Validation of the enneagram test

Validation of the enneagram test C5s (Q98) – fred Lacroix (France) – Sept 2020

Intro (validation of the enneagram test)

Many enneagram tests exist in the world and in France. While some (but very few) such as RHETI have been scientifically validated, there is none in the French language, to our knowledge, which has been validated publicly and scientifically.

Following on from the work of Dr Anna Sutton (Dr A. Sutton, psychologist and lecturer at MMUBS University in Manchester, UK) on the scientific approach to the enneagram, we want to validate the C5s enneagram test in 6 steps. We present the first results to you.

The test was validated reliably in a study of 918 participants in Sept 2020. It shows a discriminating power of 78% which is excellent for an iodographic psychometric test like the enneagram. However, the test can be further optimized, which has been carried out at the beginning of 2021 in a version v.5.

Enneagram, in few words

Among the personality theories intended to understand, anticipate, manage and change behavior, we can cite the enneagram.

The enneagram offers a model of the nine-pointed personality structure. Thus these 9 branches (or enneatypes) correspond to 9 ways of defining oneself and of giving priority to a certain image of oneself. “I am upright and hardworking”, “I love and I help”, “I am successful and I am efficient” ….

Each of its dimensions is linked to one of the three centers of intelligence, the instinctive, emotional and mental centers. As in transactional analysis or the Schutz method, there is within the enneagram a real personality, called “essence”. This is often masked by the ego, or false personality. This ego manifests itself in compulsions, a sort of avoidance mechanism for the most unacceptable behaviors for each type (failure, anger, banality …).

The enneagram is based on a typology to build a theory of personality based on the idea of a more authentic ego, which it would be possible to come closer to by freeing oneself from its inappropriate defenses.

The objective is personal development as a source of better being in one’s life or of efficiency in the professional world. Thus the 3 steps are first to know oneself, then to esteem oneself and to assert oneself in its authenticity. Finally, the last step is to improve yourself by developing your potential, managing your emotions or relationships, being more creative ….

Step 1: Enneagram research review

This part will not be developed in this page. The subject has been already well managed through an article (Anna Sutton) that you can consult from this link.

Step 2: choice of the test and objectives

The choice of a test should be made on the basis of an objective. Our goal is to make an enneagram test available free of charge and anonymously. This test should help a person, who does not know the enneagram, to identify its enneatype.

Self-report tests have been shown to be good tools for personality psychology (Ashton M., 2014, Personality Psychology … p33, p63).

In a test, it is not the machine that decides for you what you are. The questionnaire considers you as free and responsible for your choices. The human being is not an exact science. This questionnaire should propose to make you think about the different facets of yourself and to discern the main one. Also, it is unlikely that this questionnaire will lead to a certainty. Rather, it should be seen as a support to shed light on the structure of your personality.

Understanding the different enneatypes through reading, coaching or meeting people should make it possible to finalize the identification of its enneatype. Indeed, without reflective practice, an emotion produces a look at the world. If I am not aware of my affects, I project them onto the world. I attribute to external causes what comes from my emotional state. Unaware of the effect I have on myself or on others, I will attribute my behavior and that of others to my and his personality.

 

We’re all different, yet somewhere the same

We all have personalities, we all observe the personalities of others. So we all communicate about the personality and we all make decisions based on the personality. In short, we all do the psychology of the human personality. Personality is therefore a subject that goes beyond scientific research. But that does not prevent personality from being studied as a human science.

 

The test C5s

The test used was created by C5s (Enneagram Institute, consulting firm, coaching and training organization located in St Priest (69) – tel. +33 (0)6 86 96 53 22) from existing tests such as the RHETI (v2.5 of Riso and Hudson, 1999), the envolutive enneagram test (Vidal J.Ph., 2014) and an empirical approach. C5s has been practicing the enneagram for over 20 years. fred Lacroix is an Enneagram practitioner (school of the ennéagramme envolutif). The general public website (owned by C5s) https://aprisme.blog is visited by more than 40,000 visitors per year.

Our test has 90 items with a likert 5 (Not at all, Rarely, Neutral, Often, Completely).

The chosen philosophy of the test is that of the ennéagramme envolutif. It seems to us the most relevant in relation to the objective set. A set of 8 to 10 characteristic behaviors of each enneagram, presented in a linear fashion, easily recognizable by a person with a minimum of knowledge of the enneagram. The objective is to promote the identification of a profile while answering questions. It is a bias considered to be helpful for a person who is discovering the enneagram for the first time and looking for its profile.

The test is free, anonymous if people want it. For statistical reasons, gender, age and country are required.

