological networks as well as little incremental validity beyond each other (Criteria 2, 3, and 6) are mostly untenable for honesty-humility and agreeableness, irrespective of the conception of agreeableness.

Similarities of the nomological networks of honesty-humility and agreeableness

Intriguingly, the agreement of the nomological networks of honesty-humility and HEXACO-agreeableness in Sample 4 indicated that the two scales exhibited an almost perfect overlap, ICC_DE = .90. This should not be the case since the HEXACO model was designed to separate these two traits. At the same time, a host of correlation differences and substantial shares of incremental validity were observed — especially regarding criteria I posited as crucial to the distinction (see Table 3). These findings support the above reasoning according to which the two are complementary. Important to note, the ICC_DE can indicate perfect agreement even in the case of clearly distinct parameter profiles, but it can also point to almost perfect distinctiveness where there is none. Hilbig et al. (2021) demonstrated this by means of the profiles X1 = [.03, .01, .02], X2 = [.51, .52, .53], Y1 = [.29, .31, .30], and Y2 = [.30, .30, .29]. Despite obvious differences, X1 and X2 yielded an agreement of ICC_DE = −1.00, and despite obvious similarity, Y1 and Y2 yielded an agreement of ICC_DE = −.06. As can be seen, the dispersion of the coefficients within each profile is very small, erroneously suggesting redundancy or distinctiveness. Future research should, thus, consider alternative measures of profile agreement. Alternatively, the ICC_DE should not be interpreted in isolation but along with lower-level discrepancies and commonalities (e.g., correlation differences).

Limitation and future direction

I aimed to provide initial evidence militating against the redundancy of agreeableness and honesty-humility, and samples from the German population were used for this endeavor. Arguably, cultural differences could favor or hinder the distinction between agreeableness and honesty-humility due to cross-cultural differences in the manifestations of personality traits. For instance, Hofstede et al. (2010) posited the motivation towards achievement and success (i.e., competitive vs. nurturing orientation) and individualism (i.e., emphasis on social independence vs. interdependence) as two of six dimensions that distinguish between members from different cultures. Both domains imply meaningful heterogeneity in the emergence and manifestation of prosocial emotions, cognitions, behaviors, and motives, whereas German culture is rated relatively high in both (66 and 79 of 100, respectively). Thus, replications of the present study are required in cultures that differ from the German one in terms of focal dimensions from Hofstede et al.’s framework. Since honesty-humility and agreeableness turned out to be distinguishable at least at a semantical level in six Asian languages (Berezina & Gill, 2019), however, it stands to reason that the present findings from German samples generalize, despite differences in cultural values (e.g., Hofstede et al., 2010).

Second, Samples 1 and 3 entailed predominantly students (i.e., homogeneity in age, education, and other characteristics) and were imbalanced in terms of gender. In this vein, older individuals as well as women score higher on honesty-humility and lower in aversive phenomena (cf. Ashton et al., 2024; Kowalski et al., 2021). That said, variance restrictions must be assumed for very aversive validation criteria (e.g., delinquency), potentially leading to underestimated correlations.

Third, all constructs were measured with self-reports, suggesting a common method bias and inflations of the observed profile agreements. Relatedly, in some cases, shared item wordings must be considered such that predictor measures (i.e., agreeableness and honesty-humility) and the criterion measures partly overlap in assessed contents (e.g., items on interpersonal trust in the agreeableness subscale from the Big Five Inventory-2 and in the utilized measures on [mis]trust; desire to possess luxury goods and to be seen as someone special in narcissism, prestige, and honesty-humility). In Sample 4, the same response scales were used for all measures, regardless of the original response formats of the scales. Both shared item contents and shared response formats thereby exacerbate the common method effects. Future research should consider using different data sources for honesty-humility and agreeableness on the one hand and the assumed criteria on the other (e.g., behavioral measures). Alternatively, self-ratings could be combined with acquaintance-ratings to reduce the common method bias (Podsakoff et al., 2012).

Fourth, across samples, different criteria have been assessed: The measures employed in Sample 1 (and the copied survey administered in Sample 2) focused on different shades of antagonistic traits and tendencies, whereas the measures in Samples 3 and 4 highlighted both antagonistic and prosocial constructs. On the one hand, assessing heterogeneous constructs extends the range of criteria that do (not) distinguish agreeableness and honesty-humility. On the other hand, the validation criteria at hand determine the height of the profile agreement (Lawson & Robins, 2021). Thus, the profile agreements between the honesty-humility and agreeableness subscales from the HEXACO-60 scale obtained in Samples 1 and 2 cannot be easily compared to the respective coefficient obtained in Sample 4, even though the same measure is evaluated in all three samples (but see also the above criticism against the ICC_DE).

