Indigenous psychological assessment aims at assessing psychological phenomena with a special focus on the cultural context in which they occur. Indigenous psychological traditions have developed in parallel in different non-Western regions of the world. In the domain of personality psychology, these developments have been slower in Africa than in other world regions. The first aim of this chapter is to describe a large research project investigating personality from an indigenous perspective in South Africa—the South African Personality Inventory (SAPI) project. We start with a brief overview of indigenous approaches to personality assessment in general. We then introduce the diverse cultural and linguistic background of South Africa, followed by a description of historical and current political issues relevant to assessment. To place the SAPI in a broader framework, we review the established practice of Western test adaptation in South Africa and summarize other personality research and measures in Africa. We present the SAPI project in detail, focusing on its qualitative and quantitative components in turn. The second aim of this chapter is more conceptual. While in the past emic and etic studies of personality were conducted in a rather independent manner, we think that the time has come for a rapprochement. We describe how emic and etic studies are both needed for an inclusive cross-cultural psychology. The theme is discussed in the introduction, where we describe the conceptual framework of the South African study, and we return to it in the final section, where we describe implications of recent emic studies of personality structure for cross-cultural work in personality.
Indigenous psychology developed in the second half of the 20th century as a response to the dominant role in psychology of Western conceptual models and assessment tools. Two types of problems can be associated with the use of Western assessment instruments across cultures. The first group of problems revolves around the question to what extent the instrument and its components adequately measure the target concepts in cultures different from the instrument’s culture of origin. These are problems of construct and measurement equivalence (van de Vijver & Leung, 1997), addressed in adaptations of instruments such as the Wechsler intelligence scales (Georgas, Weiss, van de Vijver, & Saklofske, 2003) and the Minnesota Multiphasic Personality Inventory (Butcher, 1996; Butcher, Lim, & Nezami, 1998) for use in diverse cultures. The second type of problems refers to questions of comprehensiveness and relevance of Western models for specific cultural contexts (Dana, 2000). How relevant are Western psychological concepts in non-Western cultures, and how well do they cover the psychological domain of interest? This second type of problems is addressed in indigenous studies.
Indigenous psychological studies have developed an especially rich tradition in Asian cultures (for overviews, see Cheung, Cheung, Wada, & Zhang, 2006; Cheung & Fetvadjiev, 2016; Church, 2010; U. Kim, Yang, & Hwang, 2006). Researchers have identified a number of specific personality concepts, such as amae (indulgent dependency in Japan; Doi, 1973), cheong (group-related affect in Korea; Choi & Choi, 2001; Choi & Lee, 1999), kapwa or pakikipagkapwa (togetherness, shared identity in the Philippines; Enriquez, 1978, 1997), renqing (relational orientation in China; Ng, Fan, Cheung, Leong, & Cheung, 2011; Zhang & Bond, 1998), the notion of selfless self (India; Ho, 1995; Mosig, 2006), and yuan (predestined relational affinity in China and Taiwan; Yang & Ho, 1988). Extensive indigenous studies have also been conducted in Mexico, where the simpatía personality concept (consideration and conflict-avoidance; B. S. Kim, Soliz, Orellana, & Alamilla, 2009; Triandis, Marin, Lisansky, & Betancourt, 1984) is often encountered (diverse Mexican concepts and measures are reviewed in Ortiz et al., 2007). Many of these indigenous concepts share a substantive focus on social and relational functioning. A common feature of the early indigenous studies is that their focal concepts, derived from local cultural-philosophical traditions, were often treated in isolation from other existing psychological models. This has restricted the possibilities of making direct comparisons between local and hypothesized universal personality models (Cheung, van de Vijver, & Leong, 2011).
More recent research has directly assessed the overlap of indigenous concepts with the Five-Factor or “Big Five” Model of personality, comprised of Extraversion, Agreeableness, Conscientiousness, Emotional Stability (vs. Neuroticism), and Intellect or Openness to Experience (Goldberg, 1990; McCrae & John, 1992). Studies in the Philippines (Katigbak, Church, Guanzon-Lapeña, Carlota, & del Pilar, 2002) and Mexico (Ortiz et al., 2007) have found that most indigenous concepts could be subsumed in the Big Five. Transcending historical divisions between universal (etic) and indigenous (emic) constructs, a new wave of research aims to integrate local and universal elements of personality in a unified, emic-etic framework in model and instrument development (Cheung et al., 2011). Pioneering work in this direction was done by Cheung and her colleagues in the development of the Cross-Cultural (Chinese) Personality Assessment Inventory (CPAI-2; Cheung et al., 2001; Cheung, Fan, & Cheung, Chapter 4 in this volume). The CPAI-2, developed with a bottom-up approach starting with qualitative and ethnographic data, provides a comprehensive assessment of personality concepts deemed relevant in China. The CPAI-2 model was found to have partial overlap with the Big Five. The model includes an additional factor, Interpersonal Relatedness (IR), which is not captured by Five-Factor-Model measures. The SAPI project followed a similar approach as the CPAI-2 and forms a new major initiative in the emic-etic framework (Cheung et al., 2011).
In contrast to the rich tradition in Asia and Mexico, Africa has received scarce attention in indigenous personality studies, despite being the second largest and most populous continent, characterized by a large diversity of languages and cultures. Africa has been the ground for the emergence of important indigenous models of intelligence (e.g., Serpell, 2011) and has a variety of folklore, spiritual, and philosophical conceptualizations of personality. However, apart from a recent cross-cultural project that included psycholexical data from four African languages (two studies by Saucier and his colleagues in 2014, discussed in the section on indigenous personality in Africa), there has been no systematic psychological research on personality from an indigenous perspective in an African context. The SAPI is the first project to start filling this gap.
South Africa has four major, distinct ethnocultural groups, according to a classification system inherited from apartheid: Blacks (a term used for people of African descent, 79% of the population), Coloureds (mixed descent, 9%), Indians/Asians (Asian descent, 3%), and Whites (European descent, 9%; Statistics South Africa, 2012). The extent of separation across these groups has varied with time; as an official state policy, separation was the strongest in the apartheid regime during most of the second half of the 20th century. After the abolishment of apartheid in the 1990s, there have been efforts for greater social integration of cultural groups. Language plays an important role for both group identification and integration across groups.
The 1996 constitution of South Africa instituted 11 official languages and stipulated that at least 2 of them should be used by national and provincial government. The constitution also recognized and declared support for the use of several other nonofficial languages, including the Khoisan languages; languages of large communities of European and Indian origin; and languages with religious significance, such as Arabic, Hebrew, and Sanskrit.
