Exploring the utility of the CPAI-2 in a South African sample: Implications for the FFM

Personality and Individual Differences 81 (2015) 67–75

Contents lists available at ScienceDirect

Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid

Exploring the utility of the CPAI-2 in a South African sample: Implications for the FFM q S. Laher Department of Psychology, University of the Witwatersrand, South Africa

a r t i c l e

i n f o

Article history: Received 28 April 2014 Received in revised form 27 November 2014 Accepted 4 December 2014 Available online 22 December 2014 Keywords: CPAI-2 Culture Five-Factor model Personality Personality assessment

a b s t r a c t The Cross-Cultural Personality Assessment Inventory-2 (CPAI-2) was developed as an alternative instrument to measure personality in Asian cultures based on the argument that available instruments are largely based on the Five Factor Model of personality and may not adequately assess personality in Asian cultures. Research on the CPAI-2 was extended to non-Asian particularly Western cultures but not to African cultures. In this study a sample of 425 South African university students completed both the CPAI-2 and the NEO-PI-R to consider the utility of the CPAI-2 and the four factor model in which the instrument is located in an African context. The results were did not provide conclusive support for the four factor model as conceptualised in the CPAI-2. Support for the five factors of the FFM was evident but the results from both the four and five factor models indicated empirical support for a social relational aspect of personality. Based on these results, this paper argues for the necessity of incorporating social relational elements into the existing FFM and reflects on work done in the South African context which can inform this direction. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction South Africa is located at the tip of Africa and has a population of approximately 52 million people of diverse origins, cultures, languages, and religions (Census, 2011). The country has 11 official languages which belong to two language families: the Germanic languages (Afrikaans and English) and the Bantu languages which are further subdivided into the Sotho-Tswana group of languages (Northern Sotho, Southern Sotho, and Tswana), the Nguni group of languages (Ndebele, Swati, Xhosa and Zulu), Venda and Tsonga. According to the 2011 Census, the four most spoken first languages are Zulu (22.7%), Xhosa (16.0%), Afrikaans (13.5%) and English (9.6%). English is the language used for commerce and science even though it is not the first language for majority of the country. A legacy of South Africa’s apartheid past is the classification of the population into a Black (of African descent; 79.2%), White (of European descent; 8.9%), Coloured (of mixed race; 8.9%) or Indian/Asian group (of Asian descent; 2.5%). Apartheid’s policy of separate development ensured that these groups were not given much opportunity to mix in social and occupational spheres. This has resulted in a split in the population with Black individuals demonstrating more collectivistic orientations and White people more individualistic orientations with Indian and Coloured people being

q

This article is a Special issue article – ‘‘Young researcher award 2014’’. E-mail address: [email protected]

http://dx.doi.org/10.1016/j.paid.2014.12.010 0191-8869/Ó 2014 Elsevier Ltd. All rights reserved.

somewhere in the middle (Allik & McCrae, 2004; Valchev et al., 2014). This linguistic and ethnic diversity provides an interesting space within which to explore personality more so since the psychic unity and universal applicability of the Five Factor Model is often debated in African and Asian contexts (see Cheung et al., 2008; Imperio, Church, Katigbak, & Reyes, 2008; Laher, 2013; McCrae, Terracciano, & 79 members of the Personality Profiles of Cultures Project, 2005; Valchev et al., 2014). African research on the FFM is also limited as evidenced in the studies by McCrae et al. (2005). Laher (2013) argues that the FFM has not achieved much replicability in African contexts. McCrae et al. (2005) argue that this may be due to the lack of research in African countries, the poorer data quality and issues of translation but they equally acknowledge that this could be attributed to different ways of perceiving the world. Using this rationale as a starting point, this study investigated the utility of the CPAI-2, an instrument with a different but similar model to the FFM, in a South African sample. It also conducted a joint factor analysis between CPAI-2 and NEO-PI-R responses to comment further on the utility of the FFM. Furthermore, no studies have been conducted using the CPAI-2 in its entirety in an African context. 1.1. Contextualising the CPAI-2 Cheung et al. (2001) argued that while the FFM was recoverable using the NEO-PI-R in a Chinese setting, it did not provide a

68

S. Laher / Personality and Individual Differences 81 (2015) 67–75

comprehensive description of personality. Most notably what was lacking were the interdependent aspects of Chinese culture and by extension Chinese personality. Chinese personality is viewed in the context of the cultural-ecological model and is characterised by collectivist orientation, other orientation, relationship orientation and authoritarian orientation (Cheung et al., 2001). Thus Cheung and colleagues began work on developing an indigenous personality inventory, the culmination of which is the current Crosscultural Personality Assessment Inventory-2 (CPAI-2). The CPAI-2 is based on a four factor model (Social Potency, Dependability, Accommodation and Interpersonal Relatedness) and represents an alternate model of assessing personality that is congruent with the dominant FFM but adds to the FFM by including a more relational dimension to the assessment of personality. Cheung et al. (2008) reported Cronbach alpha coefficients of .86, .86, .76 and .70 were reported for the factor scales of Social Potency, Dependability, Accommodation and Interpersonal Relatedness in the Chinese normative sample. Subscale coefficients ranged from .51 to .80 as evidenced in Table 1. Congruence coefficients ranged between .94 and .98 for the four factor solution across the seven regions. Born and Jooren (2009 cited in Cheung, Cheung, & Fan, 2013) used a Dutch translation of the CPAI-2 with a sample of 363 Dutch college students and found support for a five factor solution. The four factor structure was not replicated with the CPAI-2 but Interpersonal Relatedness remained as a separate factor. Iliescu and Ion (2009) found a 6 factor solution with a Romanian version of the CPAI-2 with 500 adult participants. Iliescu and Ion (2009 cited in Cheung et al., 2013) reported that some of the original CPAI-2 factors were realigned but a separate Interpersonal Relatedness factor was clearly defined. Cheung, Van de Vijver, and Leong (2011) reported that the congruence of the four factor structure

was generally confirmed in cross-cultural samples but more so amongst Asian and Asian American samples. 1.2. Interpersonal Relatedness in the South African context Studies in South Africa over the last five years or so have consistently provided evidence for the social relational aspect of personality (see Valchev, Van de Vijver, Nel, Rothmann, & Meiring, 2013). Of particular note are the findings of Valchev et al. (2012) who conducted semi-structured interviews with 1216 participants from the 11 official language groups in South Africa. Analysis revealed 9 broad personality clusters, namely, Conscientiousness, Emotional Stability, Extraversion, Facilitating, Integrity, Intellect, Openness, Relationship Harmony and Soft-Heartedness. Valchev et al. (2013) analysed these responses further and found that Black individuals were more likely to use less traits and more contextualization in their responses which is consistent with individuals from collectivist cultures and White individuals were more likely to use more traits and less contextualization – a pattern that is more consistent with individualist cultures. This group of researchers used these descriptions and are in the process of developing the South African Personality Inventory (SAPI; Valchev et al., 2014). Nel et al. (2012) identified a social relational cluster within the 9 SAPI clusters made up of the Facilitating, Integrity, Relationship Harmony and Soft-Heartedness clusters. Branco e Silva and Laher (2012) explored the CPAI-2 Interpersonal Relatedness factor in relation to the SAPI social relational cluster in a sample of 489 South African students using a joint factor analysis. Results revealed 4 factors. The first factor had loadings on the Integrity, and Relationship Harmony clusters of the SAPI and Social Sensitivity from the CPAI-2 Interpersonal Relatedness factor. Factor 2 had loadings from the Harmony cluster of the SAPI and the Harmony

