The Selection of Talent: Evidence from Ethiopia [pdf] | Hacker News Reader

The Selection of Talent: Evidence from Ethiopia [pdf]

docs.wixstatic.com

67 pointsblopeur22 comments

Anyone know what font is used in that article? It's beautiful.

> Further, we measure the quality of the individuals who apply for the job through a battery of personnel selection tests that capture cognitive ability, non-cognitive ability and relevant work experience. These tests are reliable predictors of work performance and are used by firms worldwide (Heckman et al., 2006; Autor and Scarborough, 2008; Hoffman et al., 2015).2

> We use the Raven and Stroop tests for cognitive ability (Schmidt and Hunter, 1998). For non-cognitive skills we administer the Big-5 personality test and the Grit scale (John and Srivastava, 1999; Duckworth et al., 2007).

Does evidence really exists that these tests are predictive?

https://www.fontsquirrel.com/fonts/tex-gyre-pagella

I opened the PDF in Acrobat and looked at the document properties, which has a font tab with the embedded fonts.

As a bonus it's apparently a free font.

Thanks for doing the dirty work. This gets an upvote, and everyone else gets the opposite until this is the top comment.

The font is Palatino (or some shameless Palatino clone). The lowercase t and a (among other letters) are very distinctive.

https://en.m.wikipedia.org/wiki/Palatino

Hermann Zapf was one of the master type designers of the 20th century.

Very close, the font is TeX Gyre Pagella which was based on URW Palladio L, which was itself designed as a Palatino replacement.

That's one of the shameless Palatino clones, of which there are several. They're all basically indistinguishable, all inferior to the original metal version, and all ethically dubious. (Exactly copying someone's years-long labor of love and then slightly changing the name and redistributing it as a new product might be the norm for some type foundries, and the precise legal questions are murky, but it's pretty scummy.)

If you want to ethically typeset something in Palatino, try Palatino Nova. https://www.linotype.com/615716/palatino-nova-family.html

You could use e.g. WhatTheFont to find the font.

https://www.myfonts.com/WhatTheFont/

To answer your question about the font, I think it's the same font used by the current ACM LaTeX template. The font is called Linux Libertine[1].

[1] https://www.dafont.com/linux-libertine.font

This is very obviously wrong. Just look at the letters.

Again, it was just a quick guess. I have a terrible eye for these kinds of things :P

> Does evidence really exists that these tests are predictive?

Currently IQ is considered the best predictor.

If you have two applicants around the same IQ but one has higher conscientious then it could be a decider in a company, but why is it needed in this study I can't see.

No idea how good Raven and Stroop tests are for predicting IQ?

I have a suspicion that if this were a widespread practice the effect would disappear.

It sort of is widespread. Some of my engineering classmates were flown to other countries for job interviews at the employer's expense. It's probably most significant for poor people like students and Ethiopian clerical job applicants for whom application costs are high compared to their income.

The article was beautiful and I would really like to know myself if the tests have been found to be predictive.

I'm not sure I follow; the authors claim that paying job applicants for applying attracts better talent? And they further claim that Ethiopians cannot afford the cost of generating an application?

Somehow I doubt that these results would translate to white collar jobs in the first world.

I can absolutely imagine this being an effective strategy for companies hiring for high-talent jobs.

I have 15 years experience, and there's competition just to interview with a company. In my last job hunt I talked seriously to maybe 10 or 11 places. I did 7 onsites, two work-weeks of interviewing. But picking the ones I wanted to do on-sites with was close to random. All I usually had to go on was a 15-minute call with a recruiter.

I'm quite happy with where I've landed, but in the end I just didn't have time to talk to every company that sounded interesting. Who's hiring is a smorgasbord and lord help you if you do well on Triplebyte's second round and have 30 companies willing to do an on-site. I had to pass on interviewing with several companies that are quite interesting just because I couldn't be interviewing forever.

At some point, you just have to pick a few to dive deeper with, and it's obvious that it's random. If one had been willing to pay me to go through their process, that would have affected my odds of doing an on-site with them quite a bit. And isn't getting people in the door part of the process for recruitment?

The general concept is "friction" and it surely exists in developed countries and professionalized labor markets too, although the specific problem that they point at in the abstract is "credit" (being able to bear the time costs of applying to new jobs) which is probably a less significant factor in some other labor markets. But applying for jobs can be extremely time-consuming everywhere.

I'd be willing to try this for white collar jobs in the first world.

As a developer, your time is valuable. We'll compensate you for helping discover if we'd be a mutual fit.

1. I'd offer NYC-based K8s-savvy developers a $100 Amazon gift card for a CV and passing a phone screen, no obligation to interview on site.

2. I'd offer NYC-based K8s-savvy developers a $500 Amazon gift card for getting to an offer after an on-site interview day (more like half day), no obligation to accept offer.

Types of folks I'm looking for is listed in my profile.

In theory it sounds like a good idea, but I just wonder, what stops people from gaming such a system if it becomes commonplace?

$500 for a couple hours of trouble, assuming I'm local, isn't a bad deal.

They'd have to be qualified enough to get an actual offer. He didn't say anyone interviewing would get the $500. Not sure what sort of gaming could be done with that criteria. And with that carrot for qualified people, he's likely to get more applicants than otherwise, which is the point.

Well, not every developer is making a ton of cash, and a fair few qualified ones may be quite happy to do an interview every day or so for a bit of extra cash.

Heck, if enough companies offered $500 for a successful interview, some people would probably end up making more from going to interviews than they would accepting the offer.

that's the point.

the company lose more for time with the unfilled opening.... and you might like the offer.

Reply on news.ycombinator.com