Early-career setback and future career impact | Hacker News Reader

Early-career setback and future career impact

nature.com

133 pointshhs26 comments

> ..Early-career setback appears to cause a performance improvement among those who persevere.

> ..These findings are consistent with the concept that “what doesn’t kill me makes me stronger,” which may have broad implications for identifying, training and nurturing junior scientists.

Good to hear, as someone who (is not a scientist but) has had a fair share of early-career setbacks, from external causes, bad luck, as well as self-inflicted in youthful ignorance. :)

At least from an anecdotal retrospective of my current blessed situation, surviving and overcoming these setbacks have indeed made me more resilient and persistently motivated.

I like "We are the sum of our scars." And OK, it's from Stover's Caine novels.

Also, we wouldn't have fingers and toes if intervening cells in the early limb bugs hadn't died.

Damn. I meant "buds".

Funny personal story:

In 2008 i applied for an IT role at News Corp in Sydney Australia, i didn’t get the job. In 2015 they (one of their subsidiary companies) acquired the company that i founded with two other partners. A much better outcome.

I’m sure I’ll feel silly when someone finds them, but are the concepts of “narrow win” and “near miss” defines in the context of this paper? Is “career setback” defined?

In the abstract “Here we examine junior scientists applying for National Institutes of Health R01 grants. By focusing on proposals fell just below and just above the funding threshold, we compare near-miss with narrow-win applicants”

In other words, scientists who just barely missed funding “near miss” versus scientists who just barely were over the threshold to receive it “narrow win”.

Thank you

Or it’s just a filter for perseverance.

The paper tests this hypothesis and finds that there's probably a causal effect:

"...these results demonstrate that the screening effect may have played a role, but it appears insufficient to entirely account for the observed difference between near misses and narrow wins."

+1. The dual populations of dropouts and higher performers smells like survivorship bias.

Or researchers who had near misses are at institutions (or moved to institutions) that have sufficient internal resources to get early career researchers bridge funding to keep them productive during their funding lull.

I think there are so many ways for people to get lost in the current finding environment, that would be extremely difficult to be able to make any findings to how people performed. The main exception I see is the attrition rate. This is the measure they reported that I have the most confidence in.

As was said elsewhere, everything else just smalls of a survivorship bias.

Code and data is here:

https://yang-wangnu.github.io/setback/

Come work at Amazon. If you can survive at Amazon, you can literally work anywhere afterwards (assuming they hire you and you’re not an asshole).

I like the conclusion but I have to dispute it: there is a snowball effect for motivated (and resilient) people. Maybe this shows that most people are not motivated and need a kick in the pants, but if you are clever and you get ahead, you can use that to get more ahead (otherwise you were not really ahead at all)

Figure 2 and 3 legends don't start at zero. What in the world?

Another funny personal story: since someone shared one.

Started developing a taste for studying and computers at age 12-13. Did not hit a stumbling block till 40. Oh yes there were bad times but it was always one valley to another peak and at 40 the tide turned. I would not say it was the Midas touch, it was not, there were others way more successful but I look at my growth based on my aspirations and capabilities.

And then the fall. Took 5 years to realize the fubar situation and then maybe 6 years of perseverance with a course correction ;) Even before the article I realized that the lack of failures in early days probably hurt me in the long run.

Once again, this is a very simplified version. Even I could find flaws in my presentation of details if I needed to.

Anecdote: My company has recently passed on a few early career candidates who pretty clearly hadn't been mentored.

We have a take-home assignment as part of our interview process and they were able to get a working solution, but missed on design patterns that would make their code more maintainable and extensible, things that are usually learned from code-reviews by people who have had the experience of working with bad code.

This was more common in candidates who came to us from the tech departments of non-tech focused companies.

We lamented not being able to hire otherwise smart and personable candidates, but at the end of the day we only had headcount for 1 employee and we went with a different smart and personable candidate who had better early career opportunities.

So you allocated scarce resources to the best available candidate you could find? ...or there's something I'm missing...?

The parent is pointing out that they continued the pattern of the setback by denying a candidate the opportunity to learn the skills whose absence resulted in their denial. It's a catch-22 which for them started with their first job out of school.

Reading it, there's at least a 50% chance that learning at this company would've made them worse developers that focus too much on stuff that doesn't really matter.

> Anecdote: My company has recently passed on a few early career candidates who pretty clearly hadn't been mentored.

Either that or the candidates you passed on simply didn't want to spend several hours making enterprise-grade software for a take-home test that they're not getting paid for.

This. This. This.

Everyone thinks they can interpret their tests perfectly. Ultimately, the candidates are likely better off. The coded message here is not that they didn't use obvious design patterns but that they didn't code in the way the employer wanted. Which...I suppose is a perfectly fine reason to pass on a candidate.

I'm just sharing my experience as it relates to the article, no need to be inflammatory. Take home assignments are a pretty common assessment tool and many candidates prefer it to getting grilled on DS/Algos on a whiteboard.

Also, I would love to see the quantitative evidence that the value of the design patterns in maintenence are. My guess is that in fact this is a random assertion made by a single engineer.

> we went with a different smart and personable candidate who had better early career opportunities

I just had a similar situation. Was sort of stunned at the number and quality of candidates who applied. One I had to cut loose was just so good I couldn't help but offer to have a follow-up discussion about things she could work on. Curious, is this an HR no-no?

> We have a take-home assignment as part of our interview process and they were able to get a working solution, but missed on design patterns that would make their code more maintainable and extensible, things that are usually learned from code-reviews by people who have had the experience of working with bad code

Design patterns? This is a red flag for me. How do you know you're not actually a bunch of unproductive cargo-culters? Those applicants might've dodged a bullet with you there.

> We lamented not being able to hire otherwise smart and personable candidates, but at the end of the day we only had headcount for 1 employee and we went with a different smart and personable candidate who had better early career opportunities.

I see. Did you ever actually ship anything, though?

Reply on news.ycombinator.com