3 ms·

Slightly related to the article: Does anyone else feel like the general public's inability to extract information from statistics is a huge problem? I think th

Slightly related to the article: Does anyone else feel like the general public's inability to extract information from statistics is a huge problem? I think that any sort of "standardized" education should include at least a class worth of teaching on how to interpret statistics.

gnodar 8y ago
I had that class in HS. Or at least I had a general statistics class where the teacher emphasized interpreting results, common fallacies, etc. As one might expect, the students didn't pay attention to anything other than memorizing the formulas and spitting them back out on the test. They then promptly forgot the formulas the memorized. I took the optional follow up class as an elective. Even amongst the students of that class, who were interested enough in statistics to take a more advanced one as an elective, all of us had to re-learn the formulas we forgot. My point is, I know it's simple to notice a problem in society and think that the obvious solution is to make it required learning, but it won't stick unless its reinforced over and over again. It can't just be a class, it has to be reiterated in various forms throughout many classes.
- skookumchuck 8y ago
  One way to do it is to teach statistics using poker (or blackjack).
madengr 8y ago
Sure, but the average person has probably had only high school algebra, and remembers none of it.
daenz 8y ago
YES There are countless debates I've been a part of, on various topics, with parties involved cling to outliers as proof of their position, even in the face of actual statistical data. Words like "average" or "likelihood" don't seem to mean anything to them...only narrow anecdotes that are impossible to argue against without using statistics. Example: I've debated someone that men are on average taller than women. They refused to accept this because they "know plenty of tall women." And their rejection of statistics and the notion of "on average" can be summed up by: "there is no 'average' person."
anonacct37 8y ago
It's not just the general public. There's a much deeper problem here. Everyone from Medical doctors to phds to basically everyone on this forum is in the same boat. We believe what we want to believe and we overwhelmingly reject data that doesn't fit our narrative.
Bartweiss 8y ago
> I think that any sort of "standardized" education should include at least a class worth of teaching on how to interpret statistics. This is sort of a fantasy of mine - having every public school teach a brief course on "not getting tricked". In my fantasy world, it's a several-week guide to common ways people lie and mislead the public. One section is on interpreting statistics. Another is probably on comparing scientific papers to the news stories they spawn. Another on 'paltering' and all the things a given statement doesn't say. The 'fun' section covers cold reads, famous hoaxes, and a general sense of how to not end up believing in every silly story you hear. You could use The Demon Haunted World as a readable source textbook for a lot of it. It's a pipe dream, of course. You'd need teachers who could teach it, which is a tall order to begin with, and then you'd need to find a way to give out examples without offending half the parents in the school when every examples (astrology, crystal healing, and so on) is going to cross somebody. But damn, even if it didn't take completely, wouldn't it be nice to at least have a shared framework to talk about this stuff?
- rm_-rf_slash 8y ago
  When I was in middle school, my home economics class had a week on media literacy, and while I only remember one sentence: “Is a song played on the radio because it is popular, or is it popular because it is played on the radio?” ...it told me everything I needed to know at a young age. Ten years later I’m introduced to Adam Curtis’ “Century of the Self” and that filled in a few gaps.

gnodar 8y ago

I had that class in HS. Or at least I had a general statistics class where the teacher emphasized interpreting results, common fallacies, etc. As one might expect, the students didn't pay attention to anything other than memorizing the formulas and spitting them back out on the test. They then promptly forgot the formulas the memorized. I took the optional follow up class as an elective. Even amongst the students of that class, who were interested enough in statistics to take a more advanced one as an elective, all of us had to re-learn the formulas we forgot. My point is, I know it's simple to notice a problem in society and think that the obvious solution is to make it required learning, but it won't stick unless its reinforced over and over again. It can't just be a class, it has to be reiterated in various forms throughout many classes.

skookumchuck 8y ago
One way to do it is to teach statistics using poker (or blackjack).

madengr 8y ago

Sure, but the average person has probably had only high school algebra, and remembers none of it.

daenz 8y ago

YES There are countless debates I've been a part of, on various topics, with parties involved cling to outliers as proof of their position, even in the face of actual statistical data. Words like "average" or "likelihood" don't seem to mean anything to them...only narrow anecdotes that are impossible to argue against without using statistics. Example: I've debated someone that men are on average taller than women. They refused to accept this because they "know plenty of tall women." And their rejection of statistics and the notion of "on average" can be summed up by: "there is no 'average' person."

anonacct37 8y ago

It's not just the general public. There's a much deeper problem here. Everyone from Medical doctors to phds to basically everyone on this forum is in the same boat. We believe what we want to believe and we overwhelmingly reject data that doesn't fit our narrative.

Bartweiss 8y ago

> I think that any sort of "standardized" education should include at least a class worth of teaching on how to interpret statistics. This is sort of a fantasy of mine - having every public school teach a brief course on "not getting tricked". In my fantasy world, it's a several-week guide to common ways people lie and mislead the public. One section is on interpreting statistics. Another is probably on comparing scientific papers to the news stories they spawn. Another on 'paltering' and all the things a given statement doesn't say. The 'fun' section covers cold reads, famous hoaxes, and a general sense of how to not end up believing in every silly story you hear. You could use The Demon Haunted World as a readable source textbook for a lot of it. It's a pipe dream, of course. You'd need teachers who could teach it, which is a tall order to begin with, and then you'd need to find a way to give out examples without offending half the parents in the school when every examples (astrology, crystal healing, and so on) is going to cross somebody. But damn, even if it didn't take completely, wouldn't it be nice to at least have a shared framework to talk about this stuff?

rm_-rf_slash 8y ago
When I was in middle school, my home economics class had a week on media literacy, and while I only remember one sentence: “Is a song played on the radio because it is popular, or is it popular because it is played on the radio?” ...it told me everything I needed to know at a young age. Ten years later I’m introduced to Adam Curtis’ “Century of the Self” and that filled in a few gaps.