The 11 official languages of South Africa include 9 Bantu and 2 Germanic languages. The Bantu language group, covering most of sub-Saharan Africa, contains between 300 and 500 languages and is part of the Niger-Congo language family, the largest language family in the world with about 1,500 languages (Nurse & Philippson, 2003). The nine official Bantu languages in South Africa are Northern Sotho, Southern Sotho, Tswana (in the Sotho-Tswana group), Ndebele, Swati, Xhosa, Zulu (in the Nguni group), Tsonga, and Venda. There is some mutual intelligibility across these languages. The two Germanic languages are Afrikaans and English. The Bantu languages are spoken as first language primarily by Blacks. The first-language use of Afrikaans is shared primarily by Coloureds and a part of the White group; 14% of the country’s population is first-language Afrikaans speakers. Finally, English is spoken as a first language by 10% of the population, spread across all four ethnic groups.
There is substantial regional variation in language prevalence. For example, Xhosa and Zulu, the most popular Bantu languages in South Africa, are spoken as first language predominantly in the provinces along the east coast; the Sotho-Tswana languages, in the higher inland; and Afrikaans, on the west coast. Although English is the first language of only a small part of the population, it is commonly used and understood across groups and functions as a lingua franca. English is becoming increasingly popular in Blacks as the preferred language for education at the expense of Bantu languages (Luiz, 2015). It is used extensively in business and organizational settings as well as in psychological assessment.
The Bantu languages have a rich oral tradition but no written tradition in southern Africa prior to European contact. Written texts (initially translations of religious texts, later creative works of native authors) only started appearing from the 18th to 19th century on (Doke, 1954). Lexicography has only developed slowly. There are authoritative bilingual (English) dictionaries for some of the more popular languages such as Zulu, but the lexicographical resources for less popular languages such as Ndebele are very limited.
A former Dutch, then British colony, South Africa as a Union came into existence in 1910. Ethnic and racial segregation had been a reality from early stages, but the policy of apartheid was formally adopted in 1948 with the Nationalist Party coming into power.
The history of psychometric test use in the country has been closely linked to the apartheid project (Foxcroft & Roodt, 2013; Laher & Cockcroft, 2014). Already in the early 20th century, Black children were tested on cognitive tests normed for White children, and lower scores were attributed to Blacks’ inferior intelligence. Three main features characterized psychological test usage for much of the following period: a focus on developing and standardizing measures for Whites only; administering measures standardized for one group to another group without investigating whether or not the measures might be biased and inappropriate for the target group; and conclusions about group differences without considering the impact of cultural, linguistic, and socioeconomic factors on test performance (Meiring, 2007).
Psychological assessment was initially focused on educational applications and was later extended to vocational selection and assessment. With this extension, personality assessment became relevant. Some early examples of descriptions of native Africans’ personalities are found in works by Biesheuvel in the 1950s (with descriptions such as “liking of repetitive action”) and in Thematic Apperception Test data by De Ridder in the 1960s (e.g., “strong latent aggression and insufficient moderation and control”; both authors cited in Laher & Cockcroft, 2014, p. 305). With the increasing international isolation of South Africa from the 1960s on, imported tests became inaccessible, and this encouraged the development of local tests, such as the South African Personality Questionnaire (Retief, 1992). However, these local tests were usually modeled after Western templates and were developed and normed in the White group (Laher & Cockcroft, 2014).
Apartheid included a policy of job reservation, which implied that individuals from different groups could not compete for the same jobs. This policy was abolished in the 1980s, leading to a renewed need for fair and unbiased measures. The first studies on bias were conducted in the late 1980s and pointed to bias in both ability and personality measures (Meiring, 2007).
After the end of apartheid in the 1990s, there were strong public sentiments against psychological assessment in general, associated with the perception that assessment had served to legitimize and perpetuate racial segregation. A telling example is the fact that in the early drafts of the Employment Equity Act (discussed in the following section), psychometric testing for employment purposes was banned altogether (Laher & Cockcroft, 2014). It took some vigorous engagement with the legislature to overcome this resistance.
The legislative effort to protect individuals from unfair discrimination in assessment was embodied in the Employment Equity Act 55 of 1998, amended in 2014 (Republic of South Africa, 2014). It stipulates that psychometric testing of employees and applicants is “prohibited unless the test or assessment being used (a) has been scientifically shown to be valid and reliable; (b) can be applied fairly to employees; and (c) is not biased against any employee or group.” The 2014 amendment included the clause that tests are certified by a state-authorized official body. Overview studies indicate that many instruments currently in use have not been examined for bias and equivalence; where bias analyses have been conducted, evidence of bias has often been found (Foxcroft & Roodt, 2013; Meiring, 2007).
The prevailing practice in personality research and assessment in South Africa has been to use directly imported or adapted Western instruments, such as the Eysenck Personality Questionnaire, Fifteen Factor Questionnaire Plus (15FQ+), Sixteen Personality Factor Questionnaire (16PF), and Revised NEO Personality Inventory (NEO-PI-R). Meiring, van de Vijver, and Rothmann (2006) found persistent problems with low reliability of the 15FQ+ in Blacks. Language comprehension problems in non-first-language English speakers have been identified in different tests but especially in the 16PF; no systematic studies of bias and equivalence of the 16PF have been conducted (van Eeden, Taylor, & Prinsloo, 2013). There is evidence for structure equivalence of the NEO-PI-R across Blacks and Whites, which means that exploratory factor analyses yielded the same factors in both groups; yet, data have come mostly from student samples, and the argument has been made that the inventory may not cover the implicit personality concepts in South Africa comprehensively (Laher, 2013).
The construction of the Basic Traits Inventory (BTI; Taylor & De Bruin, 2005) was one of the few recent cases of local test development. The BTI uses items adapted from various sources such as the International Personality Item Pool (IPIP; Goldberg et al., 2006; http://ipip.ori.org/) to measure the Big Five, taking the South African context into account. The instrument has been found to have structure equivalence across ethnic groups (Ramsay, Taylor, De Bruin, & Meiring, 2008).
In summary, there has been an increased demand for measures with demonstrated validity and equivalence across groups in South Africa, and measures currently in use have only met this demand to a limited extent, perhaps with the exception of the BTI. Despite the perceived need to integrate local perspectives in psychological assessment, no previous research has addressed indigenous personality concepts across groups in South Africa.
The notion of a personality pattern common to the diverse cultures of Africa and distinct from other cultural contexts has long fascinated students of culture and personality. An exact term search for “African personality” in PsycINFO rendered 30 results, of which 16 refer to personality patterns and models (rather than questionnaire labels). In contrast, the search for “Asian personality” rendered only one result. Yet, the ratio of actual psychological studies conducted on indigenous personality conceptions on the two continents is rather the opposite.
LeVine (1973) came up with seven features he expected to be common to sub-Saharan Africa based on personal experience, observations, and reading of ethnographic literature, such as pronounced gender differentiation, emphasis on material transactions in interpersonal relations, tendency to blame others, and low separation anxiety. These descriptions were made with all reservations as to their validity and were meant to invite future research. However, systematic research into personality from a local perspective in Africa never gained momentum.