Conclusion

This research addressed the question of whether the broad personality trait honesty-humility meaningfully extends the readily known trait agreeableness or whether the former essentially clones the latter. The results indicate the uniqueness of honesty-humility vis-à-vis agreeableness — mostly regardless of how agreeableness is measured. Thus, the assumption of a jangle fallacy affecting these constructs is untenable. That said, this research does not answer the question of whether the HEXACO model or the Big Five model is superior, but this question is per se misleading, given the taxonomic nature of both. Instead, the present research helps researchers select a suitable personality predictor in order to model specific prosocial and antisocial outcomes in the real world (Hilbig & Moshagen, 2020; Rakhshani et al., 2020; Wiernik et al., 2020).

Supplemental Material

Supplemental Material - The empirical distinction between honesty-humility and big five-agreeableness: Initial evidence from German samples

Supplemental Material for The empirical distinction between honesty-humility and big five-agreeableness: Initial evidence from German samples by Christian Blötner in Personality Science

Footnotes

Author note

Atsushi Oshio was the handling editor of this manuscript.

Acknowledgements

I would like to thank Isabel Thielmann for her invaluable comments on an earlier draft of this manuscript and for providing me with the data used in this study as Sample 4.

Author contributions

Christian Blötner: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Software; Supervision; Validation; Visualization; Writing – original draft; Writing – review & editing.

Declaration of conflicting interests

Not applicable.

Funding

This study was partially supported by the Internal Research Grant # 373_2023_15 funded by the FernUniversität in Hagen.

ORCID iD

Christian Blötner

Data accessibility statement

All datasets can be retrieved from

Supplemental material

Supplemental material for this article is available online. Depending on the article type, these usually include a Transparency Checklist, a Transparent Peer Review File, and optional materials from the authors. All R scripts and supplemental materials can be retrieved from .

Notes

Not applicable.

References

Ashton

M. C.

Lee

(2008a). The HEXACO model of personality structure and the importance of the H factor. Social and Personality Psychology Compass, 2(5), 1952–1962. https://doi.org/10.1111/j.1751-9004.2008.00134.x

Ashton

M. C.

Lee

(2008b). The prediction of honesty–humility-related criteria by the HEXACO and five-factor models of personality. Journal of Research in Personality, 42(5), 1216–1228. https://doi.org/10.1016/j.jrp.2008.03.006

Ashton

M. C.

Lee

(2009). The HEXACO-60: A short measure of the major dimensions of personality. Journal of Personality Assessment, 91(4), 340–345. https://doi.org/10.1080/00223890902935878

Ashton

M. C.

Lee

(2019). How well do Big Five measures capture HEXACO scale variance? Journal of Personality Assessment, 101(6), 567–573. https://doi.org/10.1080/00223891.2018.1448986

Ashton

M. C.

Lee

(2020). Objections to the HEXACO model of personality structure—and why those objections fail. European Journal of Personality, 34(4), 492–510. https://doi.org/10.1002/per.2242

Ashton

M. C.

Lee

Anglim

Bucher

M. A.

Horwood

Samuel

D. B.

(2024). The HEXACO personality space before and after re-rotation to approximate the Big Five dimensions. Journal of Personality Assessment, 106(2), 145–155. https://doi.org/10.1080/00223891.2023.2229427

Ashton

M. C.

Lee

de Vries

R. E.

(2014). The HEXACO honesty-humility, agreeableness, and emotionality factors: A review of research and theory. Personality and Social Psychology Review: An Official Journal of the Society for Personality and Social Psychology, Inc, 18(2), 139–152. https://doi.org/10.1177/1088868314523838

Bader

Hilbig

B. E.

Zettler

Moshagen

(2023). Rethinking aversive personality: Decomposing the Dark Triad traits into their common core and unique flavors. Journal of Personality, 91(5), 1084–1109. https://doi.org/10.1111/jopy.12785

Berezina

Gill

C. M. H. D.

(2019). Modeling personality structure using semantic relationships: Is the HEXACO honesty-humility a distinct trait? Psychology in Russia: State of the Art, 12(1), 89–103. https://doi.org/10.11621/pir.2019.0107

10.

