-->

27 March, 2009

Myers-Briggs, Surveys, and the like

Apparently, if you send out a random survey, you can only expect about a 3-5% response rate. I guess I believe that. But it still surprises me that when I send out a survey to a known audience (i.e. my friends) instead of a random audience, that my response rate is not that much better.

I asked 127 of my Google contacts to tell me their Myers-Briggs personality type. Only 21 people responded to me. That's 16%, which is about 3-5 times more than what you'd expect to get if you sent to complete strangers. But I am not sure that's comforting. It basically means that I am only 3 times as popular with my friends, as I would be with complete strangers! Alternatively, it might be that it has nothing to do with the fact that they're my friends, and they responded to me because they knew the answer, and it took minimal effort.

Anyway, I wanted to know what type people were, because I had some theories, and was also looking to see if there were any obvious patterns. I am not going to tell you who any of the people were who responded. But I am going to tell you what their responses were:


The interpretation of the table above is as follows (for those who are not familiar with the system, and also to explain some of my "code")

E/I = extravert/introvert
N/S = intuitive/sensing
T/F = thinking/feeling
J/P = judging/perceiving

X = person scores equally on a particular measure, and was not able to be categorized

The titles of the first and third pairing (E/I and T/F) are fairly self-explanatory. But the other two groupings require a little explanation. Briefly, "intuitive" classification means that you tend to act based more on abstract ideas, whereas "sensing" classification means that you tend to act based on concrete practical realities in front of you. The "judging" classification tends to be people who are very schedule and plan oriented, whereas "perceiving" people are less schedule and plan oriented.

My list of contacts was split rather evenly between men and women. But I received only 4 responses from males.

The majority of respondents were "N" compared to "S" (13 to 5). There could be many reasons for this. Maybe, being that most of my acquaintances are either scientists, engineers, or musicians, there is a preference for abstract thinking? Not sure if that's a valid hypothesis. Another possibility is that more people test "N" because the questions that that they ask that indicate "N" sound much cooler than the ones that they ask for "S". For example, would you rather say "I am attracted to complex ideas" or "I am most comfortable with everyday things that are right in front of me". It's like saying "I am an intellectual" or "I am a dullard". So there's a possibility that the entire Myers-Briggs is invalid because nobody could possibly want to be an "S"! Of course, perhaps actual dullards would think that abstract ideas are a stupid waste of time. So maybe I'm bleeding my "N" bias into that argument. Another possibility is that it's a sampling bias. Maybe the kind of people who are willing to respond to random surveys about personality type are people who are attracted to abstract ideas, in which case I would be far more likely to receive responses from "N" than from "S" people. Actually, that is my strongest hypothesis, I think.

The majority of respondents were "T" versus "F" (13 to 5, again). This one I may be tempted to attribute more to who my friends are (scientists, engineers) rather than sampling bias, because my informal discussions have never shown there to be any tendency for "T" or "F" to be more or less "bought-in" to the Myers-Briggs subject.

The majority of respondents were "I" versus "E" (13 to 5, again). In this case, I think it likely reflects a fairly accurate representation, in that I do believe that there are probably twice as many introverts out there as there are extraverts.

Other things of note:
There were no people who responded as "SF" and I have actually never met anyone who said that they had that pairing
All of the men who responded were "T" (though I have spoken recently with several "F" males, so this was definitely an artifact of small sample size)
The fact that there were nearly equal number of "J" and "P" also seems about right, given my own experiences with people

Did I get anything useful out of this?

Other than the fact that I now know these people's Myers-Briggs types, no. And given the fact that I probably could have guessed about three-fourths of those peoples' types accurately without asking them, I didn't really get a whole lot at all.

Except for this blog entry.

No comments:

Post a Comment