The questionnaire was first tested on 30 people who knew their enneatypes very well. It was the first step of the validation of our enneagram test. The nonlinear scoring (property of C5s) was adapted in order to obtain a discrimination of each enneatype of 100%. Thus the next step was to test it on more than 900 people who a priori did not know their enneatype.

Step 3: relevance of profiles on 918 participants

This step of the validation of the C5s enneagram test must make it possible to demonstrate that the use of the questionnaire makes it possible to easily bring out an enneatype.

918 people, over the period from March 18, 2019 to April 7, 2020, performed the test. They obtained their score in the form of a graph as shown below. Each person can download their results in the form of a pdf form.

68% of the respondents were female and the average age was 36 years old. The age varying from 12 years (3.3% of respondents were minors) to 98 years.

Split male / female

Split according to the age

7% of respondents had already taken the test and 1.5% have taken the test for someone else. Based on the criteria of age and name when present, duplicate or questionable results were excluded from the analysis. Bringing the analysis to 854 respondents (61 fewer results).

On the 854 results, 667 (78%) made it possible to bring out a distinct enneatype (greater than 4% of variance on the closest enneatype).

The distribution of the enneatypes is relatively balanced as shown in the following graph. Which is important for a personality test. Imagine that 80% of the respondents end up in a single profile. It is not very discriminatory and useful for the 80% of people. It should be noted that this distribution is not representative of a population. In fact, the participants carried out the test on a voluntary basis. We cannot exclude that a given ennetaype (here, for example Enneatype 2) may not be very partisan of researching his personality type on a website.

Split of the 9 enneatypes (C5s study: 918 respondants – 854 enneatypes)

Indistinct results

Indistinct results

However, for the 187 people without a distinct enneatype on the basis of the test, it is possible to identify their enneatype from the description of each enneatype. (For more information click on the link). 

  • Indeed 29% of the indisctinct results related to a weak difference between 2 very close enneatype scores (we could consider the wing approach of the enneagram).
  • On the other hand, 24% of the other indistinct results related to 2 enneatypes linked in the dynamic approach of the enneagram (integration and disintegration). Effectively in a phase of stress or serenity, the concept of the enneagram induces a change in the characteristic behaviors of an enneatype.
  • Finally, 25% of the indistinct results corresponded to enneatypes known to be easily confused on the basis of certain behaviors. In this case it was the Enneatypes E1-E3-E5 and E8 between them (dominant aspect), E4-E7 (creativity), E7-E9 (aspect of positive vision) and E2-E9 (taking care of others) .

However, it would be necessary to be able to be in contact with these people (78% give their contact items) in order to be able to extrapolate the 22% of uncertain results. What this study did not allow.

In conclusion, the objective of step 3 is validated and satisfactory with a discrimination rate of the C5s enneagram test of 78%. The distribution between the 9 enneatypes is homogeneous, varying from 5% to 16% .

Step 3 of the validation of the enneagram test

Step 4: relevance of items on 427 participants

We did this analysis on a part of the whole pool of respondants for convenient reasons (time and resources) as the number was enough to elaborate statistics.

The choice of questions for each enneatype is based on the relevant description of each of the enneatypes.

For example the question “life is often a tragedy. This is why I am often melancholy” should be discriminating for a type 4. This is confirmed by our analysis on 427 participants. Indeed the average score of the closest enneatype is almost half (48.6%). The standard deviation is less than 1 (σ = 0.96) which indicates that the other types have scores relatively close but far from type 4. Indeed the coefficient of variation is 0.42 (r). This question is therefore quite discriminating for a type 4.

But some questions like “I like what is beautiful”, just as relevant for a type 4 can nevertheless be solicited by the other enneatypes. Which is confirmed by our study. The average deviation from the nearest enneatype is only 15%. The standard deviation is 0.88 and the coefficient of variation is 0.13. What is considered in the psychometric world as relatively weak, and not very discriminating.

Average score by enneatype for 2 items relative to enneatype 4

Internal note: According to M. Ashton (Psychology of personality …, p22), the values (coef. Of variation) between -.2 and 0.2 are generally considered to be small. Values between -0.4 and -0.2, and between 0.2 and 0.4 are considered moderate in size. Values exceeding -0.4 and 0.4 are generally considered large.

 

Résults

The table below shows the overall scores by enneatypes. It confirms a relatively good discrimination by enneatype. With more discriminating enneatypes than others. This analysis confirms the good results of step 3.

average deviation from the closest profile

But it also indicates that the questionnaire can be improved in two different and complementary ways.

Effectively by reducing the questionnaire to 70 questions instead of 90. If we remove the least discriminating questions, our analyzes indicate that the overall average deviation can be improved. It would go from 31.1% to 34.6%.

On the other hand 4 questions can be changed. This required reconfirming that these new questions are much more discriminating. This was done in 2021 through version 5 (v5) of our Enneagram questionnaire.

