Using personality type differences to form engineering design teams

Abstract

This paper argues for the greater use of personality type instruments such as the Myers-Briggs Type Indicator (MBTI) and the Keirsey Temperament Sorter II (KTS II), when forming engineering design teams. Considering the importance of teamwork in all aspects of education and industry, it is surprising that few universities in the UK use personality type information when forming design teams. This has led to many courses not getting the best out of their students, and more importantly the students not getting the most out of the teamworking experience. Various team formation methods are discussed and their relative strengths and weaknesses outlined. Normal personality type distributions in base populations are presented and compared with data from recent studies of engineering students, and the link between engineering, design and creativity is discussed. The results of this study have shown that the most important of the type preferences is the Sensing-iNtuitive (S-N) scale, with its proven link to creativity and learning styles. It is concluded that both engineers and designers have much in common, and a methodology of using personality type choice sets to select and form engineering design teams is proposed.

Introduction

Placing individuals into productive teams is one of the most important activities of any educational or business environment. However, it is also one of the least considered components. Much attention has been given to selection, performance measurement, retention and progression activities in the literature, but too little to the most fundamental task of them all - forming the team. It is little wonder that educational and business environments often fail to get the best out of their students and their employees, leading to frustration, recriminations, and poor performance. This failure, coupled with the fact that lecturers in higher education are finding themselves under increasing pressure, has resulted in group formation activities being hit-and-miss at best, and doomed to fail at worst. ‘University teachers have accordingly found themselves working harder and at the same time being required to be more business-like and more accountable.’ (Ramsden, 2003).

Research questions

• What is the range of team formation methodologies available?

• Which, if any, team formation methodologies work and why?

• Is there a better way of forming engineering design teams, than simply using traditional random selection methods?

• Where is the proof that they work?

Team formation methods

There are many alternative methods available to the individual lecturer when forming an engineering design team. Each of these methods has advantages and disadvantages; however most are fatally flawed due to the fact that they do not consider the strengths and weaknesses of the individuals involved and how to structure the mix to get the ‘best’ out of all team players. By ‘best’, we mean performing at the individual’s maximum output. If each individual member of the group is given a role which best suits their skills and knowledge, and if the team is structured such that each role is covered, but not duplicated, then we believe that the team will perform to its maximum capability. In doing so, the team will produce the best learning experience for the individuals, and also produce the best outcome — design, system or prototype.

In brief, the choices in selecting team-based groups after Race (2001) are:

1. Let the students choose their own teams.

2. Use the alphabetical class order in the register.

3. Use the university student number code order.

4. Select team members based on previous performance.

5. Select groups based on a heterogeneous mixture, i.e. sex, age, nationality, specialisation, etc.

6. Select a team leader and let them pick one additional member in turn.

7. Select team members based on sitting or standing position.

8. Select team members based on astrological ‘star sign’ or month of birth.

9. Select team members based on their Personality Type and/or Learning Style.

10. Issue coded labels to students, who then form groups based on the codes.

Of these methods, the most commonly used are (a), (b) and (c).

Allowing students to form their own teams results in formations based on friendships. Friends rarely work well in a team situation — the relationship is too cosy, things don’t get done and the atmosphere is too relaxed. This method also tends to alienate people based on differences in sex, age, nationality, race, religion, disability and social status (as does method (f)). Method (d) can be used to place the best students in the top (alpha) team and the worst students in the bottom (zeta) team. This method, whilst stimulating the bottom team to perform or die, can also have the effect of giving the top team a feeling of unrealistic superiority (the Apollo Syndrome) (Belbin, 1981) with its many negative implications. It is also obvious that you need historical data, which is not available in Year 1 classes, to employ this method. Using random selection methods such as (b), (c), (g), (h) and (j) will produce average results at best. The only methods that will guarantee above average results are (e) and (i). Heterogeneous mixtures of students usually perform well due to their blending of expertise, experience and perspectives. However, even apparently well-balanced teams such as these sometimes fail to perform due to no obvious reason.

