CPP Career Group Analytics: Action research and Artificial Intelligence (AI) analyses

By Christoff Prinsloo and Maretha Prinsloo on December 24, 2019

© 腾龙 郭/ adobe.stock.com



The goal of this study differs from that reported in part 2, which focused on the information processing tendencies of high level CPP performers or high capability groups across career fields. This study involved the use of AI techniques to identify the unique processing tendencies of various career groups, regardless of the levels of CPP performance involved. Both cognitive styles as well as Information Processing Competencies (IPCs) were considered to differentiate between groups. AI findings were also compared to typical career trends which emerged from qualitative action research by means of personal feedback on CPP results (and in some cases alternative values and personality test results as well) to approximately 5,000 individuals.

Given the challenges involved in statistically analyzing psychometric data, especially those of averaging effects, AI and action research are both well-positioned to reflect more realistic findings than is the case with “blind” number crunching.

For current purposes, an Artificial Intelligence (AI) product, namely Microsoft Power BI’s Key Influencer Visualization was applied to the CPP results of a sample of 24,600 candidates. Normalized-Standardized T-scores were used. The goal was to identify dimensions that best discriminate between various career groups by identifying “key influencers”. These differentiators indicate the likelihood of a specific career group to be characterized by certain processing tendencies at specific score ranges.

The analysis sample spanned several career fields, genders, age groups, geographical areas, educational levels and CPP results as indicated below:


Table 1: The Career groups used in this analysis


Figure 1: The gender representation of the sample


Figure 2: The sample’s educational level and educational field

Figure 3: The geographical areas represented



The AI program used in this study was aimed at analyzing the differences between groups to identify the cognitive dimensions that best differentiate a specific career group from all the others.  Whereas rank ordered scores (in the case of the Cognitive Styles) and Standardized T-scores (in the case of the IPCs) were used the first part of this blog series, rank ordered Normalized-Standardized T-scores (NST-scores) were used in part 2 and in this final part of the blog series.

The likelihood that a processing score (a specific factor) predicts the career group indicated, is also provided. For example, for older Accountants there is a likelihood of 4.19 for a factor value beyond the listed threshold value to belong to the career group of older Accountants. In other words, any candidate is 4.19 times more likely to belong to the older Accountants target group if they have an Analytical score above 28, which is the threshold value.

Another example, looking at the thinking styles of younger Accountants, shows the Intuitive Style ranking as the most powerful differentiator with a likelihood of 4.21. Of all the career groups, respondents with an Intuitive style score above 77 are thus 4.21 more likely to belong to the group of younger Accountants than of respondents with lower Intuitive Style scores. The bar chart below shows the distribution of respondents in the career groups for various levels of the specific factor.

For younger Accountants, the blue bar shows that 3.53 % of the entire sample belongs to the career group younger Accountants, whereas on average, 0.69% of the sample would have belonged to younger Accountants. That ratio (3.53/0.69) roughly indicates the 4.21 factor.

The styles and IPCs that best differentiate a specific career group from the others are not necessarily the group’s highest score dimensions, but may be the opposite or alternatively, reflect a specific score range.

Examples of the way in which the best differentiators per career group are indicated appear in Figure 4.


Figure 4: Best cognitive Style and IPC influencers of the older and younger Accountant groups


The results obtained through this AI study are reflected in Table 2 and Table 3:

Table 2: Key Cognitive Style differentiators of various career groups


Table 3: Key Information Processing Competencies (IPCs) of various Career groups


An interpretation of the findings of this AI study


The key influencer for older Accountants, or the stylistic preference that best differentiates them from other career groups is that of the:

  • Analytical style (T62 above 28 at a factor of 4.19)

And the key Information Processing Competency (IPC) influencers for this career group are those of:

  • Metacognitive Awareness and Alertness (T34 above 22 – at infinity)
  • Spontaneous comparison of elements (T09, above 80 at a factor of 5.98)

Younger Accountants (24 – 44 years of age) are best differentiated from other career groups by the stylistic preference of the:

  • Intuitive (T65 above 77 at 4.21)
  • Memory (T66 above 69 at 2.37)

And the key IPC influencers those of:

  • Extracting core elements (T23 above 80 at 41,68)
  • Speed (T56 above 75 at 9.77)
  • Learning (T44 above 72 at 8.51)

This may imply that long term involvement in an accounting career encourages the use of a high degree of Metacognitive awareness and an Analytical approach as characterized by detailed, precision and rule-based tendencies.

