# Does recruiting have scientific basis? [closed]

Recruiters look at many particular signs in a candidate and draw conclusions that, sometimes, look shallow, sometimes odd.

For example, let's take this question: If you provided an incorrect answer to a technical question on an interview, should you respond with a correct response? The answer given can be reworded as:

If you answer a question when it's too late, you are considered obsessive.

Why can't this be interpreted in other ways (e.g.: you look interested in performing a good interview)?

That was just a random question taken from the most voted ones with the interviewing tag. Actually, I'm not particularly interested in the specific case.

What I really want to know: is the interpretation of interviews based on scientific thesis? Do recruiters follow some established theories or models?

If yes, what are those theories and models?

If no, why? How can recruiters (and companies behind them) be sure of their judgment?

Personal note: it looks like to me recruiters are more obsessed than applicants

## closed as not a real question by IDrinkandIKnowThings, Rarity, CincinnatiProgrammer, jcmeloni, Adam VFeb 4 '13 at 19:55

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form. For help clarifying this question so that it can be reopened, visit the help center. If this question can be reworded to fit the rules in the help center, please edit the question.

• This is a really interesting question. I suspect the answer is "no, personal/company experience" however. – enderland Feb 4 '13 at 16:21
• en.wikipedia.org/wiki/Job_interview has some links for background on Job Interviews though I'd question how scientific are you wanting to get here? – JB King Feb 4 '13 at 16:33
• Something to consider with the technical question is whether the recruiter understands enough context behind the question to really allow one to correct their answer. If not, this is where one can seem obsessive as it is kind of like the average driver wanting to fix their car without realizing all the parts within it. – JB King Feb 4 '13 at 16:42
• @heyhey Keep in mind that you're dealing with a social science. A lot of it is statistics saying "When we tried doing X, the result was Y with this condifence interval... which seems to be better than Z." These studies are then used to validate behaviuour models, and then the behaviour models are used to direct action. But it's not determenistically-scientific in a way that pure math is. – MrFox Feb 4 '13 at 16:54
• @Chad: actually this is a simple yes/no answer plus a request of explanation of the answer. And there seems to be an answer: yes: microeconomics & management. – hey hey Feb 4 '13 at 17:35

The TL;DR version of this answer is: Yes, there's two fields that are invovled, Microeconomics (imperfect markets, game theory) & Management (case studies for hiring/people management). These are big fields, and your question deals with many sub-branches of them.

In a pure free market a company should keep adding resources until the marginal benefit of the resource equals the marginal cost. Adapted to employees, this means that if I hire someone who will increase the profits of the firm by 100k, I would be willing to pay them up to 100k (at which point I will be indifferent between hiring them or not, doing so will be neutral to the value of my firm).

Things get interesting when you're not in a pure free market. Specifically, two things do not really translate into the real world: we do not have perfect information, and the quality of resources is very non-uniform.

The lack of perfect information creates bargaining problems that branch into various game theory simulations. When hiring people, you don't actually know the value of their work, and they don't actually know the marginal benefit that they can bring to you. Additionally you don't have perfect transparency of the market. You don't know who else is out there, and you don't know how many everyone out there makes. Both recruiters and candidates have to guess these things during the course of the negotiation and using bluffs can affect the other party's expectations.

Back in the day when I was in school we had textbooks with case studies for that last part. I tried googling an example now, but I can't find anything decent. I suggest you visit your local university and type in "hiring case study" to one of the library terminals. This is a heavily studied subject.

Now, do recruiters follow any this? Who knows. Some might, others are just ex-used car salesmen looking to smooth-talk someone into maxing out their commission.

Recruiters look at many particular signs in a candidate and draw conclusions that, sometimes, look shallow, sometimes odd.

To begin with, very few companies have employees that do interviewing full time. For almost every recruiter and interviewer you meet, you encounter someone who might range from an interview once a year to never before meeting you, to interviews many people every day.