The possibility of a common African personality profile has recently resurfaced in the framework of the broad etic studies on personality profiles of cultures (McCrae et al., 2005; Schmitt et al., 2007). A recurrent finding in these studies was that some of the instances of lowest structure equivalence to the target U.S. personality structure were found for data from African cultures. A possible explanation would be that African cultures might share some personality features that make them distinct from Western cultures. However, no evidence has been found for such a common African profile; the structure equivalence among the African cultures tended to be similar to the equivalence with Western targets (McCrae et al., 2005; Zecca et al., 2013). The lack of equivalence has thus been attributed to data quality problems and lack of corresponding trait terms in some of the African languages. This still leaves the possibility open that there are important personality concepts across cultures in Africa that are not well covered by the Western models and instruments.
Data relevant to the implicit personality conceptions in Africa can be found in a recent psycholexical project by Saucier and his colleagues. In a first and broader study, Saucier, Thalmayer, and Bel-Bahar (2014) collated human-attribute terms of 12 isolated languages, including 4 languages representing the major African language families: Afar (of the Afro-Asiatic family), Khoekhoe (Khoisan), Maa (Nilotic), and Supyire (Niger-Congo). The authors suggested a two-dimensional model (along the lines of agency and communion) as most common across the languages. It is worth noting that this study only analyzed the occurrence of dictionary terms, not actual person descriptions or ratings; the terms came from bilingual (English) dictionaries and had rather low overall numbers (e.g., 678 human-attribute terms in Supyire). In addition, the aim was to derive a common, etic structure with a least common denominator rather than a rich interpretation of the implicit model in any single language or groups of languages. In a second, more focused study, Saucier, Thalmayer, Payne, et al. (2014) analyzed peer-ratings in Maa and Supyire alongside previous psycholexical data obtained using various criteria from seven other, non-African languages. This study examined exclusively a similar, two-dimensional model specified a priori as common across languages. A more intensive future exploration of indigenous personality concepts in Africa could build upon these data.
One concept that is perceived to capture important aspects of personality in Africa is “ubuntu,” a Bantu term that means humanness (“ubuntu” and “bantu” have the same root, denoting human beings). The term is encountered in different forms across Bantu languages, often expressed in proverbs and sayings (Kamwangamalu, 1999). A popular saying is that “a person is a person through other persons” (in Zulu, “Umuntu ngumuntu ngabanye abantu”). Ubuntu goes beyond personality traits and includes elements of values, norms, and moral standards. The main characteristics of ubuntu are the relational definition of the self; valuing customs and traditions; respect and sensitivity to others’ needs; the importance of solidarity and community involvement; reconciliation and restorative justice; and a general emphasis on relationship harmony (Kamwangamalu, 1999; Metz & Gaie, 2010). Ubuntu defines a broad spectrum of personality-relevant elements, but it has so far not been translated into a specific profile of personality traits.
The SAPI project was initiated to address both the practical need for a fair and unbiased personality measure and the theoretical questions of indigenous personality in South Africa. The project aimed to develop an indigenous model and an instrument for its measurement, covering the implicit personality conceptions deemed relevant across ethnic groups and languages. The project included two extensive stages, a qualitative and a quantitative stage, reviewed in the following sections.
The qualitative stage aimed to uncover the implicit personality conceptions in all 11 languages of South Africa, as manifested in free personality descriptions. The procedures and findings of this stage are presented in detail in Nel et al. (2012), Valchev, Nel, et al. (2013), and Valchev et al. (2011).
The most popular approach to indigenous personality study is the psycholexical approach, which involves the selection of personality-descriptive terms from dictionaries and analysis of self- and peer-ratings on these terms (see De Raad & Mlačić, Chapter 6 in this volume). We chose to analyze free personality descriptions instead, which offered three advantages. First, there were no well-developed dictionaries for all 11 languages of South Africa; second, free descriptions translated into English were directly accessible to all members of the research team; third, free descriptions could be expected to have enhanced ecological validity compared to single terms.
Interviews were conducted with native speakers of the 11 languages, in their own language, asking them to describe themselves and nine other persons they know well. Between 107 (Ndebele) and 122 (Tswana) persons per group were interviewed, except for Afrikaans (n = 70) and Southern Sotho (n = 62), with a total N = 1,216. Variation in gender, age, residence, and educational background was ensured. The targets included self-descriptions, family and community members, and people known from school or work and included both liked and disliked persons. The interviews were conducted, recorded, transcribed and entered into Excel, and translated into English by master’s students who were trained as field workers for this study. The total number of descriptions used in the analysis was 53,139. (For further details, see Nel et al., 2012.)
Translators and language experts independently assessed the quality of the translations, and changes were made where needed. The data analysis was discussed regularly between the project members. The first categorization outcomes were discussed at a workshop with cultural and linguistic experts on all languages. Their feedback was taken into account in adjusting the analysis categories. Individual discussions with cultural experts and a personality expert1 were held at the final analysis stages.
The data were cleaned, leaving out physical descriptions, pure evaluative terms, responses referring to life circumstances such as marital status and profession, and other idiosyncratic responses. The analysis included three stages: labeling, categorization, and semantic clustering (cf. Peabody, 1987). In the labeling stage, single terms were assigned to each response (e.g., the label loving for responses such as “he loves people” and “we are fond of each other”). There were over 900 personality-descriptive labels. In the categorization stage, we condensed the number of labels into fewer categories by grouping clear synonyms (e.g., outgoing and socializing) and antonyms (e.g., quiet and talkative) together into facets, taking into account the patterns of co-occurrence of responses. There were 188 facets: 79 were extracted in all 11 languages, 71 in 7 to 10 languages, 28 in 3 to 6 languages, and 10 in 1 or 2 languages. In the semantic clustering stage, we grouped the facets into 37 mid-level subclusters (in an earlier analysis of the Nguni language data, this level of abstraction was referred to as 26 clusters; Valchev et al., 2011) and 9 broad clusters. This analysis aimed to maximize the homogeneity within each subcluster and the heterogeneity across clusters. The 9 clusters were shared across all 11 languages.
The nine clusters (presented with their subclusters in the left panel of Table 5.1) are: Conscientiousness, Emotional Stability, Extraversion, Facilitating, Integrity, Intellect, Openness, Relationship Harmony, and Soft-heartedness. They are at a similar level of abstraction as the Big Five and are fairly similar to it. The main point at which the two models differ is the richer representation of social-relational concepts in the nine-cluster model. Facilitating (the ability to guide others in life), Integrity (dealing with the values of honesty and trustworthiness), Relationship Harmony (focusing on maintaining and promoting harmony), and Soft-heartedness (with properties such as empathy and consideration) can all be seen as an extension of the social-relational domain, mostly represented by Agreeableness in the Big Five. However, the model seemed to reach beyond the Big Five by identifying concepts similar to CPAI-2 IR (Cheung et al., 2001) and Honesty of the HEXACO model (Lee & Ashton, 2008) as well as local concepts such as Facilitating, not well known from other models.