In the case of younger Accountants who, more so than other career groups tend to show an Intuitive stylistic approach (which to some extent overlaps with a Learning style), there is a reliance on the Memory of their knowledge base to meet the detail requirements of their jobs. Over time their Intuitive approach may therefore make way for a more detailed, factual approach to problem solving and strategizing. However, the exceptional speed, flexibility and learning orientation of younger Accountants, combined with the skill of identifying and extracting core elements, ensure highly effective cognitive functioning that basically covers most of the processing bases.

Action research aimed at identifying the cognitive characteristics of younger Accountants has indicated that most show a Tactical Strategy with potential for Parallel Processing SST complexity level. More than 70% of younger Accountants show particularly well-developed Logical-Analytical skills and preferences. The Intuitive approach of younger accountants come as a surprise, but less so given the Memory component which supports both an intuitive and an analytical application.



The key stylistic differentiators seem those of the:

  • Holistic (T64 above 63 at 6.39)
  • Memory (T66 above 66 at 5.75)
  • Analytical (T62 above 64 at 5.67)
  • Low Impulsive / Reactive (T68 less than 35 at 5.59)
  • Low Random / Trial-and-error (T69 less than 32 at 5.31) approaches.

In terms of their IPCs they are best differentiated from other career groups in terms of:

  • Metacognitive strategies for Logical reasoning (T30 above 73 at 13.46)
  • General processing approach (T04 above 67 at 7.56)
  • The use of Hypotheses (T42 above 68 at 7.07)
  • Metacognitive Strategies (T37 above 60 at 6.30)
  • Metacognitive Strategies for Exploration (T06 above 64 at 6.24)
  • The tendency to Extract Core Elements (T23 above 67 at 6.18).

As in the previous studies, Actuaries again showed a highly balanced cognitive approach in that they seem metacognitively aware, do not resort to random and impulsive approaches and accommodate both detailed and holistic information. Action research of the cognitive tendencies of Actuaries has indicated that this career group tends to obtain higher scores than most other career groups on almost all the processing scores – particularly those of Rule orientation and Categorization – which is not typical of comparable professional fields. They thus apply their thinking in a most rigorous, detailed and structured manner in a way that exceeds any of the other career groups.


Finance (other) Group

In this somewhat diverse career group which include Bankers, Economists, Financial advisors, Bookkeepers, etcetera, the relatively weak AI identified stylistic influencers included:

  • Impulsive (T68 below 52 at 1.65)
  • Learning (T72 above 48 at 1.58)
  • Holistic (T64 above 50 at 1.57).

The IPC influencers were:

  • Coherence (T16 above 40 at 1.67)
  • Exploration (T05 above 30 at 1.64)
  • Card Movements (T00 67-81 at 1.64)

Given the diversity of this group, averaging effects have resulted in somewhat unremarkable results. The group as a whole do, however, seem to be characterized by somewhat uneconomical exploration processes which in turn tend to lower the effectiveness of their more complex processes.

Previous research studies of the preferences and capabilities of financial people other than Accountants, included empirical and qualitative explorations. These studies were conducted in various sectors within the Financial Industry, such as Banks and Insurance firms, as well as across geographical regions and specific jobs. Interesting results, for example, include that Economists at a Central Bank, who generally show a Tactical Strategy SST level or higher, and who have postgraduate or multiple degrees, showed either a surprisingly high Random (which is an ineffective approach) or a Holistic cognitive style (which is an effective big picture approach within both practical and theoretical contexts).

Action research also indicated that almost all Economists seem to show an ideas orientation which differs from the more factual and logical approaches which is generally characteristic of Accountants. Credit Managers from Commercial versus Corporate Banking divisions showed very clear differences in cognitive approach. In one such study, 79% of Credit managers from the Corporate division of a bank showed a marked “left-brain” preference through the application of Logical, Analytical and Reflective styles in dealing with issues of corporate viability and financial structuring. Credit managers from the Commercial division of the same bank, who mostly deal with individuals and small businesses, showed a typical “right-brain” approach though a reliance on Metaphoric, Integrative and Random approaches. The Commercial Credit Managers also showed a significantly higher need for structure than the Corporate Credit managers did. Insurers from various regions seem to work at different levels of complexity (SST): Insurance Executives from South African, the United Kingdom and one Asian insurer, seemed to work at similar levels of complexity which significantly differed from that of their Australian counterparts.