Table 1 Descriptive statistics and internal reliability coefficients for the CPAI-2. Scale

a⁄

a

.44 .98 1.39 .80 .45 .76 .25 .66 .14

.86 .69 .68 .62 .72 .58 .65 .70 .60

.90 .72 .60 .58 .68 .41 .68 .79 .68

66 10 10 16 11 10 10 11 10 10

.34 .12 .31 .75 .17 .24 .17 .00 .67 .15

.86 .70 .64 .80 .51 .62 .60 .59 .62 .66

.75 .70 .69 .83 .37 .68 .66 .57 .54 .62

7 0 0 1 1 0

40 9 10 10 10 10

.74 .88 .40 1.57 .05 .68

.76 .69 .66 .66 .60 .69

.76 .69 .62 .74 .52 .66

9 0 3 3 0 3 0

61 14 12 11 10 14 10

.28 .44 .25 .57 .04 .70 .16

.70 .65 .52 .62 .59 .51 .57

.74 .69 .39 .45 .60 .56 .37

Mean

SD

Min

Max

80 10 10 10 10 10 10 10 10

55.39 7.37 8.54 7.89 6.46 7.06 6.13 6.56 5.82

11.36 2.19 1.59 1.82 2.29 1.79 2.36 2.69 2.42

20 0 2 2 0 1 0 0 0

77 10 10 10 10 10 10 10 10

101 10 10 18 12 10 10 11 10 10

49.90 5.15 4.58 4.55 7.25 5.39 5.78 5.34 7.11 5.15

6.64 2.41 2.11 3.89 1.91 2.41 2.37 2.28 1.81 2.19

25 0 0 0 2 0 0 0 0 0

Accommodation Defensiveness Graciousness vs Meanness Interpersonal Tolerance Self vs Social Orientation Veraciousness vs Slickness

50 10 10 10 10 10

29.40 2.45 6.29 8.43 5.63 6.77

4.18 2.11 2.12 1.86 1.88 2.21

Interpersonal Relatedness Traditionalism vs Modernity Relationship Orientation Social Sensitivity Discipline Harmony Thrift vs Extravagance

67 15 12 10 10 12 8

41.26 4.92 7.95 8.31 5.41 10.36 5.22

7.41 2.74 1.82 1.68 2.35 2.12 1.82

Social Potency Novelty Diversity Divergent Thinking Leadership Logical vs Affective Orientation Aesthetics Extraversion vs Introversion Enterprise Dependability Responsibility Emotionality Inferiority vs Self-Acceptance Practical Mindedness Optimism vs Pessimism Meticulousness Face Internal vs External LOC Family Orientation

No. of items

a⁄ Cronbach Alpha coefficients from Cheung et al. (2008).

Skewness

S. Laher / Personality and Individual Differences 81 (2015) 67–75

subscale from the CPAI-2 Interpersonal Relatedness factor. The third factor had loadings from Traditionalism vs Modernity, Discipline, Thrift vs Extravagance and Ren Qing from the CPAI-2 Interpersonal Relatedness factor. The final factor represented the validity and social desirability scales. Mean differences across language and ethnic groupings were also explored. Second language English speakers and Blacks (a combination of African, Coloured and Indian groups) scored significantly higher than White English first language speakers on Interpersonal Relatedness, Traditionalism vs Modernity and Facilitating but significantly lower on Integrity. The mirroring of results between language and ethnic groups suggests the presence of some commonality between these groups. In South Africa, this is commonality is defined along collectivist and individualist dimensions. As is evident from the literature, research on the CPAI-2 has to date only been conducted on Asian and American samples. However the instrument and the personality model associated with it offers an opportunity for an alternative exploration of personality that encompasses a more collectivist orientation. South Africa provides an excellent context in which to explore the CPAI-2 and the four factor model with a view to contributing towards a more nuanced exploration of the psychic unity presented by the five factors in the FFM by virtue of its diverse population. It could be argued that such heterogeneity may be problematic in a sample. This study used a student sample to address this issue. By virtue of being students at the same university that requires a certain proficiency in English for admission, this study maintains homogeneity to a large extent. Van Dyk and De Kock (2004) argue, that student populations in South Africa tend to be more individualistic in nature, due in part to their shared exposure to similar education. The use of this sample ensures a fairly homogenous sample but one that is congruent with the ethnic and linguistic diversity amongst South Africa’s varied cultures and hence would be able to identify with both the CPAI-2 and the NEO-PI-R. Thus this study aims to: (1) Explore whether a four factor solution as proposed by Cheung et al. (2008) is replicable in the South African context (2) Explore the CPAI-2 with the NEO-PI-R in a joint factor analysis to inform the debate on the universality of the five factor structure of personality. 2. Methods 2.1. Sample A non-probability, convenience sample of 425 students at the University of the Witwatersrand (Johannesburg, South Africa) voluntarily completed the questionnaire. The students ranged in age from 17 to 50 years, with a mean age of 21.3 years (SD = 3.81). 28.71% of the sample were male (n = 122) and 71.29% female (n = 303). In terms of ethnicity, 28.47% of the sample classified themselves as Black (n = 120), 6.35% as Coloured (n = 27), 17.88% as Indian (n = 76), 44.47% as White (n = 189), 2.12% (n = 9) as Asian (e.g. Chinese, Taiwanese) and .71% (n = 3) choosing the other option but not specifying their grouping. 68.47% of the sample reported English as their home language (n = 302), 2.35% spoke Afrikaans as their home language (n = 10) while 24% (n = 102) of the sample spoke an African language and 5% (n = 22) selected the Other option, specifying Chinese, Shona, Lebanese or some other Asian, African or Middle Eastern language as their home language. 2.2. Instruments A questionnaire incorporating a demographics section, the CPAI-2 and the NEO-PI-R was used. The questionnaire requested