In a first attempt to validate the model, we assessed the internal structure of the 37 subclusters by asking 204 students in South Africa and 95 students in the Netherlands to rate the degree of relatedness between all 666 possible pairs of subclusters (Nel et al., 2012, Study 2). The ratings were subjected to a hierarchical cluster analysis. The results broadly supported the conceptual model, although they did not replicate it in detail. There was a dominant distinction between positive and negative characteristics as well as between person-focused and relationship-focused characteristics.
Although the overall nine-cluster model was common to all groups, we found that the groups differed in the emphasis they placed on its components: Blacks referred more often to social-relational descriptions and social norms, whereas Whites referred more often to personal-growth descriptions (Valchev, Nel, et al., 2013). Furthermore, Blacks more often described persons’ specific behaviors placed in context, whereas Whites tended to make more abstract and decontextualized personality descriptions (Valchev, van de Vijver, Nel, Rothmann, & Meiring, 2013); Coloureds and Indians had an intermediate pattern.
In summary, the qualitative study of implicit personality conceptions in South Africa followed a multistep approach, with consultations with language and cultural experts; an analysis of the data first in a subset of three Nguni languages; a complete analysis of all data; and a partial quantitative validation focusing on the mid-level subclusters.
The transition to the quantitative measurement stage (described in detail in Fetvadjiev, Meiring, van de Vijver, Nel, & Hill, 2015; Hill et al., 2013) involved extensive item generation, using the original person descriptions from the qualitative data as input. Each of the 188 low-level facets was defined based on the content of its responses. We then developed between 2 and 34 (M = 13) items for each facet, leading to a total of 2,574 items. We used similar formulation criteria to those used in the development of the Five-Factor Personality Inventory (Hendriks, Hofstee, & De Raad, 1999). Items were developed in English but with translatability in mind. We used simple language with no negations and specified concrete behaviors with an object whenever possible (e.g., “I help others cope with their problems”). The items often contained direct or paraphrased statements found in the qualitative data.
We examined the items in separate pilot studies for each cluster of the qualitative data. For example, all 155 Conscientiousness items were administered to a sample of 1,041 participants. The larger Relationship Harmony and Soft-heartedness clusters were split in two each. There were 117 to 482 items and 439 to 1,041 participants, per cluster. We aimed to replicate the occurrence of subclusters per cluster in exploratory factor analysis. We selected items with loadings of at least .30 (or .40 if needed for improved distinction) and low secondary loadings, with mean scores between 1.50 and 4.50 on a five-point Likert scale from 1 to 5 and with low skewness and kurtosis. These criteria still left many fitting items to choose from. The substantive criteria were thus equally important; we selected items that (a) maximized construct representation, (b) minimized content overlap within and across clusters, and (c) were most in line with the language formulation rules. We eventually selected 571 items to be translated into the remaining 10 languages.
Qualitative Model |
Quantitative Model |
||
|---|---|---|---|
| Cluster | Subcluster (Number of Facets) | Factor | Facet Scale (Number of Items) |
| Conscientiousness | Achievement Orientation (6) | Conscientiousness | Achievement Orientation (10) |
| Dedication (6) | Integritya (12) | ||
| Orderliness (7) | Orderliness (11) | ||
| Self-discipline (6) | Traditionalism-Religiosity (4) | ||
| Thoughtlessness (2) | |||
| Emotional Stability | Balance (4) | Neuroticism | Emotional Balance (8) |
| Courage (2) | Negative Emotionality (10) | ||
| Ego Strength (5) | |||
| Emotional Control (4) | |||
| Emotional Sensitivity (3) | |||
| Neuroticism (5) | |||
| Extraversion | Dominance (4) | Extraversion | Playfulness (6) |
| Expressiveness (6) | Sociability (7) | ||
| Positive Emotionality (6) | |||
| Sociability (8) | |||
| Facilitating | Encouraging Others (4) | ||
| Guidance (7) | |||
| Integrity | Fairness (2) | ||
| Integrity (8) | |||
| Intellect | Aesthetics (5) | Openness | Broad-mindedness (5) |
| Reasoning (4) | Epistemic Curiosity (6) | ||
| Skillfulness (4) | Intellect (10) | ||
| Social Intellect (3) | |||
| Openness | Broad-mindedness (9) | ||
| Epistemic Curiosity (3) | |||
| Materialism (2) | |||
| Openness to Experience (2) | |||
| Relationship Harmony | Approachability (10) | Positive Social-Relational | Facilitating (10) |
| Conflict-seeking (3) | Integrity (12) | ||
| Interpersonal Relatedness (8) | Interpersonal Relatedness (9) | ||
| Meddlesomeness (2) | Social Intelligence (4) | ||
| Soft-heartedness | Active Support (6) | Warm-heartedness (12) | |
| Amiability (6) | Negative Social-Relational | Conflict-seeking (6) | |
| Egoism (5) | Deceitfulness (3) | ||
| Empathy (8) | Hostility-Egoism (13) | ||
| Gratefulness (2) | |||
| Hostility (11) | |||
Note. For presentation purposes, the factors of the quantitative model are arranged to match the clusters of the qualitative model as closely as possible. Clusters and subclusters in the qualitative model and facet scales in the quantitative model are presented alphabetically. a In the pooled data across ethnic groups, the Integrity facet had an equal double loading on Conscientiousness and Positive Social-Relational and is thus included in both factors. More recent data suggest that Integrity may be more strongly linked with the Positive Social-Relational factor. |
|||
In the final selection round, we combined translators’ feedback with our criteria. We removed a total of 321 items, which were potentially difficult (using idiomatic expressions), complex (longer than 10 words in any of the 11 languages), or used abstract trait terms. The final selection of items to be administered jointly in a single questionnaire included 250 items.
The 250-item questionnaire was administered to a total sample of 1,364 participants, which included university students and adults from the general population from the four ethnic groups: 671 Blacks, 198 Coloureds, 104 Indians, and 391 Whites. Ages ranged from 18 to 73 years, with mean ages between 27 (in the Indian group) and 35 years (in the White group). The percentage of females ranged from 53% (in Blacks) to 85% (in Whites). (For further details, see Fetvadjiev et al., 2015.)
All 250 items were administered. In preliminary analyses, items that reduced the replicability of the factors across groups, reduced internal-consistency reliability, or had factor loadings below .30 (or .40, as explained earlier) were removed in iterative steps. The final item set contains 146 items. The items were grouped in the following 18 facet scales, derived from exploratory factor analyses per cluster: Facilitating, Integrity, Social Intelligence, IR, Warm-heartedness, Deceitfulness, Conflict-seeking, Hostility-Egoism, Emotional Balance, Negative Emotionality, Playfulness, Sociability, Achievement Orientation, Orderliness, Traditionalism-Religiosity, Intellect, Broad-mindedness, and Epistemic Curiosity. All scales were unipolar, with items formulated in the direction of the target construct, and items were presented in a random order. Aggregated across groups, Cronbach’s alpha ranged from .61 to .86 for the different scales, with a grand mean of .77. The facet scales correspond in level of abstraction to the 37 subclusters of the qualitative model.