Clearly it would be helpful to the engineering lecturer to be able to understand the personality, motivation, strengths and weaknesses and learning style of the students before forming the team. This can best be achieved by using a questionnaire style instrument to extract this information. The two most popular methods are by using instruments such as the Index of Learning Styles (ILS) developed by Felder and Soloman in 1991 and by using the Myers-Briggs Type Indicator, MBTI^® developed by Myers-Briggs some 50 years earlier. These two models share many facets and are complimentary in many respects. This article will concentrate on the use of the MBTI instrument due to its wide proliferation and its large user base. For an overview of Learning Styles please refer to the excellent article by Felder & Brent (2005).

Belbin Team Roles

As already mentioned, the work of Dr R. Meredith Belbin and his team of researchers during the 1970s were influential in terms of understanding management teams in a business setting. His work, over nine years at the Henley Management College, investigating how teams function, culminated with his theory of Team Roles. This theory is based on nine team roles which can be broken down into:

Action oriented roles: Shaper, Implementer, and Completer Finisher.

People oriented roles: Co-ordinator, Teamworker and Resource Investigator.

Cerebral roles: Plant, Monitor Evaluator, and Specialist.

This work has led to a series of business orientated books and the e-interplace^® software package incorporating the Belbin Self-Perception Inventory (SPI) — a psychological profiling tool for the individual team member (Belbin Team Roles, 2007). There is a lot of overlap between the work of Belbin and that of Myers-Briggs, however, in so far as the Belbin method concentrates on the world of commerce, we shall concentrate on methods primarily for the educational setting.

The Myers-Briggs Type Indicator

In its basic form the Myers-Briggs Type Indicator, MBTI^® is a 93-item instrument and the most widely known psychological typing tool in use today. It was estimated by Pittenger (1993) that over 2 million copies were being sold annually in 1992. This has now risen to an estimated 3.5 million annual sales worldwide (OPP, 2007). The MBTI is available in more than 21 languages.

The MBTI has been around in one shape or another for over 60 years, and has been used in a number of occupational settings. No other psychological testing instrument has been subjected to as many tests of reliability and validity (Myers & McCauley, 1985). However, it is fair to say that it has detractors as well as supporters (Mathews, 2004).

Douglass Wilde, a Research Professor in Design at Stanford University, who has used the principals of psychological type with great success to form engineering design teams over the last 20 years, has stated that:

About a hundred million people have used the MBTI, at least three-quarters of them agreed strongly with all four results. Just about everyone agrees with at least three. The other quarter may find the MBTI preference clarity concept useful for understanding uncertainty, if not eliminating it.

(Wilde, 2003)

Historical development of the MBTI Instrument

The roots of type theory can be traced back to the turn of the 20th century and the work of Carl Gustav Jung (1875–1961), the Swiss psychiatrist and contemporary of Sigmund Freud and Alfred Adler. Jung and Adler disagreed with Freud with regards to the importance of sexuality in causing psychological problems and therefore split with him in 1912. Jung’s seminal work, Psychological Types was published (in German) in 1921 after almost twenty years of practical research work (Jung, 1971).

In her excellent book, Gifts Differing, the co-founder of the MBTI, Isabel Briggs Myers (1897–1980) describes how, together with her mother Katherine Cooks Briggs they extended Jung’s theory of personality types, adding two important aspects:

1. The existence and roles of the auxiliary processes.

2. The addition of the Judging (J) and Perceiving (P) preference.

Thus, Jung’s eight pairs (2³) became the Myers-Briggs 16 types (2⁴) (Myers & Myers, 1995). As can be seen from Figure 1, this consists of four dichotomies, the interaction of these, giving the 16 individual types, i.e. ISTJ, ENFP, etc. The abbreviations in Figure 1 are used throughout the paper.

The development of the MBTI and its acceptance took many years of hard work by Isabel Myers, herself, not a qualified psychologist or statistician. The spur for this development was World War II, where most males were called to serve in the US military, thus forcing many women into industrial jobs for which they were not familiar or even well suited. Thus the origin of the MBTI dates from the summer of 1942, to quote Myers “…to do something that might help people understand each other and avoid destructive conflicts.” (Myers & Myers, 1995)

Throughout the 40s, 50s and 60s, Myers collected and developed an item pool of data on personality type, mainly using students from schools and colleges. The first MBTI manual was published by the Educational Testing Service in 1962. In 1975, the publication of the MBTI was transferred to Consulting Psychologists Press (CPP), with the Center for Applications in Psychological Type (CAPT), organised as a service for MBTI development, research and training. The CAPT maintains a research database of MBTI published works which currently holds over 9,700 records.