69

demographic information, namely, age, gender, ethnicity and home language. 2.2.1. The CPAI-2 The CPAI-2 Form B (English version) is a self-report instrument consisting of 341 items, which are answered in a dichotomous (true/false) response format (Cheung et al., 2008). The CPAI-2 takes approximately 60–90 min to complete. It is suitable for individuals in the age range of 18–70 with at least a Grade 6 level of education (Cheung, Cheung, Leung, Ward, & Leong, 2003). The CPAI-2 has 28 personality scales, which factor into 4 factors, namely Social Potency/Expansiveness (8 scales), Dependability (9 scales), Accommodation (5 scales) and Interpersonal Relatedness (6 scales). The English version of the original CPAI was validated in a sample of 675 Singaporean Chinese and 144 Caucasian American students at a Midwestern university. Similar factor structures were obtained in both samples. These were also similar to those obtained from the standardisation sample that completed the Chinese version of the CPAI (Cheung et al., 2003). Cheung, Cheung, Howard, and Lim (2006) cite evidence for the construct validity of the CPAI-2, particularly in terms of its correlations with the NEO-PI-R. Internal consistency reliability coefficients obtained for the CPAI-2 scales in the current study were comparable to those found in the normative sample for the most part (see Table 1). In this study, internal consistency reliability coefficients of .90, .75, .76 and .74 were found for the Social Potency, Dependability, Accommodation and Interpersonal Relatedness factors, respectively. Dependability had a lower internal consistency reliability coefficient than the normative sample. Subscale reliability coefficients were more variable. Poor reliability coefficients were found for the Logical vs Affective Orientation subscale (Social Potency factor; a = .41), the Practical Mindedness subscale (Dependability factor; a = .37), and the Interpersonal Relatedness subscales of Relationship Orientation subscale (.39), the Social Sensitivity subscale (.45) and the Thrift vs Extravagance subscale (.37). With the exception of the Social Sensitivity subscale, which had an alpha coefficient of .62 in the normative sample, all four of the other subscales demonstrated lower reliabilities in the normative sample as well (.58 for Logical vs Affective Orientation, .51 for Practical Mindedness, .52 for Relationship Orientation and .57 for Thrift vs Extravagance in the normative sample). The Interpersonal Relatedness subscales only had the Discipline (.60) and Traditionalism vs Modernity (.69) evidencing coefficients at or above 0.60. Other ubcale which had coefficients below .60 include the Divergent Thinking subscale (Social Potency; a = .58), Internal vs External Locus of Control (Dependability; a = .54) and Face (Dependability; a = .57) and Self vs Social Orientation (Accommodation; a = .52). 2.2.2. The NEO-PI-R The NEO-PI-R is a self-report instrument consisting of 240 items and takes approximately 45 min to complete. It measures each of the five factors postulated in the FFM, namely Neuroticism, Extraversion, Openness to Experience, Conscientiousness. Each of the domains is measured by 48 items, which are subdivided into 6 sets of 8 items called facets. NEO-PI-R items are answered on a 5-point scale ranging from ‘Strongly Agree’ (4) to ‘Strongly Disagree’ (0) (Costa & McCrae, 1992). In this study, internal consistency reliability coefficients of .91, .89, .87, .87 and .92 were found for the NEOPI-R domain scales of Neuroticism, Extraversion, Openness to Experience, Agreeableness and Conscientiousness, respectively. All facet reliability coefficients exceeded .60 except for Impulsiveness (Neuroticism domain, .59), Actions (Openness domain, .55) and Tender-Mindedness (Agreeableness domain, .50). A factor analysis on the NEO-PI-R data in this study supported a five factor structure. There was also sufficient agreement between the

70

S. Laher / Personality and Individual Differences 81 (2015) 67–75

factor matrix from this study and that of the normative sample (Costa & McCrae, 1992). 2.3. Procedure Ethical clearance was received from the Humanities Ethics Committee at the University of the Witwatersrand (Protocol No. H60417). Permission was received from Prof Fanny Cheung to use the CPAI-2 and from Psychological Assessment Resources Inc. to use the NEO-PI-R in the study. Students were approached during lecture times and briefed about the study. They completed the questionnaires voluntarily and returned them to a sealed box in the lecture room. The responses were entered and scored on computer and the relevant statistical analyses were conducted using the SAS Enterprise Guide (Version 4.0) computer program. 2.4. Data analysis Descriptive statistics were computed for the CPAI-2 scales. Internal consistency reliability for the scales was determined using the Cronbach alpha (CA) reliability coefficient. Exploratory factor analysis using varimax rotation was conducted to determine the construct validity of the scales in the CPAI-2. Varimax rotation was used as this was the method employed by Cheung et al. (2008) for the CPAI-2 as well as in studies employing the

NEO-PI-R (McCrae et al., 2005). In accordance with Cheung et al. (2008), the factor structure of the CPAI-2 was first examined by itself and then by considering a joint factor analysis of the CPAI-2 and the NEO-PI-R. The extraction of factors was guided by both theory and empirical evidence. Hence the Guttman-Kaiser eigenvalue greater-than-one rule (K1 rule), Cattell’s scree test, parallel analysis (PA), and Velicer’s minimum average partial (MAP) were used. In examining the replicability of the four factor structure of the CPAI-2, Procrustes rotation and factor congruence coefficients were using the SAS programme from McCrae, Zonderman, Bond, and Paunonen (1996). 3. Results Descriptive statistics, namely means, standard deviations, minimum and maximum values and skewness coefficients, for the CPAI-2 are presented in Table 1. All the scales were normally distributed except for Diversity (Social Potency; skewness = 1.39) and Interpersonal Tolerance (Accommodation; skewness = 1.57). 3.1. Replicating the structure of the CPAI-2 A four factor solution (see Table 2) was examined to test the model proposed by Cheung et al. (2008). At an empirical level, an examination of Cattell’s scree plot suggested seven factors. Using

Table 2 Four Factor solution for the CPAI-2 using varimax and Procrustes rotation. Varimax rotated Scale

Factor 2

Factor 3

.34 .15 .08 .40 .20

.72 .69 .80 .59 .56

.08 .23 .75

Dependability Responsibility Emotionality Inferiority vs Self-Acceptance Practical Mindedness Optimism Meticulousness Face

