Frequency of Intertype Relations Depending on Three Methods of Their Calculation

(Let us get ahead of ourselves and say at once that the main theses in the conclusions in this part of the article will be as follows: intertype relations, in their specifically socionic character, really exist and really influence the formation of pairs of people; preferred ITRs are different for friendship and romantic pairs; the most accurate method for determining ITRs in pairs of people is the so-called “trait-based” method.)

THUS -

Intertype relations in pairs of questionnaire respondents can be calculated in three ways:

1. By their self-reported sociotypes (only for those pairs in which the sociotypes of both members of the pair are reported). In this case we are usually dealing with the most “socionically enlightened” audience, and, obviously, in this case one should expect the maximum effect of unconscious “adjustment” of reported types to ITRs considered comfortable in socionics.

2. By the maximum peaks (“leading types”) of the empirical type profile obtained as a result of questionnaire diagnostics for each respondent in the pair.

3. By a method based on the full trait profiles of the members of the pair and theoretically the most accurate, because it takes into account all the nuances of the socionic profiles of both respondents, rather than only their leading type. In this case, to diagnose the magnitude of each of the 16 possible ITRs represented in the relationship, the scalar product of three vectors is calculated. Specifically, the trait profiles of both members of the pair and a third vector, different for each type of ITR, are multiplied; in its 15 components (according to the number of traits), this third vector characterizes the theoretical match or mismatch of trait polarity between the respondents in the pair for the corresponding specific intertype relation. If, according to theory, the polarity of a certain trait in two people should match for a given intertype relation, then the corresponding component of the vector (with the number of that trait) equals plus 1, and in the case of a mismatch, minus 1. Thus, for the identity relation, the polarities of all traits in the two people should match, and therefore the third vector by which we multiply the scalar product of their trait profiles will consist of 15 positive ones. For clarity, all these vectors are presented in Table 2 (the vectors by which we multiply are presented in the rows of this table).

Thus, for each pair of people, 16 numbers are obtained (according to the number of ITR types), and each such algebraic number (it may be either positive or negative) characterizes the degree to which the corresponding ITR is represented in the actual relationship of this pair of people. Next, all numbers are set to zero except the largest one (that is, the one characterizing the ITR closest to the pair and most strongly expressed for it). After that, it remains to sum, in each sample of pairs, all the remaining numbers for each ITR type and then convert these sums into proportions representing the specific ITR in the sample of pairs of subjects under consideration.

The results of calculating the frequency of intertype relations by each of the three methods listed above for same-sex friendship pairs, opposite-sex friendship pairs, and opposite-sex romantic pairs are presented in Table 1.

WHAT CONCLUSIONS CAN BE DRAWN FROM EXAMINING THIS TABLE?

• When reporting their types, members of same-sex friendship pairs overestimate the presumed mirror and dual character of their relations, while underestimating their actual identity.

• Partly the same tendency also occurs in opposite-sex friendship pairs – here too, the types reported by respondents overestimate the probability of mirror relations and extinguishment relations and underestimate their actual identity. This is probably connected with the fact that people are most sensitive to the manifestation of nuances of differences in one another’s activity and leadership qualities (one is always slightly more of a leader, while the other is always slightly more of a follower). Therefore, even when both members of a pair actually belong to the same pole of vertness, but still differ somewhat in its absolute magnitude, they “contrast” one another’s vertness manifestations, and tend to attribute extraverted polarity to the slightly more active member of the pair and introverted polarity to the slightly less active one (although, we repeat, their actual polarity of vertness may be the same). It is surprising, however, that an overestimation of the probability of extinguishment relations (instead of identity relations) does not occur in same-sex friendship pairs – there, an overestimation of probability is observed only for mirror relations. Perhaps this is merely a statistical “fluke” caused by the comparatively small sample size. Unfortunately, in general, none of the effects of heterogeneity in ITR frequency that can be found in Table 1 have high statistical significance, although they do indicate tendencies that are quite theoretically reasonable. Samples 4-5 times larger are required to identify more accurate frequency values. Perhaps we will collect such a sample someday.

