OPP® - international business psychology consultancy

Resources - Other research

What Are 600,000 People Like?

Many MBTI users tuned in to watch 'What am I Like? The Personality Test', a programme produced by Mentorn, and screened on BBC1. We are now delighted to bring you the results taken from the BBC website version of the short quiz (available via www.bbc.co.uk/science/humanbody/mind/surveys/whatamilike/). We have now been able to download the data of 608,996 people; here we give some of the preliminary results from our analysis. Simply click on the links below to explore the findings:

Who completed the Quiz?
What are the Type Preferences of those who completed the Quiz?
Type and Demographic Factors
Gender
Age
Occupation
Education
Psychometric Properties of the Quiz
Conclusions

Who Completed the Quiz?

Given that the quiz was on the BBC website, it is not surprising that the majority of the sample (490,353 people, 80.5%) were brought up in the UK. Significant numbers were, however, brought up in a wide range of different countries [1]. The table below shows all groups of more than 1,000 people:

The remainder of the analysis described in this report was carried out on the UK sample only. This group:

• Was 41% male and 59% female
  
• Had a very wide age range, from children to people in their 90's, with a median age of 29 and a mean age of 30.5
  
• Had a range of educational backgrounds:
  
• 8% no qualifications
  
• 29% GCSE or equivalent
  
• 25% A Level or equivalent
  
• 38% Degree Level
  
• Consisted of people in a wide range of occupations. Students (27%), those in administrative jobs (16%) and IT (12%) were the largest groups.

Back to top ^

What are the Type Preferences of Those who Completed the Quiz?

The Type table below shows the percentage of each type preference in the UK group. ISTJ ("Realist") is the most common type, followed by ISFJ and then ENFP. The SSR (Self-Selection Ratio) shows how this compares to the reported type of the UK populationM[2]. The larger the SSR, the more over-represented that type is. The following types are over-represented compared to the UK population:

• INTJ (Mastermind)
• INFJ (Counsellor)
• INFP (Idealist)
• ENTP (Big Thinker)
• INTP (Strategist)

The following types are under-represented:

• ESFJ (Provider)
• ESFP (Performer)
• ESTJ (Supervisor)

The table at the right shows that in the quiz group, there are more people with a preference for Introversion than for Extroversion, Sensing than Intuition, Feeling than Thinking, and Perceiving than Judging. Compared with the UK general population, this group contains 21% more Intuitives, 10% more Introverts, 10% more Perceivers and 2% more Thinkers. This suggests that there may be a type bias in sampling – those people with a preference for Intuition, Introversion, and Perceiving may be more likely than others to go on the internet and/or once there be more likely to choose to visit the BBC website and complete a personality questionnaire. Similar findings have been shown for Intuition and Introversion in other studies.[3]

Back to top ^

Type and Demographic Factors

Gender

The table below gives the percentage of each type in the group as a whole, and for males and females separately. This shows that for this group:

A difference between males and females on the T-F dimension is often found in MBTI® studies; the relatively large differences here on the other dimensions are much more unusual. This may be due in part to the content of the quiz questions, which are arguably more behavioural in nature than the questions in the MBTI® itself. With questions relating to shopping, organising others and so on, they may be more affected by the societal roles both men and women are often expected to fulfil.

Back to top ^

Age

The table below shows the average age for each type dichotomy. The biggest difference can be seen on Sensing-Intuition, with those of a Sensing preference being on average 6 years older than those with a preference for Intuition.

Back to top ^

Occupation

There are clear differences in the Type preferences of people in different occupations, as illustrated by the following graphs:

For example over 70% of retired people showed preferences for Judging.

Back to top ^

Education

The table below gives the percentage of each type in the group as a whole, and for those at each level of education separately. Note that this table is based on those aged at least 21.

From this table it can be seen that:

• Those with no qualifications are more likely to have preferences for Extraversion than those in other groups.
  
• There is a higher percentage of Sensing types amongst those educated to GCSE level and a higher percentage of Intuitives among those educated to degree level.
  
• There is a higher percentage of Thinking types amongst those educated to a higher level.
  
• The Judging-Perceiving dichotomy shows a less clear relationship with educational level.

These findings are broadly in line with other research.

Back to top ^

Psychometric Properties of the Quiz

The internal consistency reliability of the quiz, as completed on the BBC website, is similar to that found in the development stages of the questionnaire, as shown below:

Although internal consistency reliability is rather lower than would be ideal for a psychometric questionnaire, it is very respectable for a short quiz of this type. Nevertheless, it is clear that the quiz is no substitute for the full MBTI® instrument.

Back to top ^

Conclusions

This study highlights both the advantages and the disadvantages of research using open-access questionnaires on the Internet. On the positive side, very large data sets can be collected on short tests, particularly via popular sites such as www.bbc.co.uk. On the negative side, longer questionnaires will be less likely to be implemented (and will have lower response rates) and of course the sample is unlikely to be representative of the general population. Our sample appears to be skewed towards well-educated female Intuitive, Introverted and Perceiving types, who are students or working in administration or IT. This is of course of some interest in itself, as a description of who may be more likely to complete a questionnaire of this sort.

Looking at the quiz itself in more detail, it is encouraging that it does show, for its length, reasonable internal consistency reliability. It would also of course be extremely interesting to know how well "quiz" type related to real Best Fit type. If you have completed the quiz, and know your MBTI® best fit type, please do send an email to community@opp.eu.com containing your quiz type, your actual best-fit type, and your gender. If we have sufficient responses, we will analyse this data and present our findings in a future OPPcommunity article.

The large gender differences found in this study, when compared with the much smaller gender differences found in general population samples with the MBTI itself, are a useful illustration of the fact that behaviour, what one actually does, is affected by culture as well by personality type. Further analysis of the data will investigate the way men and women answer individual questions, and the interaction of gender with age and educational level on responses to the questionnaire. The impact of culture will be further examined by bringing in the data of the 118,000 people in the sample who were brought up outside the UK.

The findings in relation to age are unsurprising, although once again the differences are quite large. Further analysis will look at age in more detail, for example looking at age and whole type, and at the relationship of type and age between childhood and adulthood.

Educational level and occupation also show some interesting differences. Further analysis here will look at the effects of type dynamics and whole type.

In summary, while the short quiz is not a substitute for the MBTI, it does present the opportunity for useful and interesting research. As we carry out further analysis, we will present our findings in further OPPcommunity articles. We also expect to have a further data download from the BBC in 6 to 9 months time, and will update you with any new findings then. In the meantime, do try out the quiz, if you have not already done so – and do let us know how your "quiz" type compares to your MBTI type.

Please also feel free to give us your suggestions for any analysis you would like to see on this data.

Back to top ^

[1] 187 different countries were represented in the sample.

[2] Kendall, E (1998). Myers Briggs Type Indicator (European English Edition) Step I Manual Supplement.

[3] See www.personalitypage.com/demographics.html.

[4] Kendall, E (1998). Myers Briggs Type Indicator (European English Edition) Step I Manual Supplement.