Factor 1

Factor 2

Factor 3

Factor 4

Factor 1

Factor 2

Factor 3

Factor 4

Cong

.16 .24 .04 .39 .04

.01 .03 .08 .06 .31

.79 .78 .74 .70 .63

.09 .04 .02 .10 .30

.12 .15 .08 .29 .00

.12 .07 .12 .05 .15

.78 .67 .78 .68 .56

.20 .08 .05 .28 .28

0.08 0.30 0.09 0.32 0.06

.10 .07 .17 .14 .23

.99** .95** .97** .95** .99**

.61 .45 .42

.08 .06 .10

.09 .24 .00

.58 .43 .52

.05 .14 .53

.11 .04 .03

.12 .04 .23

.58 .49 .54

.16 .03 .58

0.15 0.16 0.03

.09 .22 .34

.98** .90* .99**

.22 .64 .78 .31 .73 .02 .56

.17 .04 .22 .06 .27 .03 .09

.02 .20 .17 .29 .03 .13 .26

.66 .27 .11 .49 .16 .66 .07

.09 .01 .35 .02 .30 .04 .05

.68 .67 .70 .62 .66 .50 .50

.03 .30 .28 .28 .22 .00 .28

.35 .10 .10 .29 .07 .34 .19

.15 .02 .31 .04 .36 .09 .15

.53 .70 .71 .55 .68 .39 .45

0.12 0.15 0.19 0.18 0.05 0.04 0.30

.45 .09 .25 .30 .20 .54 .28

Internal/External LOC Family Orientation

.48 .28

.21 .10

.22 .39

.06 .30

.17 .02

.43 .46

.33 .38

.17 .20

.26 .12

.37 .44

0.27 0.31

.22 .16

.95** .95** .98** .99** .96** .92* 192 .93* .97** .97**

Accommodation Defensiveness Graciousness vs Meanness Interpersonal Tolerance Self vs Social Orientation Veraciousness vs Slickness

.31 .47 .24 .04 .04

.17 .10 .44 .12 .02

.65 .62 .60 .28 .75

.21 .07 .07 .23 .12

.08 .06 .40 .15 .09

.30 .39 .20 .14 .38

.72 .70 .53 .42 .47

.18 .05 .19 .12 .20

.16 .11 .41 .12 .12

.20 .41 .19 .05 .11

0.71 0.65 0.64 0.32 0.69

.16 .15 .02 .17 .28

.95** .99** .95** .89* .88*

.09 .43 .15 .21 .04 .37

.32 .13 .54 .12 .30 .19

.27 .04 .30 .23 .50 .00

.51 .24 .16 .69 .35 .36

.30 .08 .39 .05 .08 .06

.09 .03 .03 .06 .28 .08

.28 .05 .01 .37 .36 .10

.46 .56 .54 .68 .47 .52

.34 .03 .46 .18 .20 .10 .97**

.20 .24 .05 .18 .18 .13 .95**

0.42 0.03 0.29 0.41 0.42 0.09 .94**

.34 .45 .37 .59 .46 .52 .93*

.95** .90* .86 .97** .97** .88*

14.58

14.57

10.61

9.66

Interpersonal Relatedness Traditionalism vs Modernity Relationship Orientation Social Sensitivity Discipline Harmony Thrift vs Extravagance Factor Cong % Variance Explained

**

Procrustes rotated

Factor 4

Social Potency Novelty Diversity Divergent Thinking Leadership Logical vs Affective Orientation Aesthetics Extraversion vs Introversion Enterprise

*

Cheung et al. (2008)

Factor 1

p < 0.05. p < 0.01.

71

S. Laher / Personality and Individual Differences 81 (2015) 67–75

the eigenvalues greater-than-one rule, seven factors were also indicated. PA and Velicer’s MAP technique both indicated the extraction of five factors. Given this, together with the fact that the FFM also argues for five factors, a five factor solution was extracted. These results are presented in Table 3. The four factor solution in this study explains 49.40% of the variance. Despite the varied loadings between the 4 factor solution found in this study and the solution obtained by Cheung et al. (2008), the congruence coefficients demonstrate adequate agreement between the scales and subscales of the CPAI-2 for the two samples with all congruence coefficients achieving significance and exceeding .88. Social Sensitivity is the only subscale which had a nonsignificant congruence coefficient of .86 but according to McCrae et al. (2005) this is adequate. In total the five factor solution explained 55.25% of the variance (see Table 3). The loadings obtained on Factor 1 are congruent with the Social Potency factor found in Cheung et al. (2008) as well as the Openness to Experience and Extraversion factors in the FFM. Factor 2 did not concur with any of the factors proposed by Cheung et al. (2008) but is representative of the Neuroticism factor of the FFM. The smaller loadings on Factor 2 also suggest a combination with some aspects of Extraversion and Openness to Experience. The loadings on Factor 3 were congruent with the Accommodation factor proposed by Cheung et al. (2008) and the Agreeableness factor of the FFM. Factor 4 loadings were not representative of any of the four factors postulated by Cheung et al. (2008) but the pattern of loadings is congruent with the Conscientiousness factor of the FFM. Factor 4 also has an Introversion

dimension, while Factor 1 incorporates an Extraversion dimension. The loadings on Factor 5 are somewhat congruent with the Interpersonal Relatedness dimension proposed by Cheung et al. (2008). 3.2. Joint factor analysis: CPAI and NEO-PI-R Theoretically the NEO-PI-R located within the FFM suggests the extraction of five factors. Cheung and her colleagues argue for four factors in the CPAI-2 (Cheung et al., 2003, 2008). If these are indeed separate factors for both instruments, the extraction of nine factors appears warranted. However, Cheung et al. (2008) argue that the FFM does not subsume their Interpersonal Relatedness factor but that their model does not adequately cover Openness, in which case six factors are warranted. At an empirical level, examination of Cattell’s scree plot indicated the possible extraction of eight factors. Using the eigenvalues greater-than-one rule, the extraction of 13 factors was warranted. However both Velicer’s MAP and the PA techniques provided support for the extraction of eight factors. Based on these arguments, an eight, six and five factor solution were extracted for the joint factor analysis of the NEO-PI-R and the CPAI-2. From the eight factor solution in Table 4 it is evident that Factor 1 can be described as the Neuroticism factor while Factor 2 describes the Conscientiousness factor. Factor 3 can be described as an Extraversion factor and Factor 4 represents the Agreeableness domain of the FFM. Factor 5 can be identified as the Openness to Experience domain as defined by the FFM. Factor 6 is not clearly defined. It has four of the five Accommodation subscales loading

Table 3 Five factor solution for the CPAI-2 using varimax rotation. Scale

Factor 1

Factor 2

Factor 3

Social Potency Novelty Diversity Divergent Thinking Leadership Logical vs Affective Orientation Aesthetics Extraversion vs Introversion Enterprise

.69 .68 .80 .56 .56 .63 .40 .38

.43 .16 .09 .53 .19 .13 .38 .77

.10 .27 .07 .32 .01 .07 .11 .02

.06 .03 .13 .02 .38 .00 .40 .05

.01 .03 .07 .06 .13 .13 .22 .18

Dependability Responsibility Emotionality Inferiority vs Self-Acceptance Practical Mindedness Optimism vs Pessimism Meticulousness Face Internal vs External LOC Family Orientation

.15 .11 .18 .03 .21 .05 .07 .22 .15

.24 .68 .70 .31 .77 .04 .38 .28 .33

.06 .28 .24 .33 .11 .15 .28 .24 .43

.69 .16 .13 .48 .08 .69 .03 .07 .19

.02 .01 .35 .03 .10 .06 .47 .49 .13

Accommodation Defensiveness Graciousness vs Meanness Interpersonal Tolerance Self vs Social Orientation Veraciousness vs Slickness

