I don’t credit any science based on subjective questionnaire rather than objective observation. Like on this, there could just be a strong correlation between affinity and self perception. Where the closer the computer gets is because it can more easily map the affinity, not necessarily be quantifying personality itself.
If a person believes they are for the environment, might have an action that they click a “like” for an environmental meme. So a computer can match the likes to self perception. But a mom might know her daughter needs to join popular causes because she never made the cheer-leading squad, so her understanding is not going to match how her daughter sees herself.
personality is not answers to questions — personality is history. so the premise behind the study is also flawed. all they are doing is saying “look! The water’s wet!” What it amounts to. Lol. yes, Virginia, computers CAN compile information.
So there is a difference in how you see yourself, as opposed to how others see you. That part is correct science. But just because a computer can formulate how you see yourself by compiling data from likes, does not mean it “knows” you better. It just knows how you see yourself better.
It’s like that one quote from the movie, “evil never sees itself as evil” … There is going to be a bias on any self-reported observation of the self. It is action that matters, not belief.
So where a computer can very easily formulate how a human will “like” or not like–that decision to click–it cannot formulate the anomaly for when a click goes outside of normal patterns. But a person who knows you, might.
Article is interesting, number one… Because it shows they are continuing to do social experiments with information from Facebook. And number two, because the scientists screwed up.
Plus Facebook is already using an algorithm to put a spin on what appears in a feed, based on past likes. So already has patterns mapped where the computer can formulate probabilities. When members go through and tag that information with other data, like what religion or sex of the individual…. That’s where the data is compiled. So if you put up a sign that says “I’m a proud mother,” that just tagged it with a female indicator. But what the data cannot quantify, is how many females feel the need to announce how proud they are of being a mother, or how many are guilted into a share or a like. So are those NOT clicking not subject to guilt, or are they male, or are they too smart to click on tagging memes? So it can create an inclusion body of data, but fails to create any substantiated exclusion… So it is invalid for science…. Only useful for bottom fish marketing.
also there was no double-blind, the users they had in the experiment received the questionnaire, and THEN did the liking-process with what i assume were memes geared in specific manner to each personality question. in other words, they were taking the same test twice, and the computer merely validated their consistency in answering the “likes” in a similar manner to the questionnaire. in fact, this kind of study is subject to “coaching” — where on a non-reported aside, the participants are told to like only the things that fall into the same categories as the personality questionnaire. and all the computer is doing is keeping score on how well they match themselves.
for an actual study, would have to take likes already made or done first — THEN apply a personality questionnaire…. though it would still have a variance according to that individual’s need to be consistent — to display a consistency of behavior.
but if they did it with the likes first, would not get the results they wanted. so if i were to challenge this study, all have to do is reverse the order in which they took the data.
and the biggest personality trait of each of their study pool — is that they are docile and predictable. and so it depends if they chose the members for study based on docile traits to begin with, or if there was a random selection. so have to ask yourself what type of people are going to volunteer for a personality test?
i didn’t mean to go on about this …. but in many ways it embodies how incorrect science is used these days to create public “belief.” so then the question is, why do they want you to believe that a computer evaluation of personality is infallible?
and keep in mind that being unpredictable is a survival technique that will occur in less-congested and more active populations. or might be the opposite. see, now that’s the study they should be undertaking.
as our numbers rise, so does the percentage of “mentally ill” — but that has a factor of how voraciously the industry of mental health goes about their job. you actually could compute it all — but would need a super computer. eh……
i have taken one of those long-winded personality tests before. the only question that stumped me, was “do you believe in God.” all the rest, i knew what they were aiming for, as in what was the best trait to display.
i decided believing in God was the answer they wanted. but then the shrink judged me according to the dress i was wearing, anyway.
it’s like IQ tests…. often the smartest people are the ones who decide to not take the test.
and the ones who wrote the test are setting the bar at their intelligence being the top ceiling. nobody asks who writes the tests, or what qualifies them to determine intelligence.
the only thing a written test can validate, is if information given is retained. that’s it. that’s all. the rest are a symptom of nazi-like dreams for superiority or exclusion. God help them ……………