Back to Enneagram Theory and Research

Do opposites attract?

What makes one person attractive to another?  Do we choose mates that are similar to ourselves, or do we need differences to ignite the romantic spark?  Which types fit best with which other types?  This page shows some of my research into this question.

The table below shows the frequency of all 81 combinations of Enneagram-type marriages.  There are 457 couples in this study.  Some combinations are strikingly more common than chance, and are highlighted in yellow or red (1 or 2 standard deviations above chance).  Some combinations are much rarer than expected by chance, and these are shaded in gray.
 

 

Table of Enneagram Marriage Frequencies

 

Husband

 

 

 

 

 

 

 

 

Wife:

Husb. is 1

Husb. is 2

Husb. is 3

Husb. is 4

Husb. is 5

Husb. is 6

Husb. is 7

Husb. is 8

Husb. is 9

Wife is 1

5

4

5

3

14

7

9

7

15

Wife is 2

10

0

6

8

12

15

4

18

12

Wife is 3

7

0

1

2

4

5

3

1

5

Wife is 4

8

6

3

5

6

3

5

11

16

Wife is 5

5

1

0

0

2

2

2

1

3

Wife is 6

6

1

8

4

11

2

2

13

13

Wife is 7

8

1

5

1

8

4

1

6

7

Wife is 8

2

2

2

4

2

3

1

3

11

Wife is 9

9

1

9

2

11

14

6

11

2

 

Color

Key

 

2 standard deviations above chance (98% confidence)

 

1 standard deviation above chance (85% confidence)

 

1 standard deviation below chance (85% confidence)

Note: because the color indicates frequency relative to chance, some red or yellow cells may be numerically smaller than cells not highlighted.

Major findings:

1.  Marriage patterns are not random.  For example, of all female 8s, 37% of them married male 9s, even though only 19% of all men were 9s.  There are a lot of such pairings that happen at far above chance levels.

2.  Same-type marriages are rare, occurring two times less often than expected by chance.  "Birds of a feather" may flock together as friends, but apparently not in marriage.  An interesting exception appears to be double-4 marriages, the only double-type pair to occur more often than chance (it is only 1/2 standard deviation above chance, so it is not colored in the above table).

3.  Men and women choose very different personalities for their mates.  Male 9s with female 4s are common (16 couples), while male 4s with female 9s are extremely rare (2 couples).  Female 8s preferred male 9s, whereas Male 8s preferred 2s and 6s.  

4.  People marry along a line of integration 38% more often than chance.  For example, 7-1 and 3-6 marriages were common, as were male 2s-female 4s, male 3s-female 9s, and several other such combinations.

5.  "Gut" types (8, 9, and 1) neither preferred nor avoided other "gut" types, and the same was true for heart and thinking types.

6.  Although people avoided marrying their own types, they did not show any avoidance of adjacent (wing) types.  A preference for wing types was not seen either.

7.  The 6 most common type pairings are:

        18 pairs:  male 8, female 2
        16 pairs:  male 9, female 4
        15 pairs:  male 6, female 2
        15 pairs:  male 9, female 1
        14 pairs:  male 5, female 1
        14 pairs:  male 6, female 9

It is notable that 9+4 marriages were among the most common, given the reputation of this combination for stormy volatility.  It should be noticed that this list is ranked purely by numerical counts, not by its frequency relative to chance.  Hence, it is skewed toward the more common types (9s and 6s).

Future questions:

1.  How does the personality of marriage partners affect their happiness, personal growth, and longevity?  These answers could be very relevant to marriage counselors.

2.  The data on this page are reported by the couples themselves, or their children, and are not verified by an objective test.  Unfortunately, even the best available Enneagram tests may have only limited accuracy (see my comparison of Enneagram tests).  Development of better tests is critical for progress in Enneagram research.
 

Overall distribution of types:

If types were distributed by chance, one would expect each type to be about 1/9 = 11% of the population.  This is clearly not the case, as we see in the table below.  9s are somewhat more than 11% of the population, while 3s are less, particularly among women.  These gender differences are very consistent with what has been reported previously by Dora and Ted Levinson (Enneagram Monthly, Sept. 1998).
 

Type:

Females

Males

1

15%

13%

2

20%

4%

3

5%

6%

4

12%

6%

5

4%

15%

6

16%

13%

7

8%

8%

8

7%

17%

9

14%

19%