Socionics Type and Intertype Relations Statistics by Filatova¶

Published by Ekaterina Filatova in “Socionics, Mentology, and Psychology of Personality”, 2000, 6.

Abstract: This article outlines the results of a statistical study that was conducted on 105 families, numbering 299 people in total. Four main socionic dichotomies were examined and uneven distributions were found. The distribution of couples according to intertype relations and the socionics relations of children to their parents were also investigated. Some correlations were identified in the frequency of sociotypes among children in relation to the sociotypes of their parents.

Key words: socionics, socionics types, intertype relations, family relations, distrubtion of types, statistics of types and intertype relations.

This work was originally conceived as a study of relations between types of children and types of their parents. For this purpose, I have collected and stored relevant data of many years of my work, including data on people who sought my counseling to resolve conflicts with their family members. Information presented here does not relate to people engaged in a certain field or profession, but is completely random, and also contains data from the relatives, friends, neighbors, etc.

Accordingly, it seemed appropriate to use this data for other socionic research that can be conducted on a sufficiently random sample of 299 people, and thus, this work was broken down into two parts. The first part presents research of distribution of types according to gender and socionics dichotomies, while the second part touches upon intertype relations in the families, and the relation of children’s sociotypes to types of their parents and other family members. This study in part also concerns similar type of research that was carried out by Bukalov, Karpenko, and Chykyrysova (1), thus some comparisons are presented.

At the end of the article a complete list of the data on which this study was based is presents. Data on the first 71 families were gathered during many years of my research work; the rest was kindly provided by E. Malskaya, A. Filimonov, D. Shulman, for which I am thankful to them.

Part 1. General Research.¶

The total number of people surveyed (parents and their children) is 299 people, out of which 161 individuals were female and 138 individual were male. The list includes 27 single-parent families. For 24 of them the type of father is unknown, and for 3 of them - the type of mother is unknown. Adding the number of missing parents to the total number of representatives we obtain an almost equal number of both sexes: 164 women and 162 men. This is consistent with the known statistics on the distribution of men and women.

The following table shows type distribution among 299 people broken down by gender.

Table 1. Summary statistics.¶

Pragmatists	Men	Women	Total	%	Socials	Men	Women	Total	%
LSE	11	8	19	6.35%	ESI	2	11	13	4.35%
LSI	10	9	19	6.35%	ESE	1	9	10	3.34%
SLE	4	9	13	4.35%	SEE	2	16	18	6.02%
SLI	5	5	10	3.34%	SEI	11	7	18	6.02%
Total:	30	31	61	20.40%	Total:	16	43	59	19.73%
Researchers	Men	Women	Total	%	Humanitarians	Men	Women	Total	%
LIE	12	9	21	7.02%	EII	3	6	9	3.01%
LII	13	14	27	9.03%	EIE	22	16	38	2.71%
ILE	9	6	15	5.02%	IEE	6	7	13	4.35%
ILI	19	14	33	11.04%	IEI	8	15	23	7.69%
Total:	53	43	96	32.11%	Total:	39	44	83	27.76%

Quite unexpected was the general inequality in the distribution of sociotypes. The most frequent types were EIE and ILI (38 and 33, respectively), while the least numerous types were EII, ESE, and SLI (9, 10, and 10, respectively). There were significantly fewer representatives of Pragmatist and Social clubs in this sample (40.13%) than Researchers and Humanitarians (59.87%).

These findings differ significantly from the data from [1] where the authors observed a uniform distribution, as follows:

Table 2. Comparative statistics.¶

	Distribution from this study	Distribution from [1]
Pragmatists [STs]	20.40%	26.5%
Researchers [NTs]	32.11%	25.2%
Socials [SFs]	27.76%	24.40%
Humanitarians [NFs]	27.76%	24.40%

Of course, we can assume that the disproportionate number of Researchers compared to other clubs is due to the fact that the circle of scientific intelligentsia, to which all those who who provided the data belong, is closest to the Researchers club. However, the children, who comprised half the list, have not yet chosen their occupation and interests. It is also noteworthy is that the second largest group is the Humanitarian club, in which the most numerous type belongs to the representative of Humanitarians - EIE. This cannot be explained the preferred social circle.