Those involved in IT banking systems from South Africa and the UK also showed relatively similar cognitive profiles which clearly differed from that of the rest of SADC African banks in terms of levels of complexity and stylistic preferences. Whereas the former group prefer a Logical and Integrative approach to problem solving, the latter SADC group applied a more operational approach characterized by Explorative and Reflective cognitive styles.



The key stylistic influencer seems to be that of:

  • Low scores on Logical reasoning (T67 below 21 at 6.44)

whereas the key IPC influencers include a:

  • Low degree of Card movements (T00 below 22 at 4.54)
  • Inadequate Exploration (T05 below 28 at 1.77)

The Administrators generally show under-developed cognitive skills, particularly where it comes to Exploration processes which are crucial to identify the relevant information which more complex processing relies on. Analytical skills training with administrators showed them responsive to the repeated practice and internalization of the metacognitive criteria aimed at optimizing Exploration skills, including: “Is this important?”, “Is this clear?” and “Do I need more information?”. Their cognitive effectiveness also tends to improve by developing the habit of making lists and representing information in simple structures or graphs.



In the case of older Engineers, the stylistic differentiators were:

  • Explorative (T61 above 26 at 8.55)
  • Integrative (T70 above 46 at 2.25)
  • Holistic (T64 above 45 at 2.17)

The IPC influencers were:

  • Exploration (T05 above 24 at infinity)
  • Detail (T17 above 75 at 4.48)
  • Precision (T11 above 45 at 2.60)

It thus seems that older Engineers, more so than most other career groups except IT developers, invest the bulk of their energy in the detailed exploration of unfamiliar situations. Their Exploration processes (as measured by “explorative tendencies” or T05 as opposed to T43 which measures “effective exploration processes” only), may not necessarily be economical or carefully directed. They therefore tend to look for information from various perspectives and may over-explore at times. While exploring as widely and in such depth as they tend to, which reflects a detail orientation, they also manage to continually integrate relevant information to understand the situation and formulate a solution.

Compared to other career groups, younger Engineers also showed a rigorous detailed and integrative approach as indicated by their preferred cognitive styles of:

  • Analytical (T62 above 64 at 2.66)
  • Logical (T67 above 61 at 2.50)
  • Integrative (T70 above 63 at 2.46)

In terms of their IPCs, the younger Engineers could be recognized by their higher scores for:

  • Metacognitive strategies for Logical Reasoning (T30 above 69 at 2.90)
  • Metacognitive Exploration Strategies (T06 above 69 at 2.76)
  • Analytical Processes (T40 above 58 at 2.43)

Given their strong metacognitive awareness and the degree of self-monitoring of their thinking processes, young Engineers thus deal effectively and economically with new and unfamiliar information. Both the older and younger engineers do, however, show a technical and rigorous approach which to some extent differentiates them from other career groups. Action research through personal feedback, aimed at identifying the cognitive characteristics of Engineers, has indicated that this career group, more so than most others, are cognitively diverse and rely on a wide range of cognitive preferences and capabilities.


IT Developers, Programmers and Data Analysts

This group seem best differentiated by their cognitive style preferences for:

  • Learning (T72 above 46 at 2.09)
  • Reflective (T74 above 50 at 1.99)
  • Logical (T67 above 50 at 1.94)

Their most characteristic IPCs included:

  • Exploration (T05 above 29 at 4.82)
  • Low Coherence (T16 below 17 at 4.27)
  • Precision (T11 above 45 at 2.32).

Here the averaging effects of group diversity somewhat obscured the strong visual mode of processing and well-developed learning and analytical orientation of programmers as found through qualitative and action research investigations. The low Coherence scores, which is measured by the CPP according to various aspects of conceptualization, using auditive modes of processing, do however reflect the visual orientation of this group which has repeatedly been confirmed in case studies and personal feedback.