Blötner

(2024). diffcor: Fisher’s z-tests concerning difference of correlations (R package version 0.8.3). CRAN. https://doi.org/10.32614/CRAN.package.diffcor

11.

Blötner

Grosz

M. P.

(2024). iccde: Computation of the double-entry Intraclass correlation (R package version 0.3.6). CRAN. https://doi.org/10.32614/CRAN.package.iccde

12.

Blötner

Spormann

S. S.

Hofmann

M. J.

Mokros

(2024). Measures of subclinical psychopathy and everyday sadism are still redundant: A conceptual replication and extension of blötner and mokros (2023). Journal of Personality. Advance online publication. https://doi.org/10.1111/jopy.12996

13.

Blötner

Ziegler

Wehner

Back

M. D.

Grosz

M. P.

(2022). The nomological network of the Short Dark Tetrad scale (SD4). European Journal of Psychological Assessment, 38(3), 187–197. https://doi.org/10.1027/1015-5759/a000655

14.

Book

A. S.

Volk

A. A.

Hosker

(2012). Adolescent bullying and personality: An adaptive approach. Personality and Individual Differences, 52(2), 218–223. https://doi.org/10.1016/j.paid.2011.10.028

15.

Borkenau

Ostendorf

(2008). NEO-Fünf-Faktoren Inventar nach Costa und McCrae (NEO-FFI). In Manual (2., neu normierte und vollständig überarbeitete Auflage). Hogrefe.

16.

Buuren

S. V.

Groothuis-Oudshoorn

(2011). mice: Multivariate imputation by chained equations in R. Journal of Statistical Software, 45(3), 1–67. https://doi.org/10.18637/jss.v045.i03

17.

Costa

P. T.

Jr. McCrae

R. R.

(1992). NEO personality inventory–revised (NEO-PI-R) and NEO five-factor inventory (NEO-FFI) professional manual. Psychological Assessment Resources.

18.

Crowe

M. L.

Lynam

D. R.

Miller

J. D.

(2018). Uncovering the structure of Agreeableness from self-report measures. Journal of Personality, 86(5), 771–787. https://doi.org/10.1111/jopy.12358

19.

Danner

Rammstedt

Bluemke

Lechner

Berres

Knopf

Soto

C. J.

John

O. P.

(2019). Das big five inventar 2 [the big five inventory 2]. Diagnostica, 65(3), 121–132. https://doi.org/10.1026/0012-1924/a000218

20.

Denissen

J. J. A.

Soto

C. J.

Geenen

John

O. P.

van Aken

M. A. G.

(2022). Incorporating prosocial vs. antisocial trait content in big five measurement: Lessons from the big five inventory-2 (BFI-2). Journal of Research in Personality, 96, Article 104147. https://doi.org/10.1016/j.jrp.2021.104147

21.

De Raad

Mlačić

(2020). Quixotry gone astray. European Journal of Personality, 34(4), 519–520.

22.

DeYoung

(2020). The Big Five model remains a good choice for personality theory and research. European Journal of Personality, 34(4), 520–522.

23.

DeYoung

C. G.

Quilty

L. C.

Peterson

J. B.

(2007). Between facets and domains: 10 aspects of the big five. Journal of Personality and Social Psychology, 93(5), 880–896. https://doi.org/10.1037/0022-3514.93.5.880

24.

Dinić

B. M.

Smederevac

(2019). Effects of HEXACO traits and experimental provocation on aggression. Zeitschrift für Psychologie, 227(3), 225–229. https://doi.org/10.1027/2151-2604/a000381

25.

Duckitt

(2020). Duckitt, John. In Zeigler-Hill

Shackelford

(Eds.), Encyclopedia of personality and individual differences (pp. 1219–1222). Springer. https://doi.org/10.1007/978-3-319-24612-3

26.

Fan

Lance

C. E.

(2017). A reformulated correlated trait- correlated method model for multitrait-multimethod data effectively increases convergence and admissibility rates. Educational and Psychological Measurement, 77(6), 1048–1063. https://doi.org/10.1177/0013164416677144

27.

Goldberg

L. R.

(1992). The development of markers for the Big-Five factor structure. Psychological Assessment, 4(1), 26–42. https://doi.org/10.1037/1040-3590.4.1.26

28.

Heene

Hilbert

Draxler

Ziegler

Bühner

(2011). Masking misfit in confirmatory factor analysis by increasing unique variances: A cautionary note on the usefulness of cutoff values of fit indices. Psychological Methods, 16(3), 319–336. https://doi.org/10.1037/a0024917

29.

