diff --git a/blog.html b/blog.html index cc11348..eab23c8 100644 --- a/blog.html +++ b/blog.html @@ -25,7 +25,6 @@

a violent introduction to linear regression - statistics for evil the reading itinerary
diff --git a/blog/statistics_for_evil.html b/blog/statistics_for_evil.html deleted file mode 100644 index 2a9bf0a..0000000 --- a/blog/statistics_for_evil.html +++ /dev/null @@ -1,84 +0,0 @@ - - - - - - luisa's blog - - - - - - - - -
- -
-
-

- statistics - for - evil -

-

july 27, 2024

-
-

did you know that hong kong's higher life expectancy relative to india is directly related to their meat consumption?

-
-
- -

of course you didn't! someone on the internet made up that correlation.

- -

actually! it is not even a correlation! mathematically you can draw a line through any two points, i could be plotting the number of birkenstock sales in the united states and the ocurrence of foot fungus per 1000 people, and those two dots would still make a line.

- -

the original statistic on meat consumption and life expectancy was actually part of a post influencing people to a carnivorous diet. apparently veganism is no longer the hip and healthy thing to do, because eating raw sticks of butter and consuming unpasteurized dairy is the trend now. i actually have too many opinions about this, but i won't babble on.

- -

anyway! here's a couple i collected from doomscrolling on reels

- -

the average american made $106,400 in 2022

- -

the average income per house hold in the united states in 2022 was $106,400, according to table A2 on income in the united states, made by the united states census bureau. this is correct but misleading - the fact that the average is 106,400 does not mean that the average joe is making 106,400.

- -

data distribution matters. especially when 1% of the population holds 30.4% of the country's wealth as of Q1 2024. here, a different measure of central tendency, the median, is much more appropriate, and that number is $74,580.

- -

measures of dispersion would give nice insight, but if your attention span can only capture one number at a time, let it be the median.

- -

nosotras period pads are up to 100% leak proof

- - this one's just outrageous. thank you for telling me this fact on your packaging! without this helpful statistic i could not have ruled out the possibility that this product might be -9% leak proof or 107% leak proof. - -

people who make their bed every morning are 206.8% more likely to be millionaires

- -

*olivia rodrigo's making the bed plays*

- -

this particular reel reeked so much of bad statistics i spent 20 minutes digging into where it came from.

- -

so hold on tight.

- -

this particular quote was not sourced to anywhere, but google searches led me to this CNBC article citing this guy, randall bell, who is a real estate damage economist. in an article titled "7 'rich habits' of highly successful people, from a man who studied them for 25 years" , they write that randall bell surveyed 5000 people and "statistically correlated various habits with different meaasures of success."

- -

i actually think this guy made up that title, if you search "real estate damage economist" you get his wikipedia page and his research group, it's a one to one mapping.

- -

i'd like to say that i have no problems with the cnbc article, if anything it's factually correct. millionaires do excercise, and do read, and they wake up early, and they have etiquette, and they write things down. sure. my problem is with reproductions of that article or citation, purposely worded to imply causation.

- -

titles like "want to be a millionaire? make your bed", and "making your bed can make you a millionaire" spin correlation into causation. which is arguably the deadliest sin of statistics. even GPT knows it's fake:

- -
- -
- -

thank you for listening to my rant about bad statistics on the internet. remember kids, don't believe everything you read, not just because statistics can be used to tell false stories, but because you can also just make things up:

- -
- -
- -
- -
- - \ No newline at end of file diff --git a/blog/the_reading_itinerary.html b/blog/the_reading_itinerary.html index 918a6e4..f17f91e 100644 --- a/blog/the_reading_itinerary.html +++ b/blog/the_reading_itinerary.html @@ -31,7 +31,7 @@

-

i read prolifically when i travel. i went to many gorgeous places this year and read many brilliant books that left me sitting in existential silence on cars, planes and trains.

+

i read prolifically when i travel. i went to many gorgeous places this year and read many brilliant books that left me in existential crisis.

without further rationale, here is my map of my reading itinerary for the first half of 2024, and a brief sometimes not so brief review of each of the books and where i read them.

diff --git a/resources/where_do_you_draw_the_line.jpeg b/resources/where_do_you_draw_the_line.jpeg new file mode 100644 index 0000000..18167e1 Binary files /dev/null and b/resources/where_do_you_draw_the_line.jpeg differ