Sorry to destroy your stereotype, but the WOW statistician who made his way here is a she, not a he.
There is no way to provide a more nuanced analysis because this is not a random sample.
If people self-select to participate in a poll, then the only nuances you can provide are about the people who participated. You cannot extrapolate about people who did not participate, even if they are members of the same internet forum.
So all the statistics that can be gathered are statistics about "people who are members of AT and choose to answer a poll".
Mostly it means you can say things about the kind of tarot readers who do online polls and join forums. And that's the limit.
You can't say anything about people who didn't participate because you don't have any data about them. You can certainly project a lot of fantasies about the things you imagine about them, but that is just fiction, not based on any data.
I was previously thinking about how long a shot it would be if the site would have an actual statistician hereabouts, but I kinda knew too that people of all persuasions and professions flock here. It's so nice that an actual one got to answer my question!
Btw my apologies if my phrasing seemed to imply it, but I make no stereotypes about statisticians. If someone where to ask me the default sex of one, I would answer that it could be either really. It's just that in school here in my country, we were taught that in English, the gender-neutral form of a pronoun usually takes the masculine one. I know I could have used "he/she" or "his/her" instead of the plain "he" or "his", but it would be cumbersome if I were to use them all the time.
When I looked at the research that the MBTI manual did in 1985 and 1998, I was astonished and couldn't help but compare them to our very own polls here, since the results were quite divergent. I mean, a 4:4 vs 7:1 I:E result for two population sets is quite revealing. The figures were certainly compelling. But I remembered a caveat from a professor in our statistics class about making sure that the survey sample is sufficiently randomized before applying it to the focus population. I am assuming that the people behind the MBTI manual know their stuff and conducted their studies on a randomized sample of the general pop. But I have misgivings if we can say that about the poll we conducted in these forums, even if the population size of 571 may not be negligible. I remember how important random sampling is and how it might not apply to our poll here, so I posed my misgivings thus:
There might be questions about how randomized our sample of test-takers are relative to the members of these forums, and relative to all the Tarotists in the general population. If a statistician (who's also into Tarot- wow) would make his way here, I think he might be able to provide a more nuanced analysis.
The assumption I am going to make may be wrong, but I think that a layman, upon looking at the results of our MBTI poll where 87+ % of the responders answered as Introverts, would conclude (or fantasize) "Ah, so most Tarotists are really introverts!" They may not think that since the poll was not performed on a random sample, it would not be correct to extend its results to the pop segment they're thinking of.
So thanks a lot for clarifying things, CrystalSeas! You taught us that in making conclusions about the poll, it is not correct to state that "Most Tarotists are introverts" or even "Most Aeclectic Forumers are introverts". The accurate statement would have been "Most of the people who took the poll are introverts, though this conclusion does not apply to the members of this forum in general or even to the population of Tarotists as a whole". It's a good thing to know stuff like this so that we can make logical inferences.