Evaluation of the confidence probability using statistics of Student (t-test) that the obtained values are not random:

Pragmatists [STs]	LSE - 0.15	LSI - 0.15	SLE - 0.95	SLI - 0.985
Researchers [NTs]	LIE - 0.95	LII - 0.99	ILE - 0.85	ILI - 0.99
Socials [SFs]	ESI - 0.95	ESE - 0.98	SEE - 0.20	SEI - 0.20
Humanitarians [NFs]	EII - 0.98	EIE - 0.99	IEE - 0.95	IEI - 0.97

These results show a sufficient high probability of uneven distribution of sociotypes in most cases.

This data may reflect the observation that in presence of significant number of inventions and discoveries, of high level of philosophical thought, culture and spirituality, our country was never distinguished by practical ability, while social services always occupied the last place. Of course, we cannot confidently draw any conclusions at this point, only forward possible arguments about the consistency of the data.

Distribution by Quadra:

Table 4. Distribution by Quadra.¶

Quadra	Distribution from this study	Distribution from [1]
First Quadra (Alpha)	23.41%	23%
Second Quadra (Beta)	31.10%	29%
Third Quadra (Gamma)	28.43%	26%
Fourth Quadra (Delta)	17.06%	22%

The fourth squadra is the smallest, while the second is almost double the size of the fourth. This trend, however, is less expressed in work of Bukalov. et al [1] that studied only married couples.

Now, let us examine the distribution of sociotypes by Jungian dichotomies. This data is presented below by gender.

Table 5. Gender distribution along the socionics dichotomy of rationality-irrationality.¶

Irrationals	Men	Women	Total	Rationals	Men	Women	Total
IEI	8	15	23	EIE	22	16	38
ILI	19	14	33	ESE	1	9	10
SEI	11	7	18	LIE	12	9	21
SLI	5	5	10	LSE	11	8	19
IEE	6	7	13	EII	23	6	29
SEE	2	16	18	ESI	2	11	13
ILE	9	6	15	LII	13	14	27
SLE	4	9	13	LSI	10	9	19
Totals:	64	79	143	Totals:	74	82	156

The distribution in this case is almost equal. Given that there are 21 more women on our list than men, we can assume women are no less rational than men. Summary of the results:

For women: rational types - 49.7 % (79/161), irrational types - 50.93% (8/161)

For men: rational types - 46,38% (64/138), irrational types - 53,62% (74/138)

Table 6. Gender distribution along the socionics dichotomy of extraversion-introversion.¶

Extroverts	Men	Women	Total	Introverts	Men	Women	Total
LSE	11	8	19	IEI	8	15	23
LIE	12	9	21	SEI	11	7	18
ESE	1	9	10	ILI	19	14	33
EIE	22	16	38	SLI	5	5	10
IEE	6	7	13	EII	3	6	9
SEI	2	16	18	ESI	2	11	13
ILE	9	6	15	LII	13	14	27
SLE	4	9	13	LSI	10	9	19
Totals:	67	80	147	Totals	71	81	162

Data on the parameter of extraversion / introversion is similar to the data on the rationality / irrationality. Just as in the previous case, the number of extraverts is comparable to the number of introverts, and given the larger number of women (21) we can assume that the distribution of extraversion / introversion trait according to gender is essentially the same.

For women: extroverted types - 49.69% (80/161), introverted types - 50.31% (81/161)

For men: extroverted types - 48.55% (67/138), introverted types - 51.45% (71/138)

Table 7. Gender distribution along the socionics dichotomy of logic-ethics.¶

Logical	Men	Women	Total	Ethical	Men	Women	Total
LSE	11	8	19	IEI	8	15	23
LIE	12	9	21	SEI	11	7	18
SLE	4	9	13	EII	3	6	9
ILE	9	6	15	ESI	2	11	13
LSI	10	9	19	IEE	6	7	13
LII	13	14	27	SEE	2	16	18
SLI	5	5	10	EIE	22	16	38
ILI	19	14	33	ESE	1	9	10
Totals:	83	74	157	Totals:	55	87	142

From the above table we can see that among men the relative number of logical types is greater than ethical types, while the reverse is true for women. The overall ratio of logical to ethical types is about even (157 to 142). This difference is within the margin of error associated with the small sample of 299 people. Possibly, with more statistics this data may even out or diverse even further. There is nothing more specific that could be said at this point.