Hilbig

B. E.

Moshagen

(2020). All models (of basic personality structure) are wrong, but some are useful. European Journal of Personality, 34(4), 527–528.

30.

Hilbig

B. E.

Moshagen

Horsten

L. K.

Zettler

(2021). Agreeableness is dead. Long live agreeableness? Reply to Vize and Lynam. Journal of Research in Personality, 91, Article 104074. https://doi.org/10.1016/j.jrp.2021.104074

31.

Hodson

(2021). Construct jangle or construct mangle? Thinking straight about (nonredundant) psychological constructs. Journal of Theoretical Social Psychology, 5(4), 576–590. https://doi.org/10.1002/jts5.120

32.

Hofstede

G. J.

Minkov

(2010). Cultures and organizations: Software of the mind (3rd ed.). McGraw-Hill.

33.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

34.

Irwing

Hughes

D. J.

Tokarev

Booth

(2024). Towards a taxonomy of personality facets. European Journal of Personality, 38(3), 494–515. https://doi.org/10.1177/08902070231200919

35.

Jolliffe

Farrington

D. P.

(2006). Development and validation of the basic empathy scale. Journal of Adolescence, 29(4), 589–611. https://doi.org/10.1016/j.adolescence.2005.08.010

36.

Kay

C. S.

Arrow

(2022). Taking an elemental approach to the conceptualization and measurement of Machiavellianism, narcissism, and psychopathy. Social and Personality Psychology Compass, 16(4), Article e12662. https://doi.org/10.1111/spc3.12662

37.

Kowalski

C. M.

Rogoza

Saklofske

D. H.

Schermer

J. A.

(2021). Dark triads, tetrads, tents, and cores: Why navigate (research) the jungle of dark personality models without a compass (criterion)? Acta Psychologica, 221, Article 103455. https://doi.org/10.1016/j.actpsy.2021.103455

38.

Lawson

K. M.

Robins

R. W.

(2021). Sibling constructs: What are they, why do they matter, and how should you handle them? Personality and Social Psychology Review: An Official Journal of the Society for Personality and Social Psychology, 25(4), 344–366. https://doi.org/10.1177/10888683211047101

39.

Lee

Ashton

M. C.

(2014). The dark Triad, the big five, and the HEXACO model. Personality and Individual Differences, 67, 2–5. https://doi.org/10.1016/j.paid.2014.01.048

40.

Lee

Ashton

M. C.

(2020). Why six factors, why it matters. European Journal of Personality, 34(4), 562–590.

41.

Lee

Ogunfowora

Ashton

M. C.

(2005). Personality traits beyond the big five: Are they within the HEXACO space? Journal of Personality, 73(5), 1437–1463. https://doi.org/10.1111/j.1467-6494.2005.00354.x

42.

Lynam

D. R.

Crowe

M. L.

Vize

Miller

J. D.

(2020). Little evidence that honesty-humility lives outside of FFM agreeableness. European Journal of Personality, 34(4), 530–531.

43.

McCrae

R. R.

(2020). IMHO, there are only five factors. European Journal of Personality, 34(4), 534–535.

44.

Moshagen

Hilbig

B. E.

Zettler

(2014). Faktorenstruktur, psychometrische Eigenschaften und Messinvarianz der deutschsprachigen Version des 60-Item HEXACO Personlichkeitsinventars [Factor structure, psychometric properties and measurement invariance of the German version of the 60- item HEXACO Personality Inventory]. Diagnostica, 60(2), 86–97. https://doi.org/10.1026/0012-1924/a000112

45.

Moshagen

Thielmann

Hilbig

B. E.

Zettler

(2019). Meta-analytic investigations of the HEXACO Personality Inventory (Revised): Reliability generalization, self–observer agreement, intercorrelations, and relations to demographic variables. Zeitschrift fur Psychologie, 227(3), 186–194. https://doi.org/10.1027/2151-2604/a000377

46.

Mussel

Paelecke

(2018). BFAS-G. Big five aspect scales - German. In Leibniz- Zentrum fur Psychologische Information und Dokumentation (ZPID) (Hrsg.), Elektronisches Testarchiv (PSYNDEX Tests-Nr. 9007737). ZPID. https://doi.org/10.23668/psycharchives.2341

47.