.16 .07 .41 .18 .03

.15 .36 .20 .19 .13

.66 .65 .63 .31 .74

.16 .09 .09 .45 .14

.32 .22 .09 .36 .05

Interpersonal Relatedness Traditionalism vs Modernity Relationship Orientation Social Sensitivity Discipline Harmony Thrift vs Extravagance

.37 .09 .50 .14 .26 .20

.17 .19 .03 .02 .06 .31

.23 .05 .33 .21 .53 .02

.30 .07 .01 .60 .25 .39

.55 .61 .41 .41 .28 .19

13.76 6.33 6.20 1.50 1.58

13.33 2.71 2.92 1.43 1.48

11.44 2.61 2.56 1.38 1.42

9.04 2.18 2.18 1.33 1.37

7.66 1.64 1.63 1.29 1.32

% Variance explained Actual Eigenvalue Random Eigenvalue MAP Random Eigenvalue PA Mean Random eigenvalue PA 95th percentile

Factor 4

Factor 5

Factor 6

Factor 7

1.14 1.09 1.25 1.28

1.03 .98 1.21 1.25

72

S. Laher / Personality and Individual Differences 81 (2015) 67–75

Table 4 Joint factor analysis of the NEO-PI-R and the CPAI-2–8, 6 and 5 factor solutions. Eight factor solution Scale

1

2

3

Six factor solution 4

5

6

7

8

1

2

3

Five factor solution 4

5

6

1

2

3

4

5

Neuroticism Anxiety Angry Hostility Depression Self-Consciousness Impulsiveness Vulnerability

.74 .60 .81 .70 .44 .66

.03 .04 .15 .09 .23 .29

.09 .07 .12 .18 .27 .04

.01 .27 .08 .12 .10 .16

.16 .11 .06 .06 .06 .00

.06 .02 .09 .24 .20 .13

.12 .06 .02 .07 .02 .28

.02 .47 .10 .10 .14 .11

.75 .61 .79 .67 .45 .71

.01 .08 .14 .01 .23 .29

.07 .06 .04 .02 .08 .17

.12 .00 .15 .23 .26 .01

.00 .48 .06 .14 .13 .11

.11 .00 .14 .28 .17 .14

.72 .50 .78 .75 .30 .69

.02 .01 .17 .01 .34 .34

.11 .04 .07 .03 .08 .12

.17 .14 .18 .20 .13 .03

.00 .46 .06 .11 .12 .09

Extraversion Warmth Gregariousness Assertiveness Activity Excitement-Seeking Positive Emotions

.06 .09 .32 .21 .04 .27

.02 .13 .32 .32 .23 .04

.74 .81 .43 .38 .52 .59

.29 .11 .31 .01 .23 .16

.28 .00 .03 .21 .06 .34

.04 .01 .05 .07 .07 .01

.01 .11 .20 .17 .28 .07

.06 .07 .28 .45 .09 .09

.00 .11 .30 .12 .01 .19

.02 .16 .32 .34 .24 .01

.17 .01 .08 .24 .17 .26

.72 .79 .51 .51 .55 .62

.37 .02 .38 .17 .21 .18

.04 .04 .07 .02 .10 .02

.00 .10 .35 .14 .04 .17

.04 .19 .28 .32 .23 .02

.21 .02 .09 .24 .15 .25

.74 .77 .44 .47 .53 .65

.30 .06 .41 .22 .28 .11

Openness to Experience Fantasy Aesthetics Feelings Actions Ideas Values

.07 .11 .18 .32 .19 .03

.25 .10 .14 .08 .12 .00

.07 .07 .21 .19 .05 .14

.01 .13 .05 .15 .06 .03

.64 .74 .64 .17 .61 .40

.01 .10 .16 .18 .07 .51

.02 .23 .00 .43 .37 .06

.12 .05 .16 .11 .02 .03

.12 .16 .29 .31 .17 .09

.31 .06 .10 .05 .09 .03

.53 .73 .51 .35 .72 .36

.05 .09 .24 .27 .01 .15

.10 .19 .03 .08 .02 .00

.01 .10 .17 .23 .08 .49

.17 .17 .21 .35 .13 .11

.23 .12 .08 .04 .19 .12

.48 .72 .59 .39 .65 .55

.10 .13 .19 .27 .06 .03

.07 .15 .01 .07 .05 .05

Agreeableness Trust Straightforwardness Altruism Compliance Modesty Tender-Mindedness

