Hacker Read

hokkos · 2016-05-27 09:14:20

I like how the graph is carefully crafted to make it like there is good age distribution when it does NOT. There is twice a subdivision in ages 20-40 than in ages 40+.

persistent | karma 157 | avg karma 2.57 · | 2019-10-24 19:40:26+00:00

But that's not what the graphs show. There is a diagonal that clearly tracks a birth year cohort, not an age.

detaro | karma 40664 | avg karma 1.79 · | 2021-07-21 06:02:40

it's using age-group specific data.

PeterGriffin | karma 134 | avg karma 1.79 · | 2014-07-07 06:22:48

The reasons the graph is centered around 30s with a halo in 20s, 40s and 50s is the same reason the graph lacks "under 10" and "over 80".

It's just how humans work. We get born, have a period of growing up, gaining experience, then a productive period with big bold dreams, then settling, retirement and death.

I guess without the finger pointing the data would be so boring that it wouldn't warrant an article, so here we go.

reply

nl | karma 29762 | avg karma 2.49 · | 2015-06-05 10:57:14+00:00

Yeah, but a normal distribution centered on 29, with a lower bound of ~18 doesn't actually imply all that many 40+ year olds.

(on mobile, someone else can reply with the math for an upvote)

reply

contravariant | karma 8428 | avg karma 2.4 · | 2016-10-12 20:48:24+00:00

I don't get why the bracketed the age group at all. Surely you can just estimate the distribution?

MichaelMoser123 | karma 4337 | avg karma 1.94 · | 2020-10-28 06:19:39+00:00

that would have to be corrected for different age distributions, somehow. But yes, such a big difference seems to make some sense.

pagutierrezn | karma 228 | avg karma 2.43 · | 2021-05-26 04:25:11+00:00

Might this age distribution merely represent the age of people willing to answer the survey?

amalcon | karma 6231 | avg karma 3.28 · | 2014-07-19 21:18:04+00:00

Indeed. It would also be interesting to see a map bucketed by median age instead of fraction above or below some threshold.

williamstein | karma 6832 | avg karma 5.85 · | 2015-01-14 21:17:30+00:00

I wonder what the age distribution is? (Not just gender and race...)

whatshisface | karma 28019 | avg karma 3.6 · | 2019-02-18 20:33:43+00:00

Wouldn't that just be another expression of age demographics?

lux | karma 1880 | avg karma 2.48 · | 2008-09-01 19:49:05+00:00

Me too! Surprised me, actually :)

I wonder how many other people in my age group there were, it didn't say... It does look like men from 20-29 mostly fared really well according to the graph, but it would be cool to see more detail anyway.

reply

yaacov | karma 1183 | avg karma 7.21 · | 2019-04-10 13:14:03

Wow, that graph is atrocious.

Another issue is that I can’t tell what age groups are included. What on earth does “modal age 23-28 mean”

reply

shoto_io | karma 2061 | avg karma 3.83 · | 2021-07-06 15:11:24+00:00

I’m pretty sure this can be explained by the age distributions. Key part is page 44 where they calculate risks.

BitGeek | karma 373 | avg karma 1.9 · | 2007-04-21 06:12:56+00:00

How does this compare to the age distribution of applicants?

eam | karma 901 | avg karma 2.48 · | 2009-10-31 17:26:41

This should be breakdown into age groups to make it more accurate in a sense.

searealist | karma 434 | avg karma 1.0 · | 2023-01-25 22:09:37

Why do you mention middle aged men when that graph contained men of all ages.

jletienne | karma 67 | avg karma 1.97 · | 2022-07-26 21:39:56

I think it does though, ages are right skewed. The median is likely under 29 in this example

jqpabc123 | karma 4014 | avg karma 1.18 · | 2022-07-19 00:02:43

Look at the charts. They specifically say "age normalized".

fighterpilot | karma 2346 | avg karma 2.24 · | 2021-04-22 08:03:38+00:00

You really do have to segment by age groups when doing this analysis.