The MBTI saw rapid growth and acceptance throughout the 80s and 90s and has grown into a multi-million pound industry. The MBTI was developed specifically as a tool for the non-psychiatric population, and is therefore inherently benign. As a founding principle, no one type is any better or worse than any other and the testee has the final say as to his or her type designation.

Figure 1 The MBTI dichotomous pairs

The MBTI Preferences

Isabel Myers determined that the sixteen personality types could best be shown using a standard Type Table as shown below:

Table 1 The standard MBTI Type Table and the UK general population distribution (%) (OPP, 2007)

ISTJ 13.7%	ISFJ 12.7%	INFJ 1.7%	INTJ 1.4%
ISTP 6.4%	ISFP 6.1%	INFP 3.2%	INTP 2.4%
ESTP 5.8%	ESFP 8.7%	ENFP 6.3%	ENTP 2.8%
ESTJ 10.4%	ESFJ 12.6%	ENFJ 2.8%	ENTJ 2.9%

Note 1. The dominant processes of each type are underlined in the table above.

Note 2. The sample consisted of 1,634 people living in the United Kingdom. 748 (46%) were male and 865 (54%) female. 94% of the sample were white and 6% came from other ethnic groups. Ages ranged from 16 to 65 years with 50% aged between 30 and 50. The sample included people of all educational levels. 69% were currently employed, with 40% at supervisory/first level management or above. A wide range of industry sectors was represented.

The MBTI instrument sets out to gain answers to the four dichotomies mentioned above, in broad terms these refer to:

Myers stated that the interaction of these orientations, functions and attitudes are what makes up the personality types. Type theory describes how, in a normal person, these functions are developed as we mature, with mastery of the dominant function, adequate but not equal development of the auxiliary, and eventual use of the third and fourth functions to an acceptable level (Wankat & Oreovicz, 1993). Further analysis and in-depth understanding of each of the 16 types can be gained by reference to the CAPT, CPP and Myers-Briggs websites (CAPT, 2007; CPP, 2007; Myers-Briggs, 2007).

The Keirsey Temperament Sorter II (KTS II)

A contemporary of Isabel Myers, David Keirsey has been very successful in his own right with his personality type system which he calls the Keirsey Temperament Sorter II, this consists of a 70-item instrument that has only two possible responses, and is available as an online test. In his bestselling books Please Understand Me and Please Understand Me II, Keirsey (1998) follows the MBTI tradition of using 16 types, however, this is where he parts company with Myers. Keirsey regards the S-N scale as the most important as it relates to the cognitive perceiving function, and in this respect he has gained a lot of followers in the area of learning and teaching styles (Felder & Brent, 2005). From his analysis, Keirsey orientates the 16 types into a tree-like structure configuring types into four Temperament groupings, which he calls Guardians, Artisans, Idealists and Rationals (see Figure 2).

To each of the 16 types, Keirsey gave an operational name, i.e. Supervisor, Inspector, Mastermind, etc. The shading in Table 2 shows the Temperament groupings (see Fig. 2).

Keirsey argues, with some justification against the use of the ‘Function Typologies’, i.e. the grouping of types based on their dominant function and towards his vision of ‘Intelligence Typologies’, i.e. Temperament groupings. Both the MBTI and KTSII have found widespread use within many fields of education and industry.

Normal type distribution in base populations

Before looking at particular studies involving engineering student populations in the UK, US and Asia, it is important to give the reader an understanding of how personality types are distributed in normal general populations. Given the length of time that the MBTI instrument has been available it is somewhat surprising that the data for the normal general populations in the US and UK were only developed in 1986 and 1998 respectively. Normal population data for many other countries in the world, including China and India does not currently exist.

