PERSONALITY CORRELATES OF MANAGERIAL TALENT: CROSS CULTURAL COMPARISON A.D. DE BOD

Ten spyte van die wêreldwye gebruik van die bestuursbeoordelingsentrum (AC) is daar tot dusver weinig kruiskulturele navorsing hieroor gedoen. 'n Verdere verwaarloosde area van AC-navorsing, is die bestudering van persoonlikheid. Hierdie studie ondersoek die verwantskap tussen bestuursdimensies en persoonlikheidsattribute van 'n groep Kanadese (N = 1199) en 'n groep Suid-Afrikaanse (N = 177) middelvlakbestuurders. Die eerste stap was om die meetinstrumente wat gebruik is te toets en die twee groepe te vergelyk. Die ingesamelde data vir beide groepe is aan 'n korrelatiewe en diskriminant-funksie ontleding onderwerp. Die ontledings is gebruik om (a) die persoonlikheidskorrelate van bestuurstalent te omlyn en (b) breë tendense bloot te lê ten opsigte van die relatiewe invloed van kultuur op die verwantskap tussen persoonlikheid en bestuurstalent.

personalities in action.They contribute very little, however, to the study of personality" (Bray, 1982, p. 183).
Any attempt to compare research results obtained in different countries with the aim to eventually make deductions as to the relative influence of culture, albeit only to detect broad tendencies, necessitates the use of similar (or at least comparable) measuring instruments.The first step in this study was therefore to compare the instruments used by the two organizations to assess managerial talent and personality.

COMPARISON OF THE MEASURING INSTRUMENTS
One of the prime features of an assessment centre is the use of managerial simulations as a method of evaluating managerial talent.Such exercises simulate the type of managerial work to which the participant would be exposed at a higher level, thus allowing his/her performance to be observed under quasi-realistic conditions.
The advantage of simulation exercises is that they allow evaluation of performance across a variety of situations and contexts, thus exposing the individual to as broad a range of managerial activity as possible.The individual can be observed in both structured and unstructured situations, assigned and non-assigned roles, competitive and co-operative settings, and on an individual or group basis.
In the assessment centres employed by the two organizations participating in this study, similar exercises are used to evaluate managerial talent, namely an in-basket, a structured leaderless group exercise, an analysis problem, a presentation exercise and an unstructured leaderless group discussion.In both assessment centres the assessors write assessment reports giving a description of the assessee's behaviour in the various exercises and presents his/her ratings on the dimensions assessed with specific behavioural examples to substantiate the ratings given.Although the dimensions which are evaluated in the two assessment centres differ By looking at the assessee from many different perspectives over the length of the assessment program (different dimensions, different assessors, different exercises), it is possible to arrive at a point where sufficient and reliable predictions of talents can be made.At both assessment centres the assessors consequently conclude (summarize) their decision in a so-called Overall Assessment Rating (OAR) which divides assessees into the following three groups:

