Blog of Sara Jakša

My Personality Based on the Text Analysis

I think I have already talked about the IBM Personality Insights. At least I used it to analyze the personality type of Tony Stark based on Iron Man movie. I have returned some of my attention to it, because I am still interested in the personality.

This time, I was analyzing myself. I took a number of blog posts from this site and put it in the machine, waiting to see, what it is going to bring out. Most of their subtraits, that it analyses, also have almost the same names as subtraits in the IPIP-NEO test that I have taken, which would normally allow me to compare them. It is too bad, that the IPIP test compares me to the women in my country only, as that way the results are not really comparable. Because I am quite sure that this test compares me with the whole sample. And it is going to be hard to tell, if the differences are thanks to the different methodology of measurement, or because the sample population is different. But that is not going to stop me form listing both of them for comparison anyway.

As before, the percentages means, that my scores are higher than that amount of people. So a score of 20% would mean, that I have a higher score than 20% of people. And a score of 80% percent would mean, that there are 80% of people with a smaller score, and 20% with a higher score.

Openness 99,7% 28%
Adventurousness 55,4% 45%
Artistic interests 93,8% 1%
Emotionally 57,8% 1%
Imagination 82,2% 89%
Intellect 99,9% 75%
Authority-challenging 99,8% 43%
Conscientiousness 38,3% 33%
Achievement striving 50,9% 12%
Cautiousness 82,6% 48%
Dutifulness 56,1% 33%
Orderliness 12,4% 66%
Self-discipline 14,9% 43%
Self-efficacy 69,4% 15%
Extraversion 11,3% 1%
Activity level 74,4% 21%
Assertiveness 88,7% 10%
Cheerfulness 01,6% 17%
Excitement-seeking 31,2% 8%
Outgoing 06,1% 1%
Gregariousness 03,9% 1%
Agreeableness 04,5% 20%
Altruism 78,8% 1%
Cooperation 50,1% 32%
Modesty 07,4% 15%
Uncompromising 45,9% 83%
Sympathy 93,1% 17%
Trust 67,9% 86%
Emotional range 48,3% 36%
Fiery 11,6% 17%
Prone to worry 37,8% 46%
Melancholy 64,5% 36%
Immoderation 36,2% 54%
Self-consciousness 87,7% 63%
Susceptible to stress 35,4% 30%

In all traits sans emotional range, I could easily map the values from both type of personality measurement. Normally all but one would have the exact name in both versions. I simply mapped the remaining one. Just in the case of agreeableness, I had to flip it.

The emotional range had most of the elements names different, but I could still map them based on the intuition, if they mean the same thing. Just keep in mind, that I could be wrong.

Looking at the general results, the results are more or less in the same ballpark in both tests. The biggest difference, at far as the main traits go, is in the openness, since one would put me in the high openness and one on the low openness. And looking at it in more detail, this comes mostly for the differences in Artistic interest, Emotionally and Authority-challanging. But in my opinion, I don't have high emotionality, nor artistic interest, so I would here lean more the the IPIP test. I could grant it, that I have average emotionality, but I not not at the top 10% in artistic interest in any case.

I mean, if I am at the top 10%, then there is not enough people for all the artists, hobby photographers, people that are actually interested in music and so on. So this is the point, where I am a bit skeptical. Actually a lot, if I was just a bit skeptical, I could wave it away, but here I can not.

But the most surprising one was in the field of Agreeableness. I don't consider myself altruistic, neither do I consider myself as a person with high sympathy. If these results would be true, than I am first of all, not understanding why there are so many people trying to provide the safe-spaces and battle for groups that are perceived (rightly or wrongs) to be at the threat and so on. I mean, I think Ayn Rand did have some point right, and I would consider her as far away from sympathy as possible. If I am so high, why are there even still charities? I mean, I don't really get these results. They don't explain, why people then act hell of a lot more sympathetic and altruistic then I do.

So, for now, I would trust the test more than this text analysis. But it might be a good way for cases, where solving a test would not be possible, for example when having to know the type of the person you negotiate with. I mean, if the Google, Facebook and others can analyze us, why could not we also analyze other people?