For women: logical types - 46%, ethical types - 54%

For men: logical types - 60%, ethical types - 40%

The difference in distribution of logical and ethical types is less among women than among men. A difference of 8% makes it less likely that there is a difference in distribution of ethical and logical types among women, while for men the results are more supportive of the existence of such difference.

These results are also markedly different from the data of [1], where in the sample group 68% of women were ethical types while 71% of men were logical types.

Table 8. Gender distribution along the socionics dichotomy of sensing-intuition.¶

Sensing	Men	Women	Total	Intuitive	Men	Women	Total
LSE	11	8	19	LIE	12	9	21
ESE	1	9	10	EIE	22	16	38
SLE	4	9	13	ILE	9	6	15
SEE	2	16	18	IEE	6	7	13
LSI	10	9	19	LII	13	14	27
ESI	2	11	13	EII	3	6	9
SLI	5	5	10	ILI	19	14	33
SEI	11	7	18	IEI	8	15	23
Totals:	46	74	120	Totals:	92	87	179

For this last dichotomy, an overall significant shift toward intuitive types is evident (179-120), and this cannot be attributed to insufficient statistics. The confidence level for both sensing and intuitive numbers is greater than 0.7 applying Student statistics. The distribution according to gender is as follows:

For women: sensing types - 46%, intuitive types - 54% For men: sensing types - 67%, intuitive types - 33%

Here, just as with the dichotomy “logic-ethics”, the differences for women lies within the statistical error (8%), while for men the difference is more significant.

In summary, dichotomies of ationality / irrationality, and extraversion / introversion show an equal distribution according to gender. Dichotomies of logic / ethics and sensing / intuition show uneven gender distributions. Our data also shows that nature takes care of the availability of all poles of the dichotomies. The only major shift in the poles, in our case, was presents in the dichotomy of intuition-sensing towards the intuitive side. The distribution of sociotypes was extreemely uneven (EIE - 39, IEI - 9), such that not everyone, in theory, would be able to find a dual.

Part 2. Study of sociotypes and intertype relations in families.¶

Now let’s examine the parameters of families, and above all, let us analyze the distribution of couples to intertype relations. The result is shown below:

Table 9. Distribution of families by intertype relations.¶

	Distribution from this study	Distribution from [1]
Duallity	12 (15.6%)	45%
Activation	12 (15.6%)	45%
Semi-duality	9 (11.7%)
Mirrow	8 (10.4%)	5%
Quasi-identity	7 (9.1%)
Benefit	6 (7.8%)	10%
Supervision	5 (6.5%)	5%
Extinguishment	4 (5.2%)
Conflict	4 (5.2%)
Business	3 (3.9%)
Superego	3 (3.9%)
Illusionary	2 (2.6%)
Kindred	1 (1.3%)
Identity	1 (1.3%)	6%
Total	77 (100%)

Note: According to our information, two families in supervision relations fell apart shortly after they had a child.

Our results diverge strongly from the data of A. V. Bukalov et al. in citation (1) which estimated dual pairs to be at 45% compared to our 15.6%. Other data can be considered qualitatively consistent, given the small sample size.

Next we present a very interesting result concerning the distribution of sociotypes of children and the sociotypes of their parents. The following table shows the child’s type’s socionics relation to the type of his or her mother and father:

Table 10. The correlation of types of children to the types of their parents.¶

	Dual	Semi -dual	M irage	M irror	Con trary	Con flict	S uperv ision
F ather	5	6	8	10	3	6	5
M other	36	6	6	5	15	7	11
	Iden tical	Ki ndred	Bus iness	Sup erego	Quasi -iden tical	Act ivity	Be nefit
F ather	31	9	8	3	5	5	3
M other	11	6	6	2	7	7	12

Two sharp outliers in intertype relations of children to their parents immediately attract attention: duality of children to their mothers and identity to their fathers are strongly represented in comparison to all other types of relations. Indeed, among the full number of children, which is 142, we were able to trace 136 having relation to mother and 108 having relation to father.