We ran exploratory factor analysis on the 18 facet scales in the pooled-within correlation matrix (i.e., the correlation matrix of all items in which the data of all groups are combined, correcting for differences in item means across groups and for differential sample sizes across groups). We identified six factors, labeled Social-Relational Positive, Social-Relational Negative, Neuroticism, Extraversion, Conscientiousness, and Openness. An overview of the factors and their defining facets is presented in the right panel of Table 5.1. The structure resembled the Big Five, except for the separate Positive and Negative Social-Relational factors, which appeared broader than the typical measures of Agreeableness.
We evaluated the correspondence of the factors across groups using Tucker’s phi (Tucker, 1951). This is a measure of factorial similarity, gauging the extent to which two factors obtained in different groups are similar. Values above .85 (Lorenzo-Seva & ten Berge, 2006) or .90 (van de Vijver & Leung, 1997) are taken as evidence of factorial similarity. The six-factor structure replicated well across the ethnic groups: Tucker’s phi congruence coefficients across groups and factors ranged from .85 to .99, and only 4 out of 24 coefficients were below .95. We replicated the structure in an independent sample of 139 Black and 270 White students. The Tucker’s coefficients ranged from .87 to .97, pointing to fair replicability.
The extent to which an indigenous personality model overlaps with a universal model is usually assessed using either a joint factor analysis of the two measures or multiple regression analysis, in which the universal model predicts the elements of the indigenous model (Cheung et al., 2001; Katigbak et al., 2002; Ortiz et al., 2007). We used both techniques, as the issue of overlap and incremental value of indigenous factors is vital for emic research. In an early stage of the SAPI development, we administered versions of only the social-relational scales (Valchev et al., 2014) together with Big Five measures; in the most recent stages, we used the complete SAPI (Fetvadjiev et al., 2015). We assessed the extent of overlap in Black and White students in South Africa as well as in a multicultural sample in the Netherlands. The Big Five was assessed using the BTI (Taylor & De Bruin, 2005), which is a well-established Big-Five instrument in South Africa, and 50 items from the IPIP (Goldberg et al., 2006). Across these different samples and measures, we consistently found that the social-relational scales form two additional factors that do not merge with the Big Five. When fewer than seven factors were extracted, other factors tended to merge, but the two social-relational factors (sometimes merged into a bipolar factor) remained distinct from the Big Five. We found no support for rival interpretations, such as bloated specifics, valence, and skewness of the score distributions (Valchev et al., 2014).
In multiple regression analyses of indigenous concepts predicted by the dimensions of universal models, R values of .40 or lower have been suggested as pointing to relatively distinct indigenous concepts (Katigbak et al., 2002; Ortiz et al., 2007). We found a number of indigenous concepts that had R values below .40 in the study that included only the social-relational scales (Valchev et al., 2014) but not in the study on the complete SAPI (Fetvadjiev et al., 2015). This may have to do with the limited sample employed in the latter study (only Whites) and with the weaker representation of some social-relational aspects in the complete SAPI. Nonetheless, we argued that a criterion value of R ≤ .40 may be too low, pointing out that some Big Five factors may also fail to be recognized as distinct factors by this criterion (Fetvadjiev et al., 2015). In line with practices in meta-analysis (Hunter & Schmidt, 1990), we argue that an R2 of .75 may be a better threshold for accepting that the target indigenous concept is fully accounted for by the universal model.
A crucial question for the interpretation of indigenous personality concepts is to what extent they offer any added predictive value beyond universal models. We examined the predictive power of the social-relational scales by using prosocialness as an outcome variable (Valchev et al., 2014). Prosocialness, which has been found to be moderately to strongly associated with Agreeableness, was assessed using Caprara, Steca, Zelli, and Capanna’s (2005) 16-item scale. The Big Five was measured with a short, 60-item version of the BTI and the SAPI social-relational domain with two broad social-relational scales (63 items). In hierarchical multiple regression, the inclusion of the SAPI scales increased the adjusted R2 value from .38 to .55, F(2, 317) = 63.81, p < .001. In summary, we found evidence for substantial incremental value of the SAPI social-relational scales. Future research will have to expand this evidence.
There are several research directions that are currently pursued in the SAPI project. One of the first aims is to develop population norms and assess differential item functioning. It will be important to assess and further develop the diverse language versions. Another central aim is to broaden the nomological network of the SAPI by examining its links to other personality models, values, social axioms, and life outcomes such as work-life balance and academic achievement. A study currently underway investigates the predictive power of the SAPI for diverse everyday behaviors in Blacks and Whites. To what extent are the cultural differences in use of abstract versus concrete personality descriptions found in the qualitative data, matched by corresponding differences in the predictive validity of personality ratings? The findings from diary and observational studies suggest that the implicit differences in the qualitative data are more substantial than any actual differences in traits’ predictive validity. On another point, in the quantitative stage so far, only literate populations have been studied. Future studies could assess the quantitative model’s applicability in illiterate individuals. Finally, it will be interesting to see how well the SAPI travels across borders. The SAPI has been developed with an explicit aim to be easily accessible and translatable for diverse groups. Findings from the Netherlands (Valchev et al., 2014) suggest that it is reasonable to expect the model to replicate in other cultural contexts. This would point to a broader need to expand the Big Five to measure and predict behavior in the social-relational domain not only in South Africa but also across cultures. We return to these broader implications in the final section.
Methods similar to those of the SAPI project have recently been applied in other psychological domains and cultural areas. The qualitative approach used in the SAPI has been directly transferred to a study aiming to develop an indigenous model of social desirability across groups in South Africa (Nel, 2014). Some of the themes that emerged were similar to concepts of the SAPI model, such as Facilitating, Intellect-Openness, and IR. In a first systematic study on indigenous personality in the Arab context in the Levant, Zeinoun, Daouk-Oyry, and van de Vijver (2014) extracted personality-descriptive terms in parallel using the mainstream psycholexical dictionary method and the method of person descriptions in interviews. The resulting models from the two approaches were complementary but not identical. The direct comparison of the psycholexical approach with the free-descriptions approach opens an intriguing avenue for future research.
The benefits of combining universal and local elements in psychological theory and assessment have become more salient in the past decades (Cheung et al., 2011). In multilingual and multicultural contexts such as South Africa, the challenges of fair assessment are particularly visible. The SAPI project makes an important contribution to addressing both theoretical questions of indigenous personality models and practical questions of instrument development. To start with the latter, the finding of good psychometric properties and structure equivalence of the new instrument is reassuring. The instrument development had a continuous focus on the inclusive representation of concepts relevant across groups and on translatability of assessment items. The outcomes show that posing equivalence as an explicit aim from the onset in instrument development can help to achieve at least structure equivalence of the measures across cultural groups.