Action research has also indicated IT developers and programmers as the most Learning oriented of all the career groups. This group generally consists of younger people who work in fast-changing environments which tends to attract those who are cognitively adaptable and agile. Qualitative action research also indicated that besides Quick insight Learning (T15), this groups is also characterized by an Intuitive (T65 and T54) approach and their personality characteristics include openness, curiosity, a need for challenge and self-confidence regarding their intellectual prowess.


Lawyers and Attorneys

Those from the Legal fraternity can be expected to differ from other groups’ stylistic preferences in terms of:

  • Random (T69 below 63 at 1.48)
  • Metaphorically inclined (T75 above 65 at 1.47).

In terms of their IPCs the key influencers were:

  • A Focus on External Clues (T1 above 22 at 11.60)
  • Self-Monitoring (T36 below 66 at 2.05)
  • Task Orientation (T35 between 32 and 63 at 1.97)

Based on action research involving personal feedback to people from this career category, they generally seem to show complexity preferences for Tactical Strategy work – in some cases with “potential” for Parallel Processing work complexity. They also, almost invariably, showed the fairly unique combination of a factual, Logical with an ideas-oriented, Metaphoric cognitive styles. Other than alternative career groups such as the Accountants, Programmers and Engineers who are analytically inclined, Lawyers thus seem to be more interested in the world of ideas, they often capitalize on metaphors and analogies when explaining concepts, tend to use abstract language and apply a metacognitively rigorous approach to problem solving. The latter findings based on informal action research are thus partly verified by the AI results on this sample of legal experts.


The Social Sciences Career Group

For this group which includes HR practitioners, Social workers and Teachers the processing style differentiators revolved around average to above average:

  • Random (T68, 50 – 71 at 1.48)
  • Low Integrative (T70 below 50 at 1.47)
  • Low Logical (T67 below 50 at 1.47)
  • Low Reflective (T74 below 50 at 1.43)

The best IPC differentiators seemed that of:

  • Speed (T56 below 70 at 2.53)
  • Low Metacognitive monitoring of Analytical Processes (T12 below 52 at 1.74)

Compared to most other professionally trained career groups, those from Social Sciences seem somewhat less effective in terms of their cognitive competence in that this group shows relatively low metacognitive awareness, as well as low integrative and logical skill. Action research as well as quantitative research amongst social sciences groups have indicated a widely diverse group, though. In terms of the levels of complexity these individuals seem to display capabilities ranging from Pure Operational to Pure Strategic SST Levels. Their stylistic preferences and information processing competencies also widely differed. It would be misleading to generalize specific processing findings for this group.


Marketing and Sales

In the case of these career groups the sample showed the key cognitive style influencers to be that of:

  • Memory (T66 below 73 at 3.89)
  • Holistic (T64 above 38 at 1.59)
  • Metaphoric (T75 above 35 at 1.57)

IPC influencers were those of:

  • Low on the Degree of Detail Focused on (T17 below 20 at 6.40)
  • High Memory (T31 below 74 at 3.19).

This particular combination means that although this career group relies on detail, this detail is mostly based on previously acquired knowledge and experience (as opposed to the tendency to pull situations apart in order to understand the building blocks and their interrelationships) as opposed to fresh analyses of unfamiliar information. Within an unfamiliar problem-solving situation, they may thus fall back on a more general problem-solving approach.

Again, action research on the cognitive characteristics of people in media, journalism, marketing and advertising showed a clear ideas orientation and tendency to capitalize on verbal conceptualization (as reflected by the Metaphoric style tendency). Those involved in the formulation of creative advertising strategies, often seem to intentionally apply a particularly Random and Impulsive approach. More characteristic of this Marketing and Sales career group than their cognition, however, is their worldview as indicated by the Value Orientations (VO) assessment.

On the VO, those in Marketing roles often show an “Orange” orientation characterized by a focus on the creation of value; skill in manipulating perceptions; a belief in their capability to create their own future; personal resilience; as well as a creative strategic orientation. This worldview has been found to prevail across regions and is as prevalent in Africa as it is in Europe, Asia or the Americas. Within Sales the Red value system as measured by the VO was the most common preferences. The latter is associated with high energy as well as a degree of forcefulness and a need for achievement and proving oneself.