Figure 2 Keirsey’s tree structure of Temperament groups

Table 2 Keirsey’s type names and their association with the MBTI type table

ISTJ Inspector	ISFJ Protector	INFJ Counsellor	INTJ Mastermind
ISTP Operator	ISFP Composer	INFP Healer	INTP Architect
ESTP Promoter	ESFP Performer	ENFP Champion	ENTP Inventor
ESTJ Supervisor	ESFJ Provider	ENFJ Teacher	ENTJ Fieldmarshal

Guardians

Artisans

Idealists

Rationals

Table 3 Distribution of dichotomous population preferences in the UK, US and Korea (CPP/OPP, 2007; Sim and Kim, 1993)

		Guardians		Artisans		Idealists		Rationals
UK		49.4%		27.1%		14.0%		9.5%
1998	E	52.6%	S	76.5%	T	45.9%	J	58.3%
N=1,634	I	47.4%	N	23.5%	F	54.1%	P	41.7%
		Guardians		Artisans		Idealists		Rationals
US		46.4%		27.0%		16.5%		10.4%
2003	E	49.3%	S	73.3%	T	40.2%	J	54.1%
N=500,000	I	50.7%	N	26.7%	F	59.8%	P	45.9%
		Guardians		Artisans		Idealists		Rationals
KR		50%		25%		11%		13%
1993	E	41%	S	75%	T	63%	J	63%
N=13,308	I	59%	N	25%	F	37%	P	37%

From the data above it is clear that the UK, US and Korean populations are broadly similar in terms of the Keirsey temperament distributions. The UK population is, however, slightly more Extravert than the US population — this is in contrast to many earlier studies which reported that the US population was more Extravert (75%) (Wankat & Oreovicz, 1993). All three populations exhibit preferences for Sensing and Judging, with Korea being the most Judging. In terms of the T-F dichotomy, Korea is very different to either the UK or US, with a strong preference for Thinking.

It should be noted that in terms of the Thinking-Feeling dichotomy it is well documented that Men prefer Thinking over Feeling (T = 60%, F=40%), whereas Women prefer Feeling over Thinking (F = 75%, T = 25%).

MBTI and the Chinese Market

The Chinese Mandarin version of the MBTI was first translated in 1994 (Miao, Huangfu, Chia and Ren, 2000) and only a few studies have been reported since then (Yao, 1993; Broer & McCarley, 1999; Osterlind, Miao, Sheng and Chia, 2004; Sharp, 2004; Hu, 2005). Surprisingly, the CAPT database has only 16 references to the term ‘Chinese’ and 7 references to ‘Taiwan’. Considering the importance of the Chinese economy in the next decade, more research is needed in this area. According to the latest Engineering UK Report (2006), China has experienced a 124 per cent increase in the number of science, technology, engineering and mathematics degrees over the past decade to 350,000 per year.

Results from the few Chinese studies (Taiwan, Hong Kong and China) that have been published have reported:

• Introversion slightly preferred over Extraversion (52–64%).

• Sensing over-represented (60–85%).

• Thinking over-represented (61–93%).

• Judging over-represented (70–85%).

Overall this gives the national identity of the Chinese population as ISTJ — inferring a group who are “Serious, quiet, earn success by concentration and thoroughness… make up their own minds about what should be accomplished and work towards it steadily, regardless of protests or distractions.” (Myers et al., 1998).

ISTJs also make up the highest proportion in the UK population and the second highest in the US population. From Myers et al. (1998) ISTJs make very good leaders, are well organised and have an entrepreneurial spirit. Due to their diligence and attention to detail they also find their way into the engineering professions in large numbers.

Use of the MBTI in Engineering Education

Over the years, studies in nearly every area of engineering have investigated the ubiquity of using the MBTI and KTS II instruments. These range from mechanical and electrical engineering, to chemical engineering and many others (McCaulley, 1990; Rosati, 1998; Jensen, Murphy and Wood, 1998; O’Brien, Bernold and Akroyd, 1998; Stone and McAdams, 2000; Jensen, Wood and Wood, 2003; Felder, Felder and Dietz, 2002; Felder and Brent, 2005; Lester, Schofield and Chapman, 2006).

One of the biggest studies into engineering students was conducted by a consortium of the American Society for Engineering Education (ASEE-MBTI) involving eight engineering schools and 3,784 students during the period (1980–87). All subject specialisms were covered and of special interest to us was Mechanical Engineering. CAPT is the Center for Applications in Psychological Type.

Comparing and contrasting the data in Tables 3 and 4 shows that Idealists are slightly over-represented. Rationals (intuitive types) are vastly over-represented by a factor of two or three, and that there is a preference for Thinking and Judging. This data has been confirmed by many later studies.