25.7% (35/136) cases when the child was dual to mother

28.7% (31/108) cases when the child was identical to father

This strong deviation is hardly accidental. Confidence level of these results according to Student’s test is greater than 0.7. At the same time, identity relation to mothers and duality to fathers is not particularly distinguished from other types of intertype relations. (Note that the numbers of children related to their mothers and their fathers were not the same because in some families it was not possible to obtain the type of one of the parents).

This different could be due to various reasons, in particular, children often tend to mimic and play into the roles of their parents (boys imitate their fathers, and girls - their mothers), which increases the potential error in determining their sociotypes. However, in this case, the identical relations deviated only in relations to the fathers, while identity to the mothers was very weak. For mothers, by contrast, according to our data, it is more characteristic to have children who are not identical but dual. Furthermore, out of 31 children who were identical to their fathers, 11 were girls and 20 were boys. It can be proposed that some of the boys ended up on this list only on the basis of roles, since there were 9 more boys than girls. Even if you subtract that number 9 of total 31 yielding 22, this number of children identical to their fathers still strongly stands out of the picture.

If, however, role imitation was not the decisive fact here, it is possible that the inheritance of sociotype is uneven according to the child’s gender: boys more often inherit the sociotype of their fathers. Course, all this is still only be considered as a preliminary results to which I would like to draw the attention of my colleagues and invite them to take part in this type of research.

Another number that stands out is contrary intertype relations of children to their mothers, while in relation to the fathers this figure is likely an underestimate. Significantly high is the number of supervision relationships of children to their mothers. Here, however, I would like to share some observations: in three of these families, with which I’m familiar enough, there is the most tender friendship, I would say even excessive attachment between the parent and the child in supervision relations. Apparently, the difference in levels of development has an effect here, as the parent is either particularly careful of the vulnerable function of the child, or, if the child is the supervisor, is in excessive admiration of the child.

It was customary to believe that the trait of rationality-irrationality is transmitted to the children provided that both parents are same on this trait. It turned out that this is not so: for 25 pairs on the above list this is indeed the case, however, two rational parents may have an irrational child, and vice versa, irrational parents may produce an irrational child as has been observed in 10 cases; that is, this is not so rare.

Since 25 +10 = 35 - this is the total number of families where parents are both rational or irrational types, we have: 25/35 = 0.71, that is, in 71% of cases, the child is on the same side of the rational-irrational dichotomy as the parents, and in 29% of cases - on the other side. This question hasn’t been considered for the other dichotomies since the fact that the other traits are not conserved is quite obvious from the data.

It may be surmised that there is reproducibility of the same sociotypes within families across several generations. Unfortunately, there is not enough data for such analysis due to the difficulty of determining the sociotypes even within three generations, for obvious reasons, but it should be noted that such proposition makes sense.

Here is a list of “kindred” families. The elder generation is at the top of the list, and below it is the younger generation.

Families, numbered 1-3. Curiously, the rational parents in this couple had two rational children, EIE son and daughter, and then suddenly an irrational child of type IEI (family 1). The same situation occurred in the family 2, in which the mother EIE (daughter of family 1) had a daughter of sociotype IEI.
Families 20-21. In family 21, a grandson appeared of same type as his grandfather, SEI.
Families 24-25. Almost all of members are of sociotype LSI. The grandson is of the same type as the grandmother.
Families 32-33. SEE grandson is of the same type as that of his grandmother.
Families 52-54. Frequent repetition of sociotype ILI. In particular, in family 54 son repeats the type of his grandfather of the family 53.
Families 83-85. Here grandfather LSE of family has grandaughter LSE in family 3.
Families 71-74. Children in family 72, LII and IEI, with their families (73 and 74) reproduced types LIE and IEI.
Families 95-96. Any patterns are not noticeable.
Families 99-100. Family 100 had a grandson of type LIE, same type as his grandfather.