Medical Doctors

This career group was best differentiated from other career groups based on the following stylistic and processing characteristics. In terms of cognitive style:

  • Reflective (T74 above 43 at 1.68)
  • Low Impulsivity (T69 below 55 at 1.64)
  • Analytical (T62 above 44 at 1.62)
  • Quick Insight (T73 above 42 at 1.60)

In terms of their IPCs, differentiators included:

  • Spontaneous comparison (T09 above 33 at 2.20)
  • Strategizing (T37 above 42 at 1.78)
  • General approach (T04 above 45 at 1.74)
  • Detail (T17 above 44 at 1.70).

A general observation is that those in healthcare and medicine are often academically astute.  They also tend to capitalize on knowledge and experience, tend to be factual, realistic and practically inclined and, probably given the risks involved, are relatively conservative in their approach – therefore the tendency to spontaneously compare elements and to be Reflective as opposed to Impulsive regardless of their characteristic Quick Insight. The majority show awareness of the integrated nature of their subject matter by apply systems thinking. However, action research results suggest that over time, career-related fatigue often seems to result in a more linear and structured approach in solving problems, as opposed to the systemic view required for medical careers.



Caution should be taken to generalize the results of this group of Politicians as the sample largely consisted of managers, director generals and others in executive roles in the public sector in South Africa. Key stylistic differentiators included:

  • Low Effectiveness in Explorative (T61 below 27 at 3.77)
  • Low Logical Reasoning (T67 below 31 at 3.50)
  • Low Reflective (T74 below 33 at 2.94)
  • High Random (T69 above 66 at 2.81).

Key IPC differentiators included:

  • Few and ineffective Card Movements (T00 below 27 at 4.30)
  • Inadequate Strategies to deal with Complexity (T25 below 28 at 4.20)
  • Low tendency to look for Logical Evidence (T29 below 29 at 3.66)
  • Undeveloped Analytical Skills (T40 below 32 at 3.16).

In terms of their results regarding cognitive complexity and ideal SST environment, the majority of these individuals showed Pure Operational and Diagnostic current levels of SST work complexity and a small proportion showed potential for Tactical Strategy work.



In constructing an evidence base for the CPP, which transcends the shortcomings of conventional Psychometrics, while also pursuing the further understanding of cognitive competence across groups, the contribution of qualitative and quantitative research techniques was also evaluated.

Based on approximately 50 formal empirical studies aimed at the validation of the CPP (as reported on in the CPP Technical Manual); countless smaller empirical  case studies aimed at quantifying and clarifying group-specific cognitive characteristics; qualitative research conducted via personal feedback to approximately 5,000 individuals where alternative test results were mostly available; and big data AI investigations, it seems that the most viable research approach in dealing with psychological data is one characterized by a focus on smaller, homogeneous and contextualized samples, where subjective insight and experience is capitalized on. Such case studies seem to produce clear and intuitively appealing results – more so than big data or even AI analyses do. It is, however, useful to combine big data AI analyses with a qualitative case-study approach to best understand the complexity and dynamics involved in cognitive functioning.

From the studies reported on in these blogs, it seems that the most effective and complex thinkers across career groups apply relatively similar cognitive approaches in that they capitalize on thinking skills “high up” in the holonic structure of the information processing model (IPM) – processes which are of an encompassing and inclusive nature. High level performers across career groups seem metacognitively aware and generally apply a Logical, Integrative and Learning orientation while capitalizing on the full spectrum of cognitive competencies.

However, it is also of interest to differentiate between the cognitive characteristics of various career groups. These differentiators are obscured by normal statistical practice which is subject to averaging effects. Artificial Intelligence (AI) techniques come to the rescue in this regard in that it reveals typical career group-specific cognitive indicators as discussed above. The AI findings seem to verify conclusions based on qualitative action research using a case study approach.


A summary of the findings

Cognitive characteristics of top performers on the CPP:


  • Integrative (T70)
  • Logical (T67)
  • Holistic (T64)
  • Learning (T72)

Information Processing competencies (IPCs):

  • Learning (T44, T15)
  • Metacognition (T37)


Table 4 summarizes all the key features associated with the career groups analyzed in the three blogs constituting this series.

Table 4: Career specific cognitive competencies

Call us on +27 11 884 0878