People can't interpret statistics

Page 3 of 3 [ 45 posts ] Go to page Previous 1, 2, 3

Previous topic | Next topic

linatet
Veteran
Veteran

Joined: 8 Sep 2013
Age: 29
Gender: Female
Posts: 934
Location: beloved Brazil

24 Feb 2014, 8:14 pm

NotaHero wrote:

You should read 'The Tiger That Isn't' by Michael Blastland which is a very interesting book about interpreting statistics and how to see past the mistakes and misinterpretation of others (Particularly by the Media)

thanks for the suggestion! I just added this one to my books list.

Verdandi
Veteran
Veteran

Joined: 7 Dec 2010
Age: 54
Gender: Female
Posts: 12,275
Location: University of California Sunnydale (fictional location - Real location Olympia, WA)

24 Feb 2014, 8:47 pm

I've noticed several people here do not understand statistical significance. More than a few times I've seen people argue that you cannot derive data from a statistical analysis of a large population unless you get data from every single person on Earth, basically. Or every autistic person on Earth.

Some are also fond of the idea that despite the fact that multiple studies show that autistic adults have a very high unemployment and underemployment rate, that somehow there is a significant unidentified population of gainfully employed autistic adults in sufficient numbers to tilt the aforementioned employment statistics in favor of a significantly higher employment rate.

Not gonna happen, though.

naturalplastic
Veteran
Veteran

Joined: 26 Aug 2010
Age: 69
Gender: Male
Posts: 34,154
Location: temperate zone

24 Feb 2014, 10:10 pm

What is the effect on infants of playing Justin Beiber CD's?

NotaHero
Toucan
Toucan

Joined: 23 Feb 2012
Gender: Male
Posts: 252
Location: Manchester

25 Feb 2014, 4:43 am

linatet wrote:

NotaHero wrote:

thanks for the suggestion! I just added this one to my books list.

Surprisingly quite interesting and I learnt some useful questions to ask myself when you hear something being reported before that information at face value.

ouroborosUK
Toucan
Toucan

Joined: 15 Dec 2013
Age: 40
Gender: Male
Posts: 291
Location: France

25 Feb 2014, 7:30 am

No people can't understand statistics. This thread reminds me of that comics

(SMBC has a few other good ones about statistics and misrepresentation)

_________________
ouroboros

A bit obsessed with vocabulary, semantics and using the right words. Sorry if it is a concern. It's the way I think, I am not hair-splitting or attacking you.

linatet
Veteran
Veteran

Joined: 8 Sep 2013
Age: 29
Gender: Female
Posts: 934
Location: beloved Brazil

25 Feb 2014, 9:01 am

Verdandi wrote:

in this case I think it's a matter of hope, not a mistake. Like: "I believe there are many successful autistics, then I can do it too"

linatet
Veteran
Veteran

Joined: 8 Sep 2013
Age: 29
Gender: Female
Posts: 934
Location: beloved Brazil

25 Feb 2014, 9:10 am

daydreamer84 wrote:

linatet wrote:

5) A last example, I was talking to my friend`s father and told him statistically women live longer than men. His answer: "that's not true, my mother died and my father is still alive" :roll:

Yeah, people don't seem to understand that statistics deal with averages and large numbers. They'll say things like "it can't be true that men are better than women at spatial tasks because I know some girls who are really good at spatial tasks." That doesn't refute a statistic showing that on average men tend to be better spatial tasks than women!

yeah, and unfortunately it can go the other way too. For instance, there are two engineers for a job, a man and a woman. Then the employer thinks: " I'm going to hire the man because men have better spatial skills" :roll:

Last edited by linatet on 25 Feb 2014, 9:00 pm, edited 1 time in total.

alwaysnow
Blue Jay

Joined: 29 Dec 2013
Age: 34
Gender: Male
Posts: 90

25 Feb 2014, 11:58 am

naturalplastic wrote:

What is the effect on infants of playing Justin Beiber CD's?