In all of these above listed families we were able to identify sociotypes of 21 grandchildren, out of whom 8 were repeating the sociotype of one of their ancestors. But we have managed to trace only a single vertical line. However, each grandchild has two grandmothers and two grandfathers, therefore, the likelihood of recurrence of sociotypes is doubled, as a result, confidence level is more than 0.99.

Thus, we see that the most frequent distribution of sociotypes is when a grandchild is repeating the sociotype of grandparents: could this be the reason for the special tender affection of grandparents for their grandchildren? At the very least, even this small statistical sample reveals such a trend quite confidently.

Another frequently encountered view in socionics is that dual parents have dual or identical children. This opinion was not confirmed: in dual families we did find 10 children who were dual or identical types to their parents, but there were also 5 children of other types; that is, sufficiently frequently this hypothesis is not supported.

The above data, namely, the appearance of children who are dual to their mothers and identical to their fathers with probability greater than 25%, and the frequent recurrence of the same sociotypes within several generations, implies a possible genetic origins of socionics types.

Indeed, a fertilized egg already carries some genetic information that in case of normal development of the fetus does not change and remains the same after birth.

The assumption of the genetic nature of the formation of sociotype, in my opinion, is also confirmed by a strong external similarities of sub-types within a type. This is particularly evident in those cases where the investigator is working with sufficient relevant data. Currently I have the photo library numbering several thousand photographs of at least a thousand representatives of different types (shots were made from three or more angles for each person). In cases where there is large amount of data for members of the same type (for example, more than 100 people of type ILI), groups of similar faces become apparent, sometimes producing an impression of well-known twins. Pictures of ILIs in my photo library naturally form around a dozen such clusters, eight pairs of which can be found in my book (2).

In this regard, I would like to remind the reader of a very interesting, from my point of view the work D. A. Ivanova (3), which has shown that identical twins have same sociotypes, as well as being very similar in appearance and various physiological parameters. In this case, their genetic sameness is obvious. Also I can recollect a private communication of director of the St. Petersburg Institute of Psychoanalysis M. M. Reshetnikov concerning the work of one physician who studied medical parameters of identical twins and found that these parameters were almost completely identical. Unfortunately, he could not provide any coordinates of this work.

Apparently, in socionics, observing groups of people who are nearly identical externally and by physiological manifestations we have come close to the “elementary”, so to speak, type, whose properties can be described as a specific genetic code, and thus, the question of total number of such “elementary” types can be resolved. Already, it seems clear that the traditional socionic count of 16 types is some kind of intermediate stage of classification. Each of the 16 psychological types naturally breaks down to a number of smaller groups (in my photo library this number varies from 14 to 16), the exact value of which can be determined with sufficiently large amount of data. Currently, the most important question in my view is the basis for these divisions.

In conclusion, I want to sincerely thank A. M. Elyashevich who took the trouble to present the material of this paper to the socionics conference in September 1999, and with a high degree of understanding of the work answered numerous questions of the audience.

Supporting data:

Each family is assigned a serial number. The next column contains sociotypes of mother, father, their socionics intertype relation, then the types of their children and their intertype relations to each parent. The letter “s” and “d” indicate the sex of the child (son or daughter), while in the column of children intertype relations with their parents the letters “m” and “f” “characterize the relationship with her mother and father, respectively.

The names of all the families are recorded in my file cabinet according to their numerical order, however, as there have been numerous requests to not publish their full names, these were removed from the list below.

It may be useful to provide complete, detailed set of data instead of just the end result for my colleagues who may want to extract additional information that may be of interest to them.

Table 11. Summary Table of intertype relations for 299 families

[Table not included. Full table can be found at: http://uchebilka.ru/voennoe/8590/index.html?page=3]

References¶

Bukalov A.V., Karpenko O.B., Chykyrysova G.V., “On statistics of relations within couples. // Socionics, Mentology, and Psychology of Perosnality”, 1, Kiev, 1999.
Filatov E.S., “Socionics for everyone”, BSK, St. Petersburg, 1999.
Ivanov D. A. “Some features of thinking that twins with mental infantilism. // Socionics, Mentology, and Psychology of Personality”, 6, Kiev, 1997