Live data from Hacker News

The Utter Uselessness of Job Interviews

nytimes.com

61–70 of 410 posts

Re: The Utter Uselessness of Job Interviews

#61

How do you know that the person you'll be marrying won't cheat on you and won't leave you in hard times? If you apply the methods we use today for interviews you'll end up with a 50/50 chance at best, a coin toss. Yet why do some marriages last forever (till death do us apart) while others fail miserably or crumble even after 20 years? The search for the global optimum cannot be performed by asking a set of questions…

  instinct and sexism/racism are often conflated.
What's perceived as instinct is often the outcome of social conditioning, which gives outcomes such as "People See Black Men as Larger, More Threatening Than Same-Sized White Men" [1]

[1] http://www.apa.org/news/press/releases/2017/03/black-men-thr...

Re: The Utter Uselessness of Job Interviews

#62
post #35
post #21

I've been a programmer for a bit over ten years. I've worked at scrappy little startups, midsized companies, now for a tech giant for a few years. The engineers I work with at the tech giant are consistently better engineers than my other coworkers have been, and I credit the very structured interview process. We're trained to ask specific questions, look for specific types of answers, and each interviewer is evaluat…

Which of the companies paid the best/has the best benefits? If it's the tech giant, it's very possible they attract better candidates because they offer more.

they may also be able to hire more, and let people leave on their own (or let them go later), keeping the 'better' ones on board. Smaller companies can't make the same number of hires in the first place.

Re: The Utter Uselessness of Job Interviews

#63

How do you know that the person you'll be marrying won't cheat on you and won't leave you in hard times? If you apply the methods we use today for interviews you'll end up with a 50/50 chance at best, a coin toss. Yet why do some marriages last forever (till death do us apart) while others fail miserably or crumble even after 20 years? The search for the global optimum cannot be performed by asking a set of questions…

This article is critiquing the unstructured get-to-know you interview. It is not critiquing highly structured interviews with clear goals, nor technical challenges, nor references checks or past performance.

In your example, how would you structure a series of questions and procedures to limit the risk of marrying someone who will abandon you or cheat on you? I think you can apply good interview process techniques to this quite well!

1. Have they been divorced or cheated on someone before? People who have been divorced before are much more likely to divorce again. The 50% divorce rate in the US is slightly misleading, as many of the divorces are concentrated in repeat divorcees. In fact, among younger (early to mid twenties) first-time marriages, the divorce rate plummets to something like 15-20%.

2. Have their parents been divorced or cheated? Children of divorced parents are much more likely to get divorced themselves.

3. Have they been physically, emotionally, or sexually abused as a child? People with traumatic early childhood experiences are much more likely to develop trust issues with long-term partners, especially if they never had extensive counseling.

4. Do they have a good relationship with their family? People who have a difficult or unstable relationship with multiple family members are more likely to see tumultuous relationships as a norm.

This is all equivalent to reference checks in a job.

Then a long dating / engagement period is necessary. How do you they treat you during this period? Do they cheat? Are they abusive? Do they leave you during a period of difficulty? Do you have the same religious views? Do you split housework evenly? Do you both want kids? How do you view money? The three most common reasons for a fight among couples are 1) money, 2) housework, 3) free time (and how to spend it).

People who are otherwise happy and well-adjusted adults who get married and then divorce bitterly after 10 years are not the norm. Most divorces can be predicted. And most divorces happen before 2 years of marriage. If you are aware of the warning signs and are not blinded by a "gut instinct" I think you can definitely minimize the potential for marrying a snake -in-the-grass.

Re: The Utter Uselessness of Job Interviews

#64

> In one experiment, we had student subjects interview other students and then predict their grade point averages for the following semester. Not sure how using inexperienced interviewers proves anything. Would have been more interesting to have lecturers interview the students.

There's a reference in the article that covers that kind of scenario.

  The additional 50 students that the school interviewed but
  initially rejected, did just as well as their other
  classmates in terms of attrition, academic performance,
  clinical performance, and honors earned.

Re: The Utter Uselessness of Job Interviews

#66
post #18
post #3

It should probably have been titled "The Utter Uslessness of Unstructured Job Interviews", because that's the kind of interview the author criticizes. In my personal experience, structured interviews can be very helpful in determining a candidates abilities.

Are you sure you're using the same definitions as the author? In practice, across a pretty decent sample of large tech companies, I've never seen a truly structured interview outside of Matasano, where I was almost murdered by my employees for instituting them. I think we'd hear far more complaints about them if they were common. It's possible that what you consider to be a structured interview is in fact what this a…

> Lists of facts

This has me mulling whether that might be a better approach to administering law-school exams than the traditional analyze-this-hypothetical-fact-situation approach. (I'm a part-time law professor.)

More generally: I wonder to what extent school examinations can draw useful lessons from job interviews.

Re: The Utter Uselessness of Job Interviews

#67
post #60

Known useless indicators: * Resumes * Skills tests (hacker rank) * Whiteboard interviews * Unstructured interviews * Employee referrals No wonder headhunters have such a good business. Not that they're more discriminating, but they can pretend to be the solution to an intractable problem.

> No wonder headhunters have such a good business. Not that they're more discriminating, but they can pretend to be the solution to an intractable problem.

Do companies view them this way though? I've always thought that hiring external recruiters to bring in a bunch of candidates is done to outsource the scummy task of bothering people via phone, linkedin etc.

Re: The Utter Uselessness of Job Interviews

#68

How do you know that the person you'll be marrying won't cheat on you and won't leave you in hard times? If you apply the methods we use today for interviews you'll end up with a 50/50 chance at best, a coin toss. Yet why do some marriages last forever (till death do us apart) while others fail miserably or crumble even after 20 years? The search for the global optimum cannot be performed by asking a set of questions…

If you ask someone to predict a coin toss, and use the methods we use today for interviews, you'll end up with a 50/50 chance at best. Yet some people guess the coin toss correctly. It must be some unmeasured "gut" that's doing a good job for those people, right?

Re: The Utter Uselessness of Job Interviews

#69

Earlier quoted context omitted.

What about when you know what the concept of big O is, know why it's often touted as important, and disagree that it's as important as it's touted? My opinion is that big-O often ends up being used as a premature optimization effort hindering "just get the right answer first". Maybe that brute force method will work in a reasonable clock-time cost, despite having egregious algorithmic cost. You won't know if you're b…

That's a big red flag, at least in a Google setting. The reason is because when your dataset is in the petabytes, any algorithm bigger than O(N log N) is not going to terminate. You're not going to get any answer at all; your MapReduce is going to sit there burning CPU time for a day, and then you'll kill it, and you won't have any idea what went wrong. (This is learned from experience, if you couldn't tell.) In a st…

In my experience as a Google engineer, this is mostly false. None of the work I or any of the people I know well at Google is on petabyte datasets; most people just aren't on the hot path of indexing/websearch/youtube. In fact almost none of the stuff I was asked about in my interview has been relevant to my work. I have thought about complex algorithms two or three times in the past year.

Re: The Utter Uselessness of Job Interviews

#70
I wonder if we can extrapolate to marriages and to how arranged marriages (at least in India) having a higher success rate. Usually the parents on either side decide on a match based on family background, financial stability, education background etc,. rather than letting the to-be married decide.
Post reply on HN