Post

Conversation

Arpit Gupta
@arpitrage
Devastating evidence against pure signaling explanations of higher education. Apparently people learn things in College which improves labor market success.
Quote
Florian Ederer
@florianederer
App usage is - contagious (sd increase of roommates’ in-college app usage raises own by 4.4%) - detrimental to academic performance and labor market outcomes (sd increase reduces GPAs by 36% and wages by 2.3%) App restriction policy would boost wages! nber.org/papers/w33054
Show more
Image
120.1K
Views
David Watson 🥑
Post your reply

Arpit Gupta
@arpitrage
Implied elasticity is something like 1 SD in within-cohort major GPA -> 6.4% change in wages (if you think all the wage impacts come through GPA)? Pretty large!
3.9K
Arpit Gupta
@arpitrage
Of course, you might think the signaling happens via your school rank rather than just the pure sheepskin effect or getting in, but now we are getting very colinear with human capital itself
3K
Ryan Bourne
@MrRBourne
What if susceptibility to app contagion/addiction just correlates strongly with susceptibility to being a crap worker? Are there controls (haven't read the paper)?
975
Arpit Gupta
@arpitrage
Agree, we can’t just assume results from the niche market of the US extrapolate to the much larger and more important context of China
425
Arpit Gupta
@arpitrage
Quote
Arpit Gupta
@arpitrage
Replying to @arpitrage
Implied elasticity is something like 1 SD in within-cohort major GPA -> 6.4% change in wages (if you think all the wage impacts come through GPA)? Pretty large!
953
Flameo
@BarneyFlames
fact that they reduce GPAs massively but wages only a tiny bit suggests signalling is still very big portion of it
470
Square profile picture
Fordham Institute
@educationgadfly
Ad
Around 500 chronically low-performing schools have experienced substantial enrollment declines since the pandemic. Could they be strong candidates for closure? Learn more (including which schools made the list) in Fordham’s new report by
@sofoklis_goulas
: fordhaminstitute.org/national/resea
Read our latest report to find out more.
From fordhaminstitute.org
658K
BensenHsu
@BensenHsu
The paper examines the effects of mobile app usage, especially gaming apps, on college students' academic performance and labor market outcomes. It also investigates the impact of peer influence on app usage. 1. Mobile app usage is contagious - a one standard deviation (s.d.)
Show more
Image
147
David Piegaro
@davidpiegaro
Or what if the app addiction makes them worse workers and it’s not related to learning - on the phone at work. Not devastating evidence at al, at best a slight update away from signaling but one paper shouldn’t move the needle much, especially not this one.
486
Sean X. Luo MD PhD
@seanluomdphd
No -- it's not "devastating evidence". The paper has a N of 20, with completely ridiculous variance estimates. Utter garbage piece of study. I would suggest that it's good practice for any serious academic to actually read the paper before propagandizing it on social media.
54
Vankous
@vankous
If susceptibility to digital distraction or even just other unobservables influencing susceptibility to peer influence are heterogenous and important for workplace productivity, couldn't GPA here just signal that? Although maybe the effect size wouldn't work out?
60
evan
@evanjdoremus
Am I trippin or does a 36% drop in GPA leading to 2% lower wages seem like learnings and grades don’t matter that much?
72
Renkitt 🦁🦁🦁
@Widerife81
Isn't app usage just a proxy for less conscientious? Wouldn't these ppl have always found some distraction? Isn't it striking the impact on GPA is so much higher than on wages? Wouldn't that suggest learning stuff is important to get a grade but not hugely for wages?
101
Angel Diaz
@angelinthemarbl
There is no one advocating for a pure signaling explanation anyway. Bryan Caplan has it as 80% signaling and 20% other stuff. Also, if your hypothesis is true, what are the things people learn, and why must one attend college to learn them?
96
bookdepth
@bookdepth
Findings from the Chinese labor market don't imply that signalling theories of higher ed don't apply in the US
440
Sal 🌐🇺🇸🇵🇪
@GirlBossGuru
Still need to read the paper but how does it differentiate between private returns from signaling and human capital?
231
Show additional replies, including those that may contain offensive content