About the Classical Enneagram Test
This only applies to the (free) classical enneagram test on this site.
Fine tuning for accuracy
To make this test more accurate, statistics were used.
Each question was checked for how it was graded on average by each Enneagram type. The questions that didn't perform well, were changed or replaced. About half of the statements are not the same anymore as when the test was first put on the web. Almost all seem to do quite well currently.
Of course it is not possible to know for certain if someone is typed correctly. To achieve at least some degree of certainty for the sake of statistics, the Enneagram type as tested was compared to the type that was entered in advance to taking the test. If these matched, it was assumed that the type was correct, and the corresponding statistical data was given greater weight when determining which questions needed to be changed.
As an example of the statistics, these are average scores for the type 3 statement:
"I am competitive and ambitious but I do not think of myself as cut throat."
Type 2: 0.478
Type 3: 0.851
Type 4: 0.392
Type 5: 0.358
Type 6: 0.430
Type 7: 0.525
Type 8: 0.626
Type 9: 0.300
Obviously, the score for type 3 is highest.
Reducing the number of questions
People find it boring when tests have lots and lots of questions, especially if they don't seem to be very relevant. But to make a test like this one more accurate, lots of questions are needed.
To deal with both needs, this Enneagram test has a built in mechanism that reduces the number of questions, without sacrificing accuracy. The test starts with questions for each type, so 9 questions per page. After a certain number of questions, statements for Enneagram types that the test taker didn't much identify with are omitted.
It is not always possible to omit questions. Some people seem to recognize themselves in statements for several different types. This may be due to them having complex personalities, or because they are not very discriminate at grading the statements, perhaps because of limited self-knowledge. Acquiring self-knowledge is not an easy task indeed and this Enneagram test is aimed at helping to improve it.
To be able to calculate the wings, the types adjacent to the type with the maximum score are not omitted. Also the type the person taking the test has entered beforehand will not be removed from the question list. So the minimum number of types one can get questions for per page is three. This makes for a total of 78 questions. The maximum number of questions is 126.
The following is a list of how many questions per page people on average have left at the end of taking the test.
4 questions - 9%
5 questions - 15%
6 questions - 22%
7 questions - 23%
8 questions - 18%
9 questions - 11%
On average this is 6.5 questions per page.