• Members of opposite-sex romantic pairs, in the types they report, overestimate the duality and semi-duality of their relations – in reality, in percentage terms across the entire sample, these turn out to be less frequent than reported, whereas activation, conflict, and supervision relations, according to the questionnaire results, on the contrary, turn out to be more frequent than the members of the pair report about themselves.

• The method based on the products of the trait profiles of the members of the pair turns out to be more accurate than the method of comparing the leading peaks of their type profiles. This is indicated by the higher value of sigma (dispersion) of the ITR proportions in all three samples of subjects considered, consisting respectively of same-sex friendship pairs, opposite-sex friendship pairs, and opposite-sex romantic pairs.

ALSO OF INTEREST IS THE PERCENTAGE OF AGREEMENT BETWEEN DIAGNOSES OF INTERTYPE RELATIONS WHEN THEY ARE CALCULATED BY THE THREE DIFFERENT METHODS DESCRIBED ABOVE.

The results of this analysis are as follows - the diagnosis of the dominant ITR agrees, in pairwise comparisons of the different methods of diagnosing ITRs, in the following proportion of cases:

by self-reported types compared with the ITR diagnosis based on the leading peaks of the type profile: 0,37 for opposite-sex romantic pairs and 0,36 for the sample of same-sex friendship pairs;

by self-reported types compared with the ITR diagnosis based on full trait profiles: 0,24 for opposite-sex romantic pairs and 0,22 for the sample of same-sex friendship pairs;

by the leading peaks of the type profile compared with the ITR diagnosis based on full trait profiles: 0,39 for opposite-sex romantic pairs and 0,35 for the sample of same-sex friendship pairs.

What do these figures indicate?

First, the percentage of agreement between ITR diagnoses calculated by different methods is QUITE SMALL. From this it follows, in particular, that the effect of self-suggestion in romantic pairs who know and report their types cannot in any way explain the high percentage of dual relations empirically obtained in these same pairs under trait-based calculation. This is because, for most pairs in these cases, the relation type itself turns out to be completely different as a result of diagnostics! (different from the reported one).

Second, the self-suggestion effect nevertheless does occur. In particular, it influences sociotype diagnosis by the method of identifying the highest peak of the type profile. In this case, the proportion of agreement between diagnosed ITRs and reported ITRs turns out to be one and a half times higher (0,37 compared with 0,24) than when reported ITRs are compared with ITRs diagnosed from full trait profiles. This, of course, could also be explained by another reason, namely by the much lower accuracy of ITR diagnosis from trait profiles compared with their diagnosis from leading type peaks – if the opposite had not been demonstrated! Specifically, the trait-based method of ITR diagnosis is more accurate – this follows from theory and is also confirmed experimentally (because when diagnosis is performed specifically by the trait-based method, the variability of ITR frequencies is greater in all samples, and this indicates lower noise contamination and greater reliability of this method).

Table 1. Comparison of proportions in samples for various ITRs calculated by three different methods

Table 2. What the second member of the pair is to the first and the resulting relation of trait poles (+1 - match, -1 - opposite poles)

Table of Intertype Relations

Abbreviations:

• Id. - identity relations
• Dual. - duality relations
• Act. - activation relations
• Mirr. - mirror relations
• Kin. - kindred relations
• SD - semi-duality relations
• Bus. - business relations
• Mirg. - mirage relations
• SE - super-ego relations
• QI - quasi-identity relations
• Ext. - extinguishment relations
• Conf. - conflict relations
• Benf. - benefit relations (you are the benefactor)
• Beny. - benefit relations (you are the beneficiary)
• Supv. - supervision relations (you are the supervisor)
• Supe. - supervision relations (you are the supervisee)

Explanation: to determine intertype relations from the table, find your type in the left vertical column and your partner’s type in the top row; the intersection of the row with your own type and the column with your partner’s type will show the intertype relation.