GO
"Yes, at this point in time, based on this individual's performance in a simulated senior executive position and in the judgement of the assessors at the assessment centre, he/she has the potential to reach the senior executive level".
NO GO "No, at this point in time, based on this individual's performance in a simulated senior executive position, and in the judgement of the assessors at the assessment centre, he/she does not have the potential to reach the senior executive level".
UNCERTAIN "It was not possible on the basis of the individual's performance in a simulated senior executive position and in the judgement of the assessors at the assessment centre to predict with any degree of certainty whether he/she has or does not have the potential to reach the senior executive level".
The OAR is the sum total of the different perspectives which are gathered from a person's participation in the various simulated exercises.Yet another perspective of the individual is often gained by employing other assessment techniques in addition to the management simulations.
Within any assessment centre program it is common for simulation exercises to be combined with one or more other assessment tool including paper-and-pencil tests, selection interviews, records of past experience, career progression and employee appraisals.Each different assessment tool adds further relevant information that will assist in making appropriate decisions about the individual.
In the two assessment centres under discussion, the assessees also complete, in addition to the simulation exercises and the in-basket test, certain tests of mental ability as well as a personality questionnaire.These tests are intended to measure some of the factors within the general domain of cognitive abilities and interpersonal relations, however not all this information is utilized by the assessors in arriving at their overall assessment rating for an assessee.
Since this study investigates the relationship between personality traits and assessment centre behavioural dimensions, only the personality questionnaires which are used in the two assessment centres will be described.The Personality Research Form (PRF) (Jackson, 1967) is used in the Canadian Assessment Centre, while the 16 Personality Factor Questionnaire (16 PF) (Cattell, 1970) is used in the South African assessment centre.
The PRF consists of 400 true-false items which are combined into 22 scales to measure the strength of various motivational factors.Generally, each scale represents a relatively independent personality trait, such as achievement, dominance, etc. which is influential in determining behaviour in common situations (See Annexure 1).The 16 PF consists of 187 multiple choice items.These items describe certain attitudes, reactions, etc. in the form of statements and the testee has to indicate whether these are true, false or uncertain as applied to himself.Responses to the items are combined to measure sixteen primary personality factors From the above it is clear that the two personality instruments are not in every respect similar, but when one carefully scrutinizes the high score descriptions of the personality traits measured (see Annexure 1 and 2), it seems safe to conclude that the questionnaires measure very much the same thing.This deduction however, is intuitive and thus unacceptable for any scientific purposes, unless it can be confirmed by either a conceptual/logical process or by empirical findings, but preferably by both.
In order to conceptually and logically verify the similarity between the PRF and the 16 PF, four clinical psychologists fully conversant with both questionnaires, were asked to independently compare the two sets of scales and to indicate which of them measure the same basic personality trait.
The postulate was that if there was unanimous agreement on any of the scales, they could be used when comparing the Canadian and South African research results.The outcome of this exercise is tabulated in Table 2.
Empirical evidence on the relationship between the PRF and 16 PF will serve to strengthen any ties that are made conceptually between scales.Unfortunately empirical evidence to substantiate the conceptual connection between the various PRF and 16 PF scales, as set out above, could not be found, thus limiting the validity with which similarities between the Canadian and South African research could be generalized.6 On the whole however, there appears to be enough similarity between the measuring instruments to allow at least a comparison of the research results with the aim to establish broad tendencies.The research was carried out against this background.

METHOD
The two subject groups for this study were formed by all the middle managers who participated in the respective assessment centres since their inception.The Canadian assessment centre was initiated in 1973 while the South African centre was initiated in 1978.
Since then 1,199 and 177 middle managers have participated in the two assessment centres respectively.These individuals constitute the two subject groups of this study.For each subject in the Canadian group 13 behavioural measures (12 assessment centre dimensions plus 1 overall assessment rating) and 22 personality (PRF scales) measurements were available.The South African subjects had a similar amount of behavioural measurements, but only 16 personality measurements.The assessment centre ratings for both groups are ratings on a 7 point scale, while the OAR is a 3 point scale of 1 for a "NO" rating, 2 for an "UNCERTAIN" rating and 3 for a "YES" rating.The personality measurements are the raw scores obtained on the respective questionnaires.The data on the two groups was treated separately.It was only after this analysis that the results were compared, but then only on the measurements which were comparable (10 behavioural measurements (see Table 1), 16 personality measurements (see Table 2), 26 measurements in total).
Table 3 shows the correlations between the 22 PRF scales and the 13 behavioural ratings for the Canadian group.
Several points emerge from Table 3: All the behavioural dimensions correlated significantly with some, though not necessarily the same sub-scales of the PRF.Judgement for instance correlated significantly with 6 scales while interpersonal relations showed 10 significant correlations.Creativity, intelligence, independence and oral communication also showed significant correlations with various personality traits.

−
Only 4 of the 22 PFR scales did not correlate at all with the behavioural dimensions (abasement, endurance, social recognition, succorance).

−
Cognitive structure, impulsivity, sentience and understanding showed the most significant correlations with the behavioural dimensions.These personality traits also correlated significantly with the overall assessment rating.
A correlational analysis was also performed on the South African data.The results are displayed in Table 4.
From the correlation matrix the following can be deducted: − Four of the behavioural dimensions (initiative, judgement, flexibility, analytical ability) showed no correlation with the 16 personality traits.The remaining dimensions significantly correlated with at least 1 of the 16 PF scales (empathy), and at the most with 7 of the 16 scales (utilization and development).