Other studies have investigated personality typing in Psychology (Chamorro-Premuzic & Furnham, 2003), Economics (Ziegert, 2000), Multimedia Engineering Design (McKenna, Mongia and Agogino, 1998), Software Engineering (Layman, Cornwell and Williams, 2006), Microelectronics (Pearson, Bell and Croley, 2003), Electrical Engineering (Chang & Chang, 2000), Post-traumatic Stress Disorder (Otis, 2005), Health Professionals (Hardigan, Cohen and Janoff, 2005), Pharmacy Students (Shuck and Phillips, 1999), Dentistry (Baran, 2005) and Career Counselling (Gruber, 2000), Doctors and Patients (Clack, Allen, Cooper and Head, 2004).

Table 4 Comparison of Mechanical Engineering students with Engineering Professionals (McCaulley, 1990)

		Guardians		Artisans		Idealists		Rationals
ASEE-MBTI		40.16%		19.12%		6.94%		33.79%
1980–87	E	45.56%	S	59.27%	T	80.32%	J	64.1%
N=518	I	54.44%	N	40.73%	F	19.68%	P	35.9%
		Guardians		Artisans		Idealists		Rationals
CAPT-ME		44.48%		14.29%		19.48%		22.08%
86	E	46.76%	S	58.44%	T	70.13%	J	62.34%
N=77	I	53.24%	N	41.56%	F	29.87%	P	37.66%
		Guardians		Artisans		Idealists		Rationals
CAPT-Eng		38.25%		14.40%		19.98%		27.38%
1986	E	47.67%	S	52.64%	T	63.59%	J	60.45%
N=986	I	52.33%	N	47.36%	F	36.41%	P	39.55%

For an overview of how these disciplines relate to distributions on the standard MBTI type table we refer readers to the Atlas of Type Tables (Macdaid, McCaulley and Kainz, 1986) and Gifts Differing (Myers & Myers, 1995).

To put the MBTI results into context, it may be helpful to note extreme values of the preferences and how they relate to certain professions:

Table

Student Leaders	Architects
E - 84%	E - 30%
I - 16%	I - 70%
School Bus Drivers	Fine Artists
S - 97%	S - 9%
N - 3%	N - 91%
Management Consultants	Priests
T - 92%	T - 20%
F - 8%	F - 80%
Retail Store Management	Food Service
J - 92%	J — 48%
P - 8%	P — 52%

Almost every reference quoted above refers to studies conducted in the US or Canada, very little work has been conducted in the UK within Higher Education (Lester, Schofield and Chapman, 2006).

Creativity and the design student

There is a scarcity of MBTI information on Design students. A study by Stephens (1973), though dated, provides some interesting data in terms of the strong relationship between creativity, introversion and intuition. This research finding is further supported by earlier work from Guilford (1966). Other studies have used the MBTI, KTS II and Gough Creativity Index (GCI) to further support this correlation (Wilde, 2004).

A work by Durling (1996) also confirms the link between creativity and intuition, and goes on to discuss the problem-solving strategies of designers. Interestingly, Durling orientated the standard MBTI Type Table along the lines of the dominant functions as stated in Table 3. By plotting data for business managers, engineers, architects, artists and designers together with the general population he was able to show the broad disposition of occupational groups on the modified Type Table. Again much of the basis for this data relates to the US base populations, which have small sample sizes and are outdated. Durling reported a study of 71 students from product design, interior design, graphic design, furniture design and design marketing, based at two UK universities. The distribution of type for this sample is shown in Table 5 (re-orientated for consistency).

Table 5 Type table distribution of 71 UK design students (Durling, 1996))

ISTJ 1.4%	ISFJ 1.4%	INFJ 2.8%	INTJ 5.6%
ISTP 2.8%	ISFP 1.4%	INFP 7%	INTP 5.6%
ESTP 2.8%	ESFP 7%	ENFP 15.5%	ENTP 26.8%
ESTJ 4.2%	ESFJ 0%	ENFJ 5.6%	ENTJ 9.9%

The sample shows that Design students have a strong preference for iNtuition (79%) and a strong preference for Perception (69%). Over a quarter of the sample were from one type - ENTP. People with the ENTP preference have been described as “Warmly enthusiastic, high spirited, ingenious and imaginative. Able to do almost anything that interests them. Often rely on their ability to improvise instead of preparing in advance.” (Durling, 1996)

Engineers and designers

As can be seen from the data presented earlier, Engineers and Designers have a certain amount in common; they both have strong levels of iNtuition (40–47%) and (79%), when compared to the normal population (24–27%) which enables both groups to be able to solve problems creatively and intuitively. However, even these groups don’t come close to Fine Artists (iNtuition, 91%).