Nasser

Wisenbaker

(2003). A Monte Carlo study investigating the impact of item parceling on measures of fit in confirmatory factor analysis. Educational and Psychological Measurement, 63(5), 729–757. https://doi.org/10.1177/0013164403258228

48.

Paulhus

D. L.

Buckels

E. E.

Trapnell

P. D.

Jones

D. N.

(2021). Screening for dark personalities: The short dark Tetrad (SD4). European Journal of Psychological Assessment, 37(3), 208–222. https://doi.org/10.1027/1015-5759/a000602

49.

Phan

L. V.

Modersitzki

Kuper

Rauthmann

J. F.

(2020). Comparing the nomological networks of the Big Five and HEXACO models. European Journal of Personality, 34, 544–545.

50.

Podsakoff

P. M.

MacKenzie

S. B.

Podsakoff

N. P.

(2012). Sources of method bias in social science research and recommendations on how to control it. Annual Review of Psychology, 63, 539–569. https://doi.org/10.1146/annurev-psych-120710-100452

51.

Rakhshani

Donnellan

Lucas

R. E.

(2020). Thinking pragmatically about structural models. European Journal of Personality, 34(4), 545–546.

52.

Revelle

(2024). psych: Procedures for psychological, psychometric, and personality research (R package version 2.4.6.26). CRAN. https://doi.org/10.32614/CRAN.package.psych

53.

Rosseel

(2012). lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(2), 1–36. https://doi.org/10.18637/jss.v048.i02

54.

Saucier

(2009). Recurrent personality dimensions in inclusive lexical studies: Indications for a big six structure. Journal of Personality, 77(5), 1577–1614. https://doi.org/10.1111/j.1467-6494.2009.00593.x

55.

Sokolovska

Dinic

B. M.

Tomasevic

(2018). Aggressiveness in the HEXACO personality model. Psihologija, 51(4), 449–468. https://doi.org/10.2298/psi170705022s

56.

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

57.

Speer

A. B.

Christiansen

N. D.

Robie

Jacobs

R. R.

(2022). Measurement specificity with modern methods: Using dimensions, facets, and items from personality assessments to predict performance. Journal of Applied Psychology, 107(8), 1428–1439. https://doi.org/10.1037/apl0000618

58.

Suessenbach

Loughnan

Schönbrodt

F. D.

Moore

A. B.

(2019). The dominance, prestige, and leadership account of social power motives. European Journal of Personality, 33(1), 7–33. https://doi.org/10.1002/per.2184

59.

Thielmann

Hilbig

B. E.

Horsten

L. K.

Scholz

(2020). The Prosocial personality Project. Open Science Framework. https://doi.org/10.17605/OSF.IO/M2ABP

60.

Thielmann

Moshagen

Hilbig

B. E.

Zettler

(2022). On the comparability of basic personality models: Meta-analytic correspondence, scope, and orthogonality of the Big Five and HEXACO dimensions. European Journal of Personality, 36(6), 870–900. https://doi.org/10.1177/08902070211026793

61.

Thielmann

Zettler

(2020). On the distinction between Honesty-Humility and Agreeableness in the HEXACO model. European Journal of Personality, 34(4), 552–553.

62.

Treiber

Thunsdorff

Schmitt

Schreiber

W. H.

(2013). Testing the German 300 item IPIP scale: The translation and convergent validation of the 300-Item IPIP scale with its well-known counterpart, the NEO-PI-R. Poster presented at the World Conference on Personality,

Stellenbosch, South

Africa. 19-23 March 2013.

63.

Weijters

Baumgartner

(2022). On the use of balanced item parceling to counter acquiescence bias in structural equation models. Organizational Research Methods, 25(1), 170–180. https://doi.org/10.1177/1094428121991909

64.

Widiger

T. A.

Oltmann

J. R.

McCabe

G. A.

Smith

(2020). The HEXACO model of personality: Compelling objections. European Journal of Personality, 34, 558–559.

65.

Wiernik

B. M.

Yarkoni

Giordano

C. A.

Raghavan

(2020). Two, five, eight (thousand): Time to end the dimension reduction debate. European Journal of Personality, 34(4), 559–560.

66.

Zettler

Thielmann

Hilbig

B. E.

Moshagen

(2020). The nomological net of the HEXACO model of personality: A large-scale meta-analytic investigation. Perspectives on Psychological Science: A Journal of the Association for Psychological Science, 15(3), 723–760. https://doi.org/10.1177/1745691619895036

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

0.74 MB