−
Only 3 of the 16 PF scales did not correlate at all with the behavioural dimensions (B, C, Q3).
− Factors E and Q2 showed the most significant correlations with the behavioural dimensions.These factors as well as factors F, H and Ql correlated significantly with the overall assessment rating.
On the whole it seems safe to say that managerial talent definitely has certain personality correlates.To further establish the validity of the relationship between managerial talent and personality, a discriminant function analysis was performed on the personality scales (the PRF scales in the case of the Canadian group and the 16 PF scales in the case of the South African group) using the OAR as the classifying variable.The breakdown of the two subject groups with respect to this variable was as follows (Table 5): The group centroids for both the Canadian and South African subject groups indicate that Function 1 differentiates the "Yes" candidates from the "No" and "Uncertain" candidates, while Function 2 differentiates the "Uncertain" candidates from the "Yes" and "No" candidates.
Based on the discriminant analysis, together with the correlational data already reported, the following deductions can be made: − Canadian middle managers with managerial talent ("yes" candidates) can be characterized as being higher on impulsivity sentience and understanding and somewhat higher on endurance and harmavoidence.In contrast the managers without managerial talent ("No" candidates) tended to score higher on nurturance and somewhat higher on abasement, aggression, defendence and dominance.

−
South African middle managers with managerial talent ("Yes" candidates) can be characterized as being higher on factors Q1 and E, somewhat higher on factors C and F and low on factors Q2 and G.In contrast the managers without managerial talent ("No" candidates) tended to score higher on Q2 and L and low on factors E and Q1.Planning and organising (planning and organising) Oral communication was the only behavioural dimension that correlated significantly with all five personality traits.Interpersonal relations and planning and organising were the only dimensions which correlated with one personality trait only.The remaining dimensions correlated with at least two or more personality traits.
Among the personality traits, impulsivity (E) was the only trait which correlated with all but one of the behavioural dimensions.The remaining personality traits correlated with two or more behavioural dimensions.
All the correlational similarities between the personality traits and behavioural dimensions for the Canadian and South African subject groups were positive, except for those between cognitive structure (Q2) and the related behavioural dimensions, which were negative.On the surface the latter finding seems to be illogical, but when the high score descriptions of cognitive structure and Q2 are scrutinised, it becomes clear why talented managers tend to score low on these two personality traits (See Annexure 1 and Annexure 2).
Further outstanding correlational similarities were found in the relationship between the five personality traits and overall assessment rating (OAR).Again all the correlations were positive, except those between cognitive structure (Q2) and OAR.
A comparison of the two discriminant function analysis also indicated similarities between the Canadian and South African groups.Managers with potential ("yes" candidates) scored high on impulsivity (E) and understanding (Ql), while the managers without potential ("No" candidates) scored high on aggression (L), and defendence (L).

CONCLUSION
Based on these comparisons the personality profiles of the Canadian and South African managers appear to be similar.The overall conclusion reached is that the personality correlates of managerial talent seem to be culturally independent both for a total group of managers and for "high" and "low" talent managers.It is therefore possible to construct a capsule description of the personality profile and behavioural repertoire of a high talented manager.Such a description would run along the following lines: The high talented manager: the Canadian and South African subject groups emerged from both the correlational and discriminant function analysis.Very positive correlational similarities between the Canadian and South African subject groups were found.The results of the comparison are shown in Table12.The various similarities in the results of the two correlational analyses, were found in the relationship between five personality traits and seven behavioural dimensions (delegation (utilization anddevelopment) ExhibitionWants to be the center of attention; enjoys having an audience; engages in behaviour which wins the notice of others; may enjoy being dramatic or witty.Harmavoidance Does not enjoy exciting activities, especially if danger is involved; avoids risk of bodily harm; seeks to maximize personal safety.ImpulsivityTends to act on the "spur of the moment" and without deliberation;gives vent readily to feelings and wishes; speaks freely; may be volatile in emotional expression.Nuturance Gives sympathy and comfort; assists others whenever possible, interested in caring for children, the disabled, or the infirm; offers a "helping hand" to those in need; readily performs favours for others.OrderConcerned with keeping personal effects and surroundings neat and organized; dislikes clutter, confusion, lack of organization; interested in developing methods for keeping materials methodically organized.Play Does many things "just for fun"; spends a good deal of time participating in games, sports, social activities, and other amusements; enjoys jokes and funny stories; maintains a light-hearted, easy-going attitude toward life.Sentience Notices smells, sounds, sights, tastes, and the way things feel; remembers these sensations and believes that they are an important part of life; is sensitive to many forms of experience; may maintain an essentially hedionistic or aesthetic view of life.
Managerial Talent: A Cross-Cultural Comparison (A.D. de Bod & L.W. Slivinski) Social Desires to be held in high esteem by acquaintances; concerned about Recognition reputation and what other people think of him; works for the approval of recognition of others.