Both groups design, develop, validate and create products and services for use by people. However, many people would assume that by their very nature designers are more creative than engineers — we believe that this is a common myth. It is true to say that designers tend to operate on a divergent model theory — many solutions (possibilities); whereas engineers tend to operate on a convergent model theory — one best solution (probabilities). It is only when we combine these models together to form the divergent-convergent model do we get the optimum solution, which we might call the Engineering Design solution.

Team formation using personality type

By analysing the standard type table and its associated type profiles, it has been possible to group common personality traits into five tiers (tiers 0 to 4 in Figure 3). From this, we have concluded that there is a range of types best suited to both the dual roles of Engineering and Design.

Figure 3 Exploded type table showing choice sets for Engineering Design

From analysis of the 16 MBTI personality types and their dominant features, we suggest that the following eight types in tiers 0 and 1 would be the most suitable for Engineering Design teams. Ideally, the team leader role should be chosen from either the ISTJ or ESTJ personality types (tier 0) due to their ability to lead, organise, control, motivate and coordinate team activities:

Tier 0 (Team Leadership):

ISTJ — Inspector — Pragmatic, detailed, organised.

ESTJ — Supervisor — Practical, realistic, decisive.

Note: we advise not to place ISTJs and ESTJs within the same team, due to the inherent potential for a power struggle to develop.

Tier 1 (Development Team):

ISTP — Operator — Hands-on, problem solver, curious.

ESTP — Promoter — Problem solver, mechanic, adaptable.

INTJ — Mastermind - Original thinker, creative, organised. (Possible Leader)

INTP — Architect - Scientist, logical, analytical.

ENTP — Inventor — Ingenious, outspoken, resourceful.

ENTJ — Fieldmarshal — Frank, decisive. (Possible Leader)

These choice sets and their relationship to Keirsey’s temperaments can be seen in Table 6.

To show how this might work in practice, data from Durling (1996) and data supplied by the CAPT databank (Macdaid et al., 1986) has been combined in Table 7.

Analysis of Table 7 shows that by using only Tiers 0 and 1 selection sets (eight types) which encompasses 46% of the general UK population, it would have been possible to have selected 60% of the Design students and 70% of the Mechanical Engineers in this example. Further selection sets could be employed when necessary to enlarge the pool or act as a filtering mechanism for team formation purposes. We accept that a lot will depend on the size of the team being formed and the availability of the personality types in the pool. However, it is also clear that certain types (ISFP and ESFP) would make poor Engineering Designers.

It is common for teams to range from two to eight students in an educational environment, and perhaps up to ten in a business context. Clearly, a balanced team consisting of a range of personality types is the desired goal.

Discussion and conclusions

For the first time, accurate data on various aspects of personality type, such as population preferences, temperament statistics and type table data, has been brought together within one source.

Table 6 Choice sets for the selection of Engineering Design teams

rowsep="1" colsep="1">	rowsep="1" colsep="1">Guardians	rowsep="1" colsep="1">Artisans	rowsep="1" colsep="1">Idealists	rowsep="1" colsep="1">Rationals
Tier 0	ISTJ — Inspector ESTJ — Supervisor
Tier 1		ISTP — Operator ESTP — Promoter		INTJ — Mastermind INTP — Architect ENTP — Inventor ENTJ — Fieldmarshal
Tier 2	ISFJ — Protector ESFJ - Provider		INFP — Healer ENFP - Champion
Tier 3			INFJ — Counsellor ENFJ - Teacher
Tier 4		ISFP — Composer ESFP - Performer

Note: The Tier 0 group consists of half the Guardians, and is the grouping from which the Team Leader role should ideally be selected.

