1 hour ago, JohnSnowsHair said:

Do you know how percentiles work? 

A bit less than 3 million are admitted as freshmen each year. 0.1% of that would be a bit less than 3k. 

Based upon reading the source study, the income percentile was based on the nation income distribution bins and not the admitted Freshmen population.

12 minutes ago, BBE said:

Based upon reading the source study, the income percentile was based on the nation income distribution bins and not the admitted Freshmen population.

That's how I read the chart as well.

20 minutes ago, BBE said:

Based upon reading the source study, the income percentile was based on the nation income distribution bins and not the admitted Freshmen population.

Right. And if acceptance rates are so much higher for those in the 0.1% then you're probably looking at an even higher number.

Children of college educated persons who tend towards higher income levels also tend to go to college at higher rates. 

The point was that 0.1% isn't "17 people"; it's actually a dispproportional percentage of the overall freshman population. so if it's say 0.15% (probably higher but whatever) of the freshman population are in the 0.1% percentile of income, now you're at 4500ish individuals. 

34 minutes ago, JohnSnowsHair said:

Right. And if acceptance rates are so much higher for those in the 0.1% then you're probably looking at an even higher number.

Children of college educated persons who tend towards higher income levels also tend to go to college at higher rates. 

The point was that 0.1% isn't "17 people"; it's actually a dispproportional percentage of the overall freshman population. so if it's say 0.15% (probably higher but whatever) of the freshman population are in the 0.1% percentile of income, now you're at 4500ish individuals. 

No, you have to look at the sample composition. 

 

And the y-axis is rate of admission, not total admissions.  

 

You really need to look at the paper the article is sourcing.

 

27 minutes ago, BBE said:

No, you have to look at the sample composition. 

 

And the y-axis is rate of admission, not total admissions.  

 

You really need to look at the paper the article is sourcing.

 

I understand that. 

The y-axis is not the rate of admission, but the likeliness of admittance against expectations based on standardized test scores.

Roughly speaking, an applicant from the top 0.1% with an 1100 SAT is over 2x as likely to get admitted vs the mean acceptance rate for an 1100 SAT earner. 

Applicants aren't evenly distributed across parental income brackets, but I'm going out on a limb and supposing that families in the 0.1% percentile are sending a higher percentage of their children to college than most of not all other percentiles. If that assumption is not off base, then projected out at least 3k of the freshman class comes from 0.1% families. Probably closer to double that. 

Yes I'm extrapolating, but if anything I'm doing so very conservatively.

You are also assuming the super wealthy are applying to college at the same rate. It would not shock me one bit that the super wealthy have a lower application rate than middle class who, you know, need to get a job

I’d also imagine a major factor here is that wealthy parents are more inclined to be making donations to the school and/or have a higher chance that their child is a legacy admission.  
 

I didn’t read the study yet to see if that was accounted for.

12 hours ago, paco said:

You are also assuming the super wealthy are applying to college at the same rate. It would not shock me one bit that the super wealthy have a lower application rate than middle class who, you know, need to get a job

this is acceptance rate, which limits the sample to whose that apply. those who are applying to Ivys in the 0.1% have twice the likelihood of getting in than a socio-economic normy with the same test scores. 

so while there may be lots of 0.1% kids who just chillax on their parents dime and eschew higher education, only those that at least apply show up here. 

They tended to have higher SAT scores and finely honed résumés, and applied at a higher rate — but they were overrepresented even after accounting for those things.