Step 5: Correlation with a gold standard

If the Rheti test is the most recognized test, nevertheless it presents a sensitivity of around 70 to 80%. As in pharmaceutical testing, the goal is to take what is called a gold standard reference. With the enneagram, the Gold Standard is the oral identification of the enneatype through discussion (or coaching) with people. This correlation is the most difficult in terms of performance demonstration. Effectively we could consider that 2 tests having the same bias would correlate better (A respondant would make the same kid of reply through the 2 tests).

We took advantage of our last Enneagram training session to correlate the test with the validation of the enneatype of the participants.

After 48 hours spent together, the enneatype of the 11 people present was confirmed. To know:

  • type 1 = 1 person
  • type 2 = 1
  • the type 3 = 1
  • type 4 = 1
  • type 5 = 2
  • the type 6 = 0
  • type 7 = 0
  • type 8 = 2
  • the type 9 = 3 people

 

Results :

All people took the enneagram test at the start of the training. 9 out of 11  without having heard about the enneagram before the training. All without knowing with certainty their enneagram type.

5 people (45%) had their correct enneatype (types 1, 2, 3, 5 and 8) as soon as they get their score.

4 people (36%) had their correct enneatype after validation of their result (These people had 2 enneatypes that scored equally; one type 4, one 8 and two 9)

A type 5 was convinced he knew his enneatype. We can imagine that he answered the questionnaire with this influence (he confirmed it). However, his true enneatype stood out with a score of 26 versus 33 (type 8 and 1). These scores are already low, which demonstrates a lack of self-knowledge.

Finally the last participant always has a doubt about his enneatype. His result shows an Enneatype 1 with a Wing 9 when the validation of his Enneatype brings out an Enneatype 9 with a Wing 1 (the difference in score is 3, so very low). However, this result is considered as not correlating in our exercise.

In the end, a direct correlation of 45% in first intention and 81% in second intention emerges. Which is a very good result for the C5s Enneagram test.

Step 6: Correlation with a psychometric test

The C5s test has been correlated with a Big 5 test.

Here, again our test shows a good correlation with this psychometric test the most used in the world by scientists (Big 5 test or OCEAN model).

See the results in our article.

Discussion and objections

2 approaches oppose or complement each other in the psychology of the personality.

  • The idiographic approach is to study the personality of a single person and bring out the predominant traits. It is this approach that is used in the clinic by a psychotherapist for example. Thus the objective is not to bring out a personality trait (extrovert / introvert) but to identify an (un) adapted pattern in its entirety.
  • The nomothetic approach (Big 5, MBTI …) consists of studying certain personality traits in a large number of people and evaluating the links between these variables. It is therefore a statistical approach which makes it possible to validate certain trends. It is therefore an approach used mainly in social psychology and research.

The enneagram is closer to the idiographic approach showing the predominant features of a single person.

1st criterion: a psychological test must be useful

Today, the nomothetic approach (Big 5, OCEAN, NEO …) is the most used in the world in the field of research. The big 5 model makes it possible to isolate 3,125 (Likert 5) kinds of people. HEXACO (likert 7) could describe 118,000 different kinds of people. But its use requires an expert and a large statistical analysis system. Which makes it a tool reserved for science and not usable by anyone like you or me.

Scientists have tried to simplify the model into 3 or 6 personality types to make it more accessible (Asendorpf, 2003). Unfortunately most people have combinations of personality trait levels that do not match either personality type (Ashton & Lee, 2009; Costa, Herbst, McCrae, Samuels, & Ozer, 2002) . The argument of using personality traits, reduced to a few personality types, for a clinical or recruitment approach (NEO) therefore faces the same difficulties as the idiographic approach such as the enneagram.

In conclusion, in a personal development or clinical therapeutic approach, it is more interesting for a therapist, a recruiter or for oneself to use a simplified personality model, easily understandable and usable by everyone. It is therefore the utility criterion that confirms our choice of the enneagram model and the iodographic approach.

2nd criterion: A rigorous approach

The second criterion, out of the 3 major criteria when evaluating a theory of personality (Sutton, 2007) is the need to be scientifically rigorous. It means a theory that makes clear and testable predictions. A scientifically rigorous theory will not pretend that complicated things are simple. But she will make clear and verifiable predictions on these complicated things. This is what we tried to demonstrate with the 6 steps of the validation of the C5s Enneagram test.

The limits of self-assessment

The accuracy of a self-assessment, however, depends on several conditions. First; people need to know their behaviors, thoughts and feelings well enough. Second, people have to agree to share these behaviors openly.