.32 .07 .04 .10 .31 .04

.06 .13 .21 .05 .10 .08

.39 .02 .37 .17 .12 .22

.48 .74 .53 .53 .70 .55

.15 .11 .25 .19 .04 .19

.08 .07 .14 .17 .07 .17

.10 .05 .03 .14 .13 .04

.14 .04 .16 .51 .11 .03

.25 .04 .09 .11 .31 .09

.08 .18 .19 .09 .04 .09

.02 .09 .15 .07 .02 .10

.36 .01 .33 .27 .14 .21

.54 .65 .59 .71 .63 .53

.12 .14 .13 .19 .09 .12

.24 .06 .17 .06 .28 .17

.08 .16 .22 .03 .06 .12

.05 .04 .10 .06 .08 .06

.41 .03 .43 .09 .10 .30

.51 .66 .53 .69 .65 .48

.28 .07 .10 .11

.68 .77 .68 .78

.15 .04 .02 .11

.13 .02 .26 .02

.18 .10 .12 .15

.01 .03 .03 .01

.14 .06 .08 .10

.02 .08 .01 .24

.27 .05 .08 .06

.66 .75 .68 .78

.23 .09 .14 .18

.19 .06 .05 .18

.10 .01 .24 .11

.02 .01 .04 .02

.25 .02 .08 .09

.67 .71 .67 .75

.18 .08 .14 .18

.20 .08 .07 .15

.13 .03 .23 .13

.24 .05

.74 .62

.07 .21

.04 .10

.01 .02

.07 .12

.11 .05

.09 .27

.23 .09

.75 .60

.06 .01

.12 .25

.00 .19

.08 .16

.26 .00

.71 .64

.07 .11

.10 .18

.00 .19

.39 .15 .11 .37 .21

.03 .05 .14 .17 .30

.26 .19 .10 .31 .11

.03 .14 .04 .29 .09

.22 .20 .46 .15 .28

.00 .29 .14 .10 .07

.63 .66 .63 .46 .53

.17 .03 .06 .20 .17

.42 .20 .16 .39 .29

.07 .09 .14 .18 .29

.51 .52 .73 .37 .56

.37 .25 .16 .41 .08

.10 .13 .01 .33 .02

.05 .28 .12 .08 .03

.37 .26 .14 .33 .23

.13 .09 .21 .24 .38

.45 .57 .69 .27 .47

.41 .25 .20 .43 .01

.16 .13 .03 .40 .04

.01 .18

.00 .05

.03 .73

.04 .02

.66 .05

.14 .15

.29 .21

.00 .16

.03 .15

.03 .05

.70 .11

.07 .78

.07 .00

.14 .16

.03 .22

.03 .11

.68 .19

.10 .72

.05 .05

.68

.11

.22

.18

.07

.18

.35

.09

.69

.12

.24

.30

.19

.21

.72

.14

.24

.29

.19

.19 .72 .62

.71 .14 .31

.11 .01 .09

.10 .10 .12

.07 .06 .02

.03 .04 .44

.11 .06 .15

.03 .32 .10

.18 .74 .60

.72 .11 .31

.14 .08 .07

.07 .04 .15

.06 .24 .15

.01 .04 .47

.16 .66 .76

.73 .17 .24

.09 .16 .20

.04 .06 .06

.06 .25 .09

.25

.41

.11

.38

.20

.13

.21

.07

.26

.47

.04

.05

.26

.19

.31

.43

.00

.05

.29

.79

.11

.13

.09

.02

.05

.20

.06

.77

.12

.12

.19

.05

.12

.73

.17

.06

.25

.04

.09 .48 .29

.70 .05 .14

.15 .17 .07

.03 .20 .11

.10 .07 .11

.08 .39 .53

.02 .08 .14

.26 .24 .03

.03 .40 .28

.67 .00 .13

.05 .10 .18

.20 .08 .03

.08 .04 .11

.00 .49 .53

.02 .56 .48

.65 .06 .05

.05 .24 .35

.21 .14 .14

.11 .12 .03

.24

.22

.00

.10

.11

.26

.08

.40

.28

.19

.12

.08

.28

.18

.32

.16

.08

.07

.32

.18 .43

.02 .01

.01 .12

.20 .35

.01 .04

.70 .51

.13 .03

.15 .15

.18 .39

.03 .00

.08 .00

.00 .10

.25 .40

.68 .54

.44 .56

.10 .08

.33 .17

.12 .05

.35 .47

.19

.02

.19

.27

.06

.54

.40

.18

.22

.03

.26

.20

.32

.52

.39

.04

.43

.14

.37

.02

.09

.53

.04

.26

.20

.19

.21

.01

.12

.35

.44

.19

.13

.08

.22

.23

.38

.19

Conscientiousness Competence Order Dutifulness Achievement Striving Self-Discipline Deliberation Social Potency Novelty Diversity Divergent Thinking Leadership Logical vs Affective Orientation Aesthetics Extraversion vs Introversion Enterprise Dependability Responsibility Emotionality Inferiority vs SelfAcceptance Practical Mindedness Optimism vs Pessimism Meticulousness Face Internal vs External LOC Family Orientation Accommodation Defensiveness Graciousness vs Meanness Interpersonal Tolerance Self vs Social Orientation

73

S. Laher / Personality and Individual Differences 81 (2015) 67–75 Table 4 (continued) Eight factor solution Scale

1

Veraciousness vs Slickness

3

Six factor solution 4

5

6

7

8

1

2

3

Five factor solution 4

5

6

1

2

3

4

5

.06

.11

.07

.69

.11

.43

.05

.06

.09

.16

.08

.07

.61

.47

.09

.04

.13

.13

.68

.02

.13

.07

.11

.41

.63

.02

.12

.10

.16

.34

.10

.12

.63

.17

.28

.59

.07

.04

.30

.02

.23

.08

.07

.29

.21

.38

.19

.01

.05

.15

.26

.39

.36

.09

.09

.27

.17

.04 .22 .12 .22

.12 .50 .17 .17

.38 .07 .09 .17

.29 .09 .27 .22

.20 .20 .17 .05

.02 .46 .09 .26

.36 .03 .21 .30

.18 .06 .60 .06

.01 .16 .20 .18

.12 .51 .13 .22

.34 .13 .26 .21

.37 .08 .00 .12

.37 .08 .54 .13

.01 .48 .01 .24

.05 .34 .12 .29

.15 .58 .21 .29

.31 .31 .20 .10

.43 .01 .12 .05

.31 .01 .51 .09

11.8

9.58

7.51

6.97

6.47

6.08

5.84

3.97

8.35

8.23

10.6 10.3

5.55 5.9

4.93 4.86

4.31 4.28

3.13 3.13

2.15 2.06

1.59 1.56

1.41 1.39

1.71

1.64

1.59

1.54

1.50

1.46

1.42

1.39

1.79

1.7

1.63

1.58

1.54

1.5

1.46

1.42

Interpersonal Relatedness Traditionalism vs Modernity Relationship Orientation Social Sensitivity Discipline Harmony Thrift vs Extravagance % Variance explained Actual Eigenvalue Random Eigenvalue MAP Random Eigenvalue PA Mean Random eigenvalue PA 95th percentile

2

on it together with Values from the Openness domain ( ve), Inferiority vs Self-Acceptance (+ve) and Internal vs External Locus of Control ( ve) from the Dependability domain and Traditionalism vs Modernity (+ve) and Discipline (+ve) from the Interpersonal Relatedness domain. Factor 7 appears to describe more of an Openness/Intellect dimension. Factor 8 may be described as a collectivist expression of Agreeableness since the foci of the scales that load on it are on compliant behaviour, maintenance of family ties and societal and familial harmony. Therefore both Factors 6 and 8 seem to be measuring some dimension of Agreeableness but Factor 6 represents Agreeableness as it pertains to the Accommodation factor identified by Cheung et al. (2008), and Factor 8 represents Agreeableness as it pertains to the Interpersonal Relatedness factor identified by Cheung et al. (2008). Results for the six factor joint CPAI-2 and NEO-PI-R analysis are presented in Table 4. The loadings on Factor 1 are best described as the Neuroticism factor. Family Orientation from the CPAI-2 Dependability factor also loads on Factor 1. Factor 2 is representative of the Conscientiousness domain and Factor 3 represents the Openness to Experience domain of the FFM. All the Openness to Experience facets load together on Factor 3 except for Values, which loads on Factor 6. Factor 4 represents the Extraversion factor and Factor 5 represents the Agreeableness factor of the FFM. Factor 6 represents the social relational elements discussed earlier. However the loadings of the scales are not wholly congruent with the Interpersonal Relatedness dimension proposed by Cheung et al. (2008) suggesting that Interpersonal Relatedness may have a different expression in the South African context. Table 4 also presents the results for the five factor solution. In the five factor solution, Factor 1 can be described as the Neuroticism factor, Factor 2 – the Conscientiousness factor and Factor 3 the Openness to Experience factor of the FFM. The Accommodation subscale of Self vs Social Orientation loaded only on Factor 3 with a loading of .38. Factor 4 represents the Extraversion factor and Factor 5 represents the Agreeableness factor of the FFM. The CPAI-2 Interpersonal Relatedness subscale Relationship Orientation did not load at the .40 cut-off on any of the factors but loads at .36 on Factor 1. The Dependability subscale of Family Orientation did not load at the .40 cut-off on any of the factors but was cross-loaded on Factor 1 with a negative loading of .32 and Factor 5 with a positive loading of .32. This finding is similar to the