Table 7 Comparison of Mechanical Engineers and Design Student type selection using only Tiers 0 & 1

rowsep="1" colsep="1">Rank	rowsep="1" colsep="1">Mechanical Engineers	rowsep="1" colsep="1">% Cumulative	rowsep="1" colsep="1">Rank	rowsep="1" colsep="1">Design Students	rowsep="1" colsep="1">% Cumulative
1 2 3 4 4 4 7 8 8 8 8 8 8 14 14 16	ISTJ (19.5%) ESTJ (18.2%) ENTP (9.1%) ISTP (6.5%) INFP (6.5%) INTJ (6.5%) ENFJ (5.2%) INTP (3.9%) INFJ (3.9%) ESTP (3.9%) ENFP (3.9%) ESFJ (3.9%) ISFP (3.9%) ENTJ (2.6%) ISFJ (2.6%) ESFP (0%)	19.5 37.7 46.8 53.3 - 59.8 - 63.7 - 67.6 - - - 70.2 - -	1 2 3 4 4 6 6 6 9 10 10 10 13 13 13 16	ENTP (26.8%) ENFP (15.5%) ENTJ (9.9%) INFP (7%) ESFP (7%) INTP (5.6%) INTJ (5.6%) ENFJ (5.6%) ESTJ (4.2%) ISTP (2.8%) ESTP (2.8%) INFJ (2.8%) ISTJ (1.4%) ISFJ (1.4%) ISFP (1.4%) ESFJ (0%)	26.8 - 36.7 - - 42.3 47.9 - 52.1 54.9 57.7 - 59.1 - - -

Some recent data on the distribution of extroversion in the US refutes earlier research which suggested that the US population was 75% Extroverted, we now know that there are more extroverts in the UK (53%) than in the US (49%), and that both populations are remarkably similar in other respects.

The authors of this paper argue that any of the methods described (MBTI, Keirsey, Belbin or Learning Styles) which seeks to group teams based on an understanding of their underlying personality traits, skills and knowledge is better than any of the alternative random selection methods.

As an example of its success, the work of Wilde (2003) at Stanford University stands out, due to his method of adapting both the MBTI and KTS II, formulating his own method of team selection. This has proved to be a highly successful strategy in terms of the annual National Lincoln Prize awards in the US:

1. No selection strategy (27% of awards won).

2. Preference information guidance (57% of awards won).

3. Creative roles used (73% of all awards won).

Further work is ongoing at Stanford to develop this system into an even more effective tool.

For the average lecturer, the use of any of these methods will come down to how much time they have to prepare, and how much money is available to spend. The answer to both these questions is usually little or none. Therefore, there either needs to be a fundamental change in the philosophy of team selection and its importance to the learning environment or cheaper alternatives need to be found.

In the search for a solution to this problem we have been experimenting with freely available pseudo-MBTI questionnaires which are available over the internet. One of the best that we have found so far is from Similarminds (2007). Of course, going down this path provides a solution to the problems of time and money, but, this is at the cost of reliability and validity, which are largely unknown. More empirical studies in this area are clearly needed.

Much of the original research into personality type was conducted in the 50s, 60s and 70s and therefore can be regarded as somewhat out of date. However, ongoing work by various agencies such as the CAPT, CPP Inc and the OPP Ltd has rejuvenated some of these data sets with information not available to earlier research studies.

The Chinese (South East Asian) economy is clearly set to rapidly expand over the next decade, and as such MBTI data, not currently available, will be extremely useful for both education and business purposes alike. From our research it is clear that the Chinese populations are different in several respects when compared to western nations. The Chinese have stronger preferences for Introverted, Sensing, Thinking and Judgment. This makes them good in organisation, detailed thinking and control, but not so good in terms of creativity, openness, warmth and perception.

Having analysed the sixteen MBTI personality types for their relevance to the fields of engineering and design, we can conclude that there are eight types which are best suited to the area of engineering design. By ranking these in order of selection preference (Tiers 0–5) we have effectively provided the educator or business leader with the tools to select a successful engineering design team, consisting of a complimentary set of skills, and managed by a strong leader type (ISTJ or ESTJ).

There is clearly much work to be done in understanding engineering designers, and the way that they think, however, there are a lot of associations to be found between creativity, intuition, learning styles and personality types.

In conclusion, the authors would like to encourage all engineering educators to make greater use of type theory when selecting and forming engineering design teams and delegating leadership roles. The benefits of this, we hope, will be recognised by the mainstream engineering education community, and just as importantly by our industrial colleagues (Dym et al., 2005).

Acknowledgements

The authors would like to thank Ms. Jamelyn R. Johnson, Coordinator of Research Services at the Center for Applications of Psychological Type for preparing the Selection Ratio Type Tables (SRTT) used in this research.