With respect to model development, it is interesting to note that despite the careful procedures for translating the elements of the qualitative model into assessment items, the qualitative model was not reproduced exactly in the quantitative model; in fact, this was not an objective. The qualitative model served as a conceptual framework to organize the main components of the implicit personality conceptions across the 11 languages in South Africa. The quantitative model represents the empirical structure underlying the association patterns of these components. Although the models are not identical, they agree in the overall structure, and both point to a Big-Five-like structure with an expanded social-relational domain. So, the two approaches elucidate different parts of the indigenous personality model in South Africa. We believe both qualitative and quantitative approaches are indispensable in indigenous research.
Another noteworthy point is that the SAPI social-relational factors are related both conceptually and empirically to the CPAI-2 IR dimension (Valchev et al., 2014). This is interesting because, unlike the CPAI-2 project, the SAPI project did not start with a strong expectation of culturally specific elements in South Africa. Ubuntu was seen as a broad framework but was not necessarily expected (and was not found) to emerge as a personality concept in its own right. Perhaps the qualitative methods, similar across the two projects, may predispose toward the more pronounced emergence of social-relational concepts, which are also most numerous in psycholexical data. Future research should address the extent to which there is such a method effect across cultures.
It is in the nature of emic studies to unravel psychological structures, personality in our case, in a specific cultural context. If the results of the South African study are compared with other indigenous personality studies we reviewed from Mexico, the Philippines, and China, it becomes clear that emic studies do not find entirely dissimilar and culture-specific structures even if the cultures that are studied are very dissimilar. More specifically, in cultures as diverse and far apart as Mexico, the Philippines, China, and South Africa, the typically Western personality structure needed to be adapted in the social-relational domain. It is easy to construe such findings as conflicting with Western results. The core of the conflict would then revolve around the question of whether Agreeableness as found in Western studies can encompass all the components of social aspects of personality found in non-Western cultures. We would argue that our findings in South Africa suggest that social-relational aspects of personality, as measured by the SAPI, go beyond Agreeableness to predict (self-reported) prosocial behavior. Still, we appreciate that staunch defenders of a Big Five model may want to disagree with this conclusion and argue that additional factors, such as response styles, personal values, or other constructs outside of the personality domain, are captured by social-relational scales and that a linear combination of Big Five scales gives the only and best prediction of prosocial behavior from the perspective of personality.
The latter viewpoint would be more plausible if our findings would stand alone. However, that is not the case. Therefore, we are inclined to think that in the social-relational domain, personality structure can and should be enriched by factors currently not considered in traditional personality models to make the models more comprehensive and predictive of behavior. Moreover, and this is central to our reasoning about the implications of our project for the field, we would argue that our findings are not conflicting with but complementary to Big Five studies, in line with the need to combine emic and etic approaches (Cheung et al., 2011; van de Vijver, 2013).
Extant personality models have implicitly started from a “one size fits all” framework in which personality structure is assumed to be universal. However, the assumption itself has never been critically questioned. If anything, non-Western indigenous studies of personality show that personality structure is neither completely universal nor completely culture-specific. There is no a priori reason to expect that an indigenous approach would be more suitable to identify most personality factors, such as extraversion, conscientiousness, and emotional stability (as most studies have done), but would be less suitable to identify agreeableness (or relational aspects of personality in general). The convergence in findings of the indigenous studies suggests that factors like extraversion, emotional stability, conscientiousness, and presumably also openness are relevant to describe individual differences in many cultures and that these factors are also reflected in the implicit personality theories of speakers of many languages. However, the convergence also suggests that this cross-cultural stability is less the case in the social-relational domain. A possible reason for this lack of cross-cultural stability in the social-relational domain may be that traits in this domain are influenced by social norms and presumably more so than a trait like emotional stability.
Viewing the results of emic and etic studies as complementary rather than conflicting and developing conceptual frameworks that can capture both sets of results can be seen as one of the goals of cross-cultural psychology (Berry, Poortinga, Breugelmans, Chasiotis, & Sam, 2011; Cheung et al., 2011; van de Vijver, 2013). A brief historical detour is needed here. In the early days of cross-cultural psychology, say in the 1960s and the 1970s, most studies compared different cultures on usually Western instruments, with the aim of identifying differences (and to a small extent similarities) across cultures. At a later stage, when many of these comparative studies have been conducted, it becomes possible to identify the patterning of cross-cultural differences. Models of cross-cultural differences, such as Hofstede’s (1980, 2001) model of work-related values, are then used to examine cross-cultural differences. For example, many studies have used individualism-collectivism as a conceptual framework to understand differences between Eastern and Western countries (Oyserman, Coon, & Kemmelmeier, 2002; Triandis, 1995). In the final stage of cross-cultural psychology, the first two stages are integrated by extending or developing psychological theories so as to include cross-cultural differences.
Applied to personality psychology, these three stages could be translated as firstly conducting cross-cultural studies using Western instruments (with an implicit assumption that the Western personality structure is universal); this is followed by a stage in which the patterning of the cross-cultural differences is addressed. Examples are studies of mean scores obtained in different countries (“the geography of personality”; Allik & McCrae, 2004). The third stage would refer to what is proposed here, namely starting from the Big Five and amending the model so as to accommodate both similarities and cross-cultural differences that were found in non-Western studies. In our view this would amount to a model in which most factors of the Big Five seem to have the same structure across many cultures, whereas personality factors in the social-relational domain are more context-sensitive. Therefore, we conclude that emic studies in personality can and should have an impact beyond the cultures that they study and that an adequate appreciation of what is universal and what is culture-specific in personality can make psychology more culture-informed.
Allik, J., & McCrae, R. R. (2004). Toward a geography of personality traits patterns of profiles across 36 cultures. Journal of Cross-Cultural Psychology, 35, 13–28.
Berry, J. W., Poortinga, Y. H., Breugelmans, S. M., Chasiotis, A., & Sam, D. L. (2011). Cross-cultural psychology: Research and applications (3rd ed.). Cambridge, England: Cambridge University Press.
Butcher, J. N. (1996). International adaptations of the MMPI-2: Research and clinical applications. Minneapolis, MN: University of Minnesota Press.
Butcher, J. N., Lim, J., & Nezami, E. (1998). Objective study of abnormal personality in cross-cultural settings: The Minnesota Multiphasic Personality Inventory (MMPI-2). Journal of Cross-Cultural Psychology, 29, 189–211.
Caprara, G. V., Steca, P., Zelli, A., & Capanna, C. (2005). A new scale for measuring adults’ prosocialness. Journal of Psychological Assessment, 21, 77–89.
Cheung, F. M., Cheung, S. F., Wada, S., & Zhang, J. (2006). Indigenous measures of personality assessment in Asian countries: A review. Psychological Assessment, 15, 280–289.