11.6

9.67

8.6

6.5

13.1

10.13

9.4

8.32

8.3

Interpersonal Relatedness subscale Thrift vs Extravagance, which does not load at the .40 cut-off on any of the factors but is crossloaded at .29 on Factors 1 and 2. 4. Discussion 4.1. Replicability of the four factor model An exploratory factor analysis with varimax rotation was conducted with the extraction of four factors, to determine if the model postulated by Cheung et al. (2008) would hold in a South African student sample. A re-alignment of the axes using Procrustes rotation indicates that it is possible for there to be significant agreement between the Chinese normative factor structure and the South African factor structure. This result is congruent with the argument put forward by Cheung et al. (2011) that the congruence of the four factor structure was generally confirmed in crosscultural samples. However it is necessary to consider the other evidence provided in this study. The reliability coefficients and the joint factor analysis with the NEO-PI-R as well as examination of the four factor solution by itself suggest some reconceptualisation of the four factors as proposed by by Cheung et al. (2008). The five factor solution for the CPAI-2 indicated that three of the four factors postulated by Cheung et al. (2008) were recoverable, namely Social Potency, Accommodation and Interpersonal Relatedness. The Dependability factor did not replicate clearly with the Dependability subscales, loading differentially on Factors 2, 3, 4 and 5. An examination of the Dependability subscales suggests that this factor specifically needs more research in terms of its conceptualisation. The Emotionality, Inferiority vs Self-Acceptance and Optimism vs Pessimism Dependability subscales appear to be congruent with Neuroticism in the FFM. The Responsibility, Practical Mindedness and Meticulousness subscales appear to articulate with the Conscientiousness domain in the FFM. The remaining subscales of Face, Family Orientation and Internal vs External Locus of Control seem to fit better with the Interpersonal Relatedness factor in the model proposed by Cheung et al. (2008). An examination of the literature on the CPAI and the CPAI-2 does indicate some difficulties with the scales that load on Dependability, with Face being the most problematic subscale (see Cheung et al., 2003, 2006, 2008).

74

S. Laher / Personality and Individual Differences 81 (2015) 67–75

4.2. Joint factor analysis of the NEO-PI-R and the CPAI-2 A similar pattern of results is observed on joint factor analyses of the CPAI-2 and the NEO-PI-R. The four factors as postulated by Cheung et al. (2008) tend to be subsumed with the five factors of the FFM in both the eight and five factor solutions. In the six factor solution an Interpersonal Relatedness factor is discernible but not in the same form as it appears to be understood in Cheung et al. (2003, 2008). It appears to be broader and more defined in terms of the individualistic vs collectivist dimensions or the social relational concepts discussed in the literature (see Triandis, 2001; Valchev et al., 2014). Social Potency appeared to correlate well with Openness to Experience and this concurs with the findings of Cheung et al. (2008). Accommodation tends to concur with the Agreeableness dimension. However Dependability was not clearly discernible, with the subscales loading on many and varied factors. It would seem that in the South African sample, the CPAI-2 subscales have validity but not in terms of the factors in which they are arranged. A realignment of subscales appears to be necessary, particularly with regards to the Dependability factor. There are also consistent differences with the loadings on the Family Orientation, Relationship Orientation and Thrift vs Extravagance subscales of the CPAI-2. These were differentially loaded on the three solutions and did not load above the .40 cut-off except in the six factor solution where Relationship Orientation and Thrift vs Extravagance loaded on the Interpersonal Relatedness factor. In the three previous studies (Cheung et al., 2001; Cheung et al., 2003; Cheung et al., 2008), a common difficulty was noted with the Family Orientation scale, which either did not load on the Dependability factor at all or did load but with a small loading that corresponded to the ± .40 cut-off. Relationship Orientation did not load above the .40 benchmark in the five factor solution in Cheung et al. (2008), and Thrift vs Extravagance and Social Sensitivity had their highest loadings on the Interpersonal Relatedness factor in the six factor solution (Cheung et al., 2008) but both loadings were below .40. Thus from the results of the current study and those of other studies, the Interpersonal Relatedness scales do appear to have merit but need to be researched further and the Family Orientation scale specifically (currently located within the Dependability factor) would benefit from more research. A final point of interest is the negative loading on the Internal vs External Locus of Control subscale in the eight and six factor solutions respectively. Internal vs External Locus of Control appeared to be loading in the direction of an external locus of control whenever it loaded with the Interpersonal Relatedness dimensions in the current study and in previous studies. Given my arguments earlier on the Interpersonal Relatedness dimension representing the individualism/collectivism distinction, it would seem that collectivist dimensions are associated with an external locus of control. Individuals with an external locus of control tend to stress the importance of external conditions, are resigned to destiny and believe in chance, fate or luck, subscribing to the belief that the success or failure of an event depends on external factors (Rotter, 1975). Given this, it is self-evident why an external locus of control would be associated with collectivist cultures. Collectivist societies emphasise people’s interdependence within the group, group goals are given priority and people’s behaviour is largely regulated by group norms rather than personal attitudes. Therefore, people in a collectivist society are mainly interested in maintaining relationships with others and avoiding conflict (Triandis, 2001). Specifically in the South African context, this is supported with research by Valchev et al. (2014) who found that Black people in the South African context (whose culture is located on the collectivist continuum) endorsed items about conveying knowledge, giving guidance, empowering others and maintaining harmonious relationships more readily than White people (who tend to be

more aligned with individualist cultures). The same was true of the negative aspect of the social-relational spectrum where Black people were also more likely to demonstrate behaviours that were disruptive of interpersonal relationships and social harmony. These findings are interrogated by Valchev et al. (2014) within the context of the understanding of Agreeableness as it is conceptualised in the FFM with the suggestion that there is a need to look more closely at how the construct is understood and measured in light of evidence provided from the South African Personality Inventory Project. The split of Agreeableness loadings in the eight factor solution is of particular interest as this provides further evidence that the factor may be represented in different ways across individualist and collectivist expressions.