Although not infants, "Justin Bieber's Music Will Make Your Kids Better People, Ridiculous Study Finds"

http://www.spin.com/articles/justin-bie ... udy-finds/

(Refreshing to see an article reporting on these "news-interpreted studies" like they should.)

linatet
Veteran
Veteran

Joined: 8 Sep 2013
Age: 29
Gender: Female
Posts: 934
Location: beloved Brazil

25 Feb 2014, 4:40 pm

zer0netgain wrote:

90% of statistics are made up.

:lol:

AAAAH! NOW I understood your joke ! :lol:

I wasn't the only one that misunderstood it.
you guys have a sharp humour for an autism forum!

Verdandi
Veteran
Veteran

Joined: 7 Dec 2010
Age: 54
Gender: Female
Posts: 12,275
Location: University of California Sunnydale (fictional location - Real location Olympia, WA)

25 Feb 2014, 7:08 pm

linatet wrote:

Verdandi wrote:

in this case I think it's a matter of hope, not a mistake. Like: "I believe there are many successful autistics, then I can do it too"

I've been in too many of those conversations. They're grasping at straws. Whatever the motivation for such an assertion, they clearly do not understand how statistics work and argue that the absence of evidence is evidence of absence.

Something can be hopeful and a mistake. That's why there are sayings like "you can wish in one hand and s**t in the other, and see which one fills up first."

zer0netgain
Veteran
Veteran

Joined: 2 Mar 2009
Age: 56
Gender: Male
Posts: 6,613

26 Feb 2014, 9:53 am

linatet wrote:

zer0netgain wrote:

90% of statistics are made up.

:lol:

AAAAH! NOW I understood your joke ! :lol:

I wasn't the only one that misunderstood it.
you guys have a sharp humour for an autism forum!

And that's with me adding the :lol:

Drehmaschine
Veteran
Veteran

Joined: 27 Feb 2013
Gender: Male
Posts: 781
Location: Bundesrepublik Deutschland

26 Feb 2014, 12:30 pm

[img][800:445]http://payload120.cargocollective.com/1/9/312305/4706840/Threadless1dfs%20copy_o.jpg[/img]

Callista
Veteran
Veteran

Joined: 3 Feb 2006
Age: 41
Gender: Female
Posts: 10,775
Location: Ohio, USA

26 Feb 2014, 8:52 pm

naturalplastic wrote:

What is the effect on infants of playing Justin Beiber CD's?

It increases the speed of teething; the babies are eager to bite the person who made them listen to such a vacuous, unimaginative excuse for music.

Verdandi wrote:

linatet wrote:

in this case I think it's a matter of hope, not a mistake. Like: "I believe there are many successful autistics, then I can do it too"

I've been in too many of those conversations. They're grasping at straws. Whatever the motivation for such an assertion, they clearly do not understand how statistics work and argue that the absence of evidence is evidence of absence.

Something can be hopeful and a mistake. That's why there are sayings like "you can wish in one hand and sh** in the other, and see which one fills up first."

Do remember, though, that there's one thing you can prove with a single example: You can prove that something is possible. If one autistic person has, for example, become the CEO of a Fortune 500 company, we know it is possible, in at least one instance, for an autistic person to do that. The odds might be small; you might have better odds at winning the lottery; but with one example, you can prove something is possible. It may be hope, but it's not completely illogical.

Even small samples aren't totally useless. Give me a properly random sample and a large difference from the control group, and I can still make a pretty good guess even with a small sample. Larger samples improve your chances of drawing the right conclusion, but the bigger the effect is, the smaller a sample you need to find it.

_________________
Reports from a Resident Alien:
http://chaoticidealism.livejournal.com

Autism Memorial:
http://autism-memorial.livejournal.com

Page 3 of 3 [ 45 posts ] Go to page Previous 1, 2, 3

Jump to:

People can't interpret statistics Post Reply New Topic

People can't interpret statistics