Cheung, F. M., & Fetvadjiev, V. H. (2016). Indigenous approaches to testing and assessment. In F. T. L. Leong, D. Bartram, F. M. Cheung, K. F. Geisinger, & D. Iliescu (Eds.), The ITC international handbook of testing and assessment (pp. 333–346). New York, NY: Oxford University Press.
Cheung, F. M., Leung, K., Zhang, J. X., Sun, H. F., Gan, Y. G., Song, W. Z., & Xie, D. (2001). Indigenous Chinese personality constructs: Is the Five-Factor Model complete? Journal of Cross-Cultural Psychology, 32, 407–433.
Cheung, F. M., van de Vijver, F. J. R., & Leong, F. T. L. (2011). Toward a new approach to the assessment of personality in culture. American Psychologist, 66, 593–603.
Choi, S. C., & Choi, S. H. (2001). Cheong: The socioemotional grammar of Koreans. International Journal of Group Tensions, 30, 69–80.
Choi, S. C., & Lee, J. J. (1999). The psychological interior and social–cultural functions of Korean ‘‘Cheong.” Korean Journal of Social & Personality Psychology, 13, 219–234.
Church, A. T. (2010). Current perspectives in the study of personality across cultures. Perspectives on Psychological Science, 5, 441–449.
Dana, R. H. (Ed.). (2000). Handbook of cross-cultural and multicultural personality assessment. New York, NY: Routledge.
Doi, T. L. (1973). The anatomy of dependence. Tokyo, Japan: Kodansha International.
Doke, C. M. (1954). The southern Bantu languages. London, England: Oxford University Press.
Enriquez, V. G. (1978). Kapwa: A core concept in Filipino social psychology. Philippine Social Sciences and Humanities Review, 42(1–4), 100–108.
Enriquez, V. G. (1997). Filipino psychology: Concepts and methods. In H. S. R. Kao & D. Sinha (Eds.), Asian perspectives on psychology: Cross-cultural research and methodology series (Vol. 19, pp. 40–53). Thousand Oaks, CA: Sage.
Fetvadjiev, V. H., Meiring, D., van de Vijver, F. J. R., Nel, J. A., & Hill, C. (2015). The South African Personality Inventory (SAPI): A culture-informed instrument for the country’s main ethnocultural groups. Psychological Assessment, 27, 827–837.
Foxcroft, C., & Roodt, G. (2013). An introduction to psychological assessment in the South African context (4th ed.). Cape Town, South Africa: Oxford University Press Southern Africa.
Georgas, J., Weiss, L., van de Vijver, F. J. R., & Saklofske, D. H. (Eds.). (2003). Cultures and children’s intelligence: A cross-cultural analysis of the WISC-III. New York, NY: Academic Press.
Goldberg, L. R. (1990). An alternative “description of personality”: The Big-Five factor structure. Journal of Personality and Social Psychology, 59, 1216–1229.
Goldberg, L. R., Johnson, J. A., Eber, H. W., Hogan, R., Ashton, M. C., Cloninger, C. R., & Gough, H. C. (2006). The International Personality Item Pool and the future of public-domain personality measures. Journal of Research in Personality, 40, 84–96.
Hendriks, A. A. J., Hofstee, W. K. B., & De Raad, B. (1999). The Five-Factor Personality Inventory (FFPI). Personality and Individual Differences, 27, 307–325.
Hill, C., Nel, J. A., van de Vijver, F. J. R., Meiring, D., Valchev, V. H., Adams, B. G., & De Bruin, G. P. (2013). Developing and testing items for the South African Personality Inventory (SAPI). SA Journal of Industrial Psychology, 39, 1–13.
Ho, D. Y. (1995). Selfhood and identity in Confucianism, Taoism, Buddhism, and Hinduism: Contrasts with the West. Journal for the Theory of Social Behaviour, 25, 115–139.
Hofstede, G. (1980). Culture’s consequences: International differences in work-related values. Beverly Hills, CA: Sage.
Hofstede, G. (2001). Culture’s consequences: International differences in work-related values (2nd ed.). Thousand Oaks, CA: Sage.
Hunter, J. E., & Schmidt, F. L. (1990). Methods of meta-analysis: Correcting error and bias in research findings. Newbury Park, CA: Sage.
Kamwangamalu, N. M. (1999). Ubuntu in South Africa: A sociolinguistic perspective to a pan-African concept. Critical Arts Journal, 13, 24–41.
Katigbak, M. S., Church, A. T., Guanzon-Lapeña, M. A., Carlota, A. J., & del Pilar, G. H. (2002). Are indigenous dimensions culture-specific? Philippine inventories and the Five-Factor Model. Journal of Personality and Social Psychology, 82, 89–101.
Kim, B. S., Soliz, A., Orellana, B., & Alamilla, S. G. (2009). Latino/a values scale development, reliability, and validity. Measurement and Evaluation in Counseling and Development, 42, 71–91.
Kim, U., Yang, K.-S., & Hwang, K.-K. (Eds.). (2006). Indigenous and cultural psychology: Understanding people in context. New York, NY: Springer.
Laher, S. (2013). The NEO-PI-R in South Africa. In S. Laher & K. Cockcroft (Eds.), Psychological assessment in South Africa: Research and applications (pp. 257–269). Johannesburg, South Africa: Wits University Press.
Laher, S., & Cockcroft, K. (2014). Psychological assessment in post-apartheid South Africa: The way forward. South African Journal of Psychology, 44, 303–314.
Lee, K., & Ashton, M. C. (2008). The HEXACO personality factors in the indigenous personality lexicons of English and 11 other languages. Journal of Personality, 76, 1001–1054.
LeVine, R. A. (1973). Patterns of personality in Africa. Ethos, 1, 123–152.
Lorenzo-Seva, U., & ten Berge, J. M. F. (2006). Tucker’s congruence coefficient as a meaningful index of factor similarity. Methodology, 2, 57–64.
Luiz, J. M. (2015). The impact of ethno-linguistic fractionalization on cultural measures: Dynamics, endogeneity and modernization. Journal of International Business Studies, 46, 1080–1098.
McCrae, R. R., & John, O. P. (1992). An introduction to the Five-Factor Model and its applications. Journal of Personality, 60, 175–215.
McCrae, R. R., Terracciano, A., & 78 Members of the Personality Profiles of Cultures Project. (2005). Universal features of personality traits from the observer’s perspective: Data from 50 cultures. Journal of Personality and Social Psychology, 88, 547–561.
Meiring, D. (2007). Bias and equivalence of psychological measures in South Africa. Amsterdam, The Netherlands: Labyrint.
Meiring, D., van de Vijver, F. J. R., & Rothmann, S. (2006). Bias in an adapted version of the 15FQ+ in South Africa. South African Journal of Psychology, 36, 340–356.
Metz, T., & Gaie, J. B. R. (2010). The African ethic of Ubuntu/Botho: Implications for research on morality. Journal of Moral Education, 39, 273–290.
Mosig, Y. D. (2006). Conceptions of the self in Western and Eastern psychology. Journal of Theoretical and Philosophical Psychology, 26, 39–50.