Part of what makes the Enneagram so useful in the app is the fact that it describes often unconscious processes and motivations. Motivations to which we may not initially have easy access. Those who have not yet recognized their subconscious processes or default modes of operation simply will not be able to accurately account for it.

It is therefore to be expected that the enneagram questionnaire, in general, will exhibit lower reliability than the questionnaires which measure explicit personality. In our study, of the 918 participants, 22% could not bring out a distinct enneatype. This is consistent with the scientific studies already carried out (see the research review – step 2).

Finally, you have to distinguish what I am, what I think I am, what I want to be and what others see. What I think I am is not who I am. Consciousness is not knowledge. Hence the importance of self-knowledge through the gaze of others and objective knowledge of oneself through knowledge. What Carlo Moïso, Italian psychiatrist and psychotherapist and big name in transactional analysis calls the prince, the toad and the mask (social attitude).

In other words our responses can be guided by what we think is correct, or what we think or would like to be. This whatever the iodographic or nomothetic approach.

The dynamic structure of the enneagram

One of the strengths of the Enneagram model is that it is dynamic. That is to say that you are not locked in a box (a typology).

There is no scientific evidence yet to prove this. Therefore, the following remain assumptions and are interpretative.

However, the enneagram offers an influence of one or two wings. The wings are the enneatypes that precede and follow an enneatype. Thus a type 3 can be closer to 2 and another 3 closer to 4. We will speak of a dominant wing.

This may be the explanation for 29% of the non-exploitable profiles (the 187 profiles without distinct enneatypes) which present high scores but close between 2 consecutive enneatypes.

On the other hand, the model of the enneagram indicates that in times of stress or on the contrary of perfect serenity (or maturity) an enneatype will be influenced by a determined enneatype profile. For more explanation on this principle, see in detail the integration and disintegration profiles of each enneatype. In the study 24% of the non-exploitable profiles have a score close to the integration or disintegration profile, preventing the identification of a given enneatype.

However, every enneagram practitioner knows that it can be misleading to identify an enneatype if the person is, for example, in disintegration. It can easily be considered that among the 918 participants in the study a certain% was in the disintegration phase during the questionnaire. But this remains interpretative and is not proven by this study.

Participant satisfaction as a criterion of utility

Respondents to the C5s Enneagram test can complete a satisfaction questionnaire after reading their scores. Over a 16-month period (April 2019-Sept 2020), 101 people who took the Enneagram test responded to this questionnaire.

  • 54% of respondents considered the result as relevant or meeting their expectations.
  • Only 5 people (5%) did not find it relevant
  • Other people did not comment on the suitability of the test, which may be understandable in a results discovery approach

Here are the positive points and the points to improve of this validation of the C5s enneagram test

Positive points: design, free, short, relevant, innovative, downloadable

Points to improve: design, not friendly, too long, not relevant, don’t answer my expectations, report too short, not personnalized

To illustrate this satisfaction, we share 2 testimonials that speak for themselves:

  • “My enneagram results are much more like my personality to the point of making my son and my partner smile, especially on the character’s flaws ;-).”
  • “I have provided a brief report. But I find that normal in view of the work already done to make it available on the internet for free. Thank you”

This last step enables us to present in a scientific way a validation of our enneagram test

Conclusion of the validation of the enneagram test from C5s

The Enneagram is a great tool to help people develop their self-awareness. It gives us insight into our own behavior and motivations and those of others. Thus it facilitates personal and professional relationships. This allows us to respond to the other person’s intention rather than misinterpreting their behavior. It thus lends itself to team building and relationship development. The advice he provides for overcoming personality biases is an integral part of it.

The results of the validation of the C5s enneagram test show well differentiating enneatypes (918 participants). Thus the test is reliable, but we must be careful about its validity. Indeed we have no guarantee that it is really our type that has been identified. Nevertheless, the objective of making people think about the different facets of oneself is achieved.

This test is suitable for people who do not know the enneagram. Its structure, by enneatype, influences the respondent to identify its potential enneatype. Which is the primary objective of the test. But it reinforces the person already knowing his enneatype in the choice of his profile (pygmalion effect).

 

The C5s Test as the first step in discovering the enneagram

We have made the arbitrary choice that the self-report test is the first step in discovering the enneagram and its personality. The following steps are the detailed discovery of the different profiles by reading and interacting with a coach or ideally other people of different profiles, respecting the oral tradition of the enneagram.

A person, knowing his profile perfectly, has no reason to take the C5s Enneagram test. Unless she wants to challenge her self-knowledge, or the enneagram or our arbitrary choice of this type of test. In this case she must do it with an interlocutor, ideally an Enneagram coach, with his argument, otherwise at the risk of being labeled “hater”. Nevertheless all your comments are welcome, and you can contact us at any time.

We are proud to present this enneagram test validation !

To know more on :