5. Conclusion It is evident from the results that it is possible to find agreement for the four factor model proposed by Cheung et al. (2008) within the factor space. However the evidence obtained from the various analyses do not allow for a definitive conclusion on the replicability of the model in a South African sample. The results on the Dependability factor in particular suggest that some reconceptualisation of the four factor model is necessary before the CPAI-2 and the four factor model can be described as cross-culturally applicable. The replicability of the FFM for the most part adds to an already burgeoning body of evidence in favour of the universality of the FFM (see McCrae et al., 2005). The findings would also support the FFT (McCrae & Costa, 2008), which argues for a biological basis for personality traits (Basic Tendencies) that get expressed as Characteristic Adaptations. The consistent support for an Interpersonal Relatedness factor however suggests that the FFM is not adequately capturing this dimension. This is made even more salient by the use of students in this sample. South African students come from various language and ethnic groups but are described as fairly acculturated (towards a western culture) and individualist (Van Dyk & De Kock, 2004). Evidence for the social relational aspects of personality was found in a sample that would be described as more individualist suggesting that collectivist expressions of personality are as universal as current individualist expressions. The comparison of factor structures across ethnicity and language in this study also revealed agreement between the groups further supporting the argument that collectivist expressions of personality are universal, not dependent on a particular culture or group. It is acknowledged that the sample size is small and more research needs to be conducted on bigger, more representative and diverse South African samples. However the implications for the FFM are evident. The findings demonstrate support for the current conceptualisations of the five factors to be revised so as to include more collectivist expressions in the definitions of the five factors that will serve to increase the psychic unity of the model. More research in the South African and African context is needed to explore further what form this inclusion may take in this context. The SAPI relational cluster represents an attempt to do this. Future research would need to include the SAPI, an emic instrument, with the CPAI-2 and the NEO-PI-R in order to inform the adaptations necessary to make the FFM more universally applicable. Laher (2013) suggests that further research be undertaken with regards to the South African concept of Ubuntu which originates from an African aphorism, ‘‘umuntu ngumuntu ngabantu’’ (isiZulu version) or ‘‘motho ke motho ka batho’’ (Sesotho version), which translates as, ‘‘a person is a person through other persons’’ (p. 214). Laher (2013) argues further that traits associated with Ubuntu (generous, hospitable, friendly, caring, compassionate,

S. Laher / Personality and Individual Differences 81 (2015) 67–75

open and available to others, affirming of others, does not feel threatened that others are able and good) can be subsumed within the Extraversion and Agreeableness factors in the FFM. This argument lends support to incorporating more collectivistic expressions of the five factors into the FFM rather than adding a new factor. There is also a body of literature that calls for a two factor structure of personality (see Saucier et al., 2013). In the South African context, Valchev, Nel, Van de Vijver, Meiring, De Bruin and Rothmann (2012) provide empirical evidence from the South African Personality Inventory (SAPI) project that distinguishes between a personal growth cluster and social relational cluster. The consistent element in all of this research is the inclusion of a collectivist, social relational dimension. This suggests that the question is no longer whether a collectivist dimension exists in the structure of personality but rather what form this dimension will take within the five factor space. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.paid.2014.12.010. References 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. Branco e Silva, L. & Laher, S. (2012). Exploring the collective dimension of personality using the SAPI and the CPAI-2. Paper presented at the 30th International Congress of Psychology, 22–27 July, CTICC, Cape Town, South Africa. Cheung, F. M., Cheung, S. F., & Fan, W. Q. (2013). From Chinese to cross-cultural personality inventory: A combined emic-etic approach to the study of personality in culture. In M. J. Gelfand, C.-Y. Chiu, & Y.-Y. Hong (Eds.), Advances in culture and psychology (Vol. 3). Oxford University Press. Cheung, S. F., Cheung, F. M., Howard, R., & Lim, Y.-H. (2006). Personality across the ethnic divide in Singapore: Are ‘‘Chinese traits’’ uniquely Chinese? Personality and Individual Differences, 41, 467–477. Cheung, F. M., Cheung, S. F., Leung, K., Ward, C., & Leong, F. (2003). The english version of the Chinese Personality Inventory. Journal of Cross-Cultural Psychology, 34, 433–452. Cheung, F. M., Cheung, S., Zhang, J., Leung, K., Leong, F., & Yeh, K. H. (2008). Relevance of Openness as a personality dimension in Chinese culture: Aspects of its cultural relevance. Journal of Cross-Cultural Psychology, 39, 81–108. Cheung, F. M., Leung, K., Zhang, J. X., Sun, H. F., Gan, Y. Q., Song, W. Z., et al. (2001). Indigenous Chinese personality constructs: Is the Five-Factor Model complete? Journal of Cross-Cultural Psychology, 32, 407–433.

75

Cheung, F. M., Van de Vijver, F., & Leong, K. (2011). Toward a new approach to the study of personality in culture. American Psychologist, 66, 593–603. Costa, P. T., Jr., & McCrae, R. R. (1992). Revised Neo Personality Inventory (NEO-PI-R) and Neo Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources. Imperio, S. M., Church, T. A., Katigbak, M. S., & Reyes, A. S. (2008). Lexical studies of Filipino person descriptors: Adding personality-relevant social and physical attributes. European Journal of Personality, 22, 291–321. Laher, S. (2013). Understanding the Five Factor Model and Five Factor Theory through a South African Cultural Lens. South African Journal of Psychology, 43, 208–221. McCrae, R. R., & Costa, P. T. Jr., (2008). The Five Factor Theory of personality. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and research (3rd ed., pp. 157–180). New York: Guilford Press. McCrae, R. R., Terracciano, A., & 79 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. McCrae, R. R., Zonderman, A. B., Bond, M. H., & Paunonen, S. V. (1996). Evaluating replicability of factors in the Revised NEO Personality Inventory: Confirmatory factor analysis versus procrustes rotation. Journal of Social and Personality Psychology, 70, 552–566. 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. Rotter, J. B. (1975). Some problems and misconceptions related to the construct of internal versus external control of reinforcement. Journal of Consulting and Clinical Psychology, 43, 56–67. Saucier, G., Thalmayer, A. G., Payne, D. L., Carlson, R., Sanogo, L., Ole-Kotikash, L., et al. (2013). A basic bivariate structure of personality attributes evident across nine languages. Journal of Personality. http://dx.doi.org/10.1111/jopy.i2028. Statistics South Africa (2011). Census 2011. Retrieved 8 September 2014 from . Triandis, H. C. (2001). Individualism-collectivism and personality. Journal of Personality, 69, 907–924. Valchev, V. V., Nel, J. A., van de Vijver, F., Meiring, D., De Bruin, G. P., & Rothmann, S. (2012). Similarities and differences in implicit personality concepts across ethnocultural 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., et al. (2014). Beyond agreeableness: Social-relational personality concepts from an indigenous and cross-cultural perspective. Journal of Research in Personality, 48, 17–32. Valchev, V., Van de Vijver, F., Nel, 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. Van Dyk, G. A. J., & De Kock, F. S. (2004). The relevance of the IndividualismCollectivism (IC) factor for the management of diversity in the South African National Defence Force. South African Journal of Industrial Psychology, 30, 90–95.