Nel, J. A. (2014, July). An exploration of social desirability within the White Afrikaans group. In F. J. R. van de Vijver (Chair), Psycholexical studies: Bridging emic and etic approaches. Symposium at the 22nd International Congress of the International Association for Cross-Cultural Psychology, Reims, France.
Nel, J. A., Valchev, V. H., Rothmann, S., van de Vijver, F. J. R., Meiring, D., & De Bruin, G. P. (2012). Exploring the personality structure in the 11 languages of South Africa. Journal of Personality, 80, 915–948.
Ng, A., Fan, W., Cheung, F. M., Leong, F. T., & Cheung, S. F. (2011). The CPAI-2 as a culturally relevant personality measure in differentiating among academic major groups. Journal of Career Assessment, 20, 196–207.
Nurse, D., & Philippson, G. (Eds.). (2003). The Bantu languages. London, England: Routledge.
Ortiz, F. A., Church, A. T., Vargas-Flores, J. J., Ibáñez-Reyes, J., Flores-Galaz, M., Iuit-Briceño, J. I., & Escamilla, J. M. (2007). Are indigenous personality dimensions culture-specific? Mexican inventories and the Five-Factor Model. Journal of Research in Personality, 41, 618–649.
Oyserman, D., Coon, H. M., & Kemmelmeier, M. (2002). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128, 3–72.
Peabody, D. (1987). Selecting representative trait adjectives. Journal of Personality and Social Psychology, 52, 59–71.
Ramsay, L. J., Taylor, N., De Bruin, G. P., & Meiring, D. (2008). The Big Five personality factors at work: A South African validation study. In J. Deller (Ed.), Research contributions to personality at work (pp. 99–114). Munich, Germany: Rainer Hampp Verlag.
Republic of South Africa. (2014, January 16). Employment Equity Amendment Act, 2013. Government Gazette, 583 (37238). Retrieved April 6, 2014, from http://www.labour.gov.za/DOL/downloads/legislation/acts/employment-equity/eea_amend2014.pdf
Retief, A. (1992). The cross-cultural utility of the SAPQ: Bias or fruitful differences? South African Journal of Psychology, 17, 202–207.
Saucier, G., Thalmayer, A. G., & Bel-Bahar, T. S. (2014). Human attribute concepts: Relative ubiquity across twelve mutually isolated languages. Journal of Personality and Social Psychology, 107, 199–216.
Saucier, G., Thalmayer, A. G., Payne, D. L., Carlson, R., Sanogo, L., Ole-Kotikash, L., … Zhou, X. (2014). A basic bivariate structure of personality attributes evident across nine languages. Journal of Personality, 82, 1–14.
Schmitt, D. P., Allik, J., McCrae, R. R., Benet-Martínez, V., Alcalay, L., Ault, L., … Zupanèiè, A. (2007). The geographic distribution of Big Five personality traits: Patterns and profiles of human self-description across 56 nations. Journal of Cross-Cultural Psychology, 38, 173–212.
Serpell, R. (2011). Social responsibility as a dimension of intelligence, and as an educational goal: Insights from programmatic research in an African society. Child Development Perspectives, 5, 126–133.
Statistics South Africa. (2012). Census 2011: Statistical release (revised). Retrieved September 20, 2014, from http://www.statssa.gov.za
Taylor, N., & De Bruin, G. (2005). Basic Traits Inventory: Technical manual. Johannesburg, South Africa: Jopie van Rooyen & Partners SA.
Triandis, H. C. (1995). Individualism and collectivism. Boulder, CO: Westview.
Triandis, H. C., Marin, G., Lisansky, J., & Betancourt, H. (1984). Simpatía as a cultural script of Hispanics. Journal of Personality and Social Psychology, 47, 1363–1375.
Tucker, L. R. (1951). A method for synthesis of factor analysis studies (Personnel Research Section Report No. 984). Washington, DC: Department of the Army.
Valchev, V. H., Nel, J. A., van de Vijver, F. J. R., Meiring, D., De Bruin, G. P., & Rothmann, S. (2013). Similarities and differences in implicit personality concepts across ethno-cultural groups in South Africa. Journal of Cross-Cultural Psychology, 44, 365–388.
Valchev, V. H., van de Vijver, F. J. R., Meiring, D., Nel, J. A., Laher, S., Hill, C., & Adams, B. (2014). Beyond Agreeableness: Social-relational personality concepts from an indigenous and cross-cultural perspective. Journal of Research in Personality, 48, 17–32.
Valchev, V. H., van de Vijver, F. J. R., Nel, J. A., Rothmann, S., & Meiring, D. (2013). The use of traits and contextual information in free personality descriptions across ethnocultural groups in South Africa. Journal of Personality and Social Psychology, 104, 1077–1091.
Valchev, V. H., van de Vijver, F. J. R., Nel, J. A., Rothmann, S., Meiring, D., & De Bruin, G. P. (2011). Implicit personality conceptions of the Nguni cultural-linguistic groups of South Africa. Cross-Cultural Research, 45, 235–266.
van de Vijver, F. J. R. (2013). Contributions of internationalization to psychology: Toward a global and inclusive discipline. American Psychologist, 68, 761–770.
van de Vijver, F. J. R., & Leung, K. (1997). Methods and data analysis for cross-cultural research. Thousand Oaks, CA: Sage.
van Eeden, R., Taylor, N., & Prinsloo, C. H. (2013). The Sixteen Personality Factor Questionnaire in South Africa. In S. Laher & K. Cockcroft (Eds.), Psychological assessment in South Africa: Research and applications (pp. 203–217). Johannesburg, South Africa: Wits University Press.
Yang, K. S., & Ho, D. Y. (1988). The role of yuan in Chinese social life: A conceptual and empirical analysis. In A. C. Paranjpe, D. Y. F. Ho, & R. W. Rieber (Eds.), Asian contributions to psychology (pp. 263–281). New York, NY: Praeger.
Zecca, G., Verardi, S., Antonietti, J.-P., Dahourou, D., Adjahouisso, M., Ah-Kion, J., … Rossier, J. (2013). African cultures and the Five-Factor Model of personality: Evidence for a specific pan-African structure and profile? Journal of Cross-Cultural Psychology, 44, 684–700.
Zeinoun, P., Daouk-Oyry, L., & van de Vijver, F. J. R. (2014, July). Understanding “Arab personality”: Methodology and initial findings of a study of personality descriptors from Syria, Lebanon, Jordan and Palestine. In F. J. R. van de Vijver (Chair), Psycholexical studies: Bridging emic and etic approaches. Symposium at the 22nd International Congress of the International Association for Cross-Cultural Psychology, Reims, France.
Zhang, J., & Bond, M. H. (1998). Personality and filial piety among college students in two Chinese societies the added value of indigenous constructs. Journal of Cross-Cultural Psychology, 29, 402–417.