Very few companies use structured intereviewing to recruit candidates. You are more likely to encounter technique in phone surveys, or in the computer-based personality tests that many retailers and entry level companies screen applicants with (read Punchinin In for more details). Typically retail companies have to hire lots of staff due to high turnover, so they have a vested interest in determining what sort of personality does well, and what fits badly. This means that the sort of person who would do well at Home Depot for example, is very likely to do poorly at Abercrombie Fitch; both stores have wildly different customer bases and product lines.

More and more intereviewers appear to be using competency based interviewing. These are less about the details of what you know, and more about how you think and solve problems. For people who memorize a lot of things, these sort of interviews are very hard to pass.

One of my pet research areas is in decisionmaking. People do not make decisions in logical, repeatable manners. Interviewing and recruiting employees is just another example of this human foible. For an introduction to why this is the case, I recommend 3 books:

Sources of Power. This book describes a new theory of decisionmaking (recognition primed decision) that claims you recognize patterns in previous instances and use those memories to make decisions.

Predictably Irrational. The author does research in behavioral economics. So this book is a summary of some of the research he and his colleagues have done on how people make decisions that make perfect sense at the time yet are quite irrational from the economic concept of "rational consumer."

Checklist Manifesto. Our lives are rather complicated, and we tend to forget things at inconvenient times, especially when things are framed in particular manners (advertising and politics use framing to divert your attention from what is important to you).

• +1 for coming up with some reading. I think this answer is better than mine. – MrFox Feb 4 '13 at 18:50

Of course it is not scientific. Yes recruiters and hiring managers make a lot of poor choices. What they are hoping for is that they will err on the side of excluding a good person rather than erring on the side of hiring a bad one.

There is sometimes some pseudo science involved (personality tests come to mind or those designed supposedly to find who will steal from the company) but these are more to remove the blame fora bad hire from the hiring offical. If you do something that looks scientific then obviously a bad hire is the fault of the process not your judgement. There is a large industry of people selling these products, but the ones I have run into are shaky at best. For instnce I failed one because my answers were too positive. They apparently didn't believe that I don't drink to excess, don't beat up people, don't come to work high, and don't steal even little things. I personally would not have thought that behavior was unusual. But it did serve to show me that I didn't want that job anyway.

There is nothing scientific about hiring and there never will be because a huge part of it is not only assessing skills (which isn't even easy to do in a scientific way or lots of current developers would not have jobs!) but assesing how the person will fit in with the existing employees (totally subjective judgement) and how the person rates in comparison to the others you interview.

Yes answers that will get you hired one place will be the ones that throw you out of competion somewhere else becasue we are all looking for something differnt and we are all assesing a diffent group of candidates. Yes different hiring managers will interpret various actions differently (personally I am perfectly happy if someone leaves and interview and then takes the time to find the correct answer to something they knew missed). My solution to that is to be myself. A company that likes who I am is one I am far more likely to be successful at anyway. One that doesn't like who I am is a place where I won't want to work anyway no matter the salary or benefits offered.

• Of course it is not scientific Citation needed? I suspect there are quite a few companies which do internal research on these sorts of things and I wouldn't be overly surprised if the academic world has research on interview techniques/tactics. – enderland Feb 4 '13 at 16:36
• Doing research doesn't necessarily imply that it's scientific research, especially for companies doing their own internal research. – René Wolferink Feb 4 '13 at 16:40
• @HLGEM You posted this at roughly the same time I did, but I like your answer better. – maple_shaft Feb 4 '13 at 16:47
• "There is nothing scientific about hiring and there never will be" - Holy Lets Not Generalize, Batman! Are you saying that there's no statistical correlation between fizz buzz and ability to code? Or no correlation between past performance of salesmen and future performance? Or frequency of being late for assembly-line workers (a very expensive situation) with future expected frequency of being late? – MrFox Feb 4 '13 at 18:26

They probably do have models, in much the same way that financial analysts had models about derivatives of mortgage backed securities reducing the inherent risk of subprime mortgages by bundling them with other mortgages.

We see how well that worked out for them...

So the real question is why do they choose to believe them? There is an underlying motive in forming a pseudo science behind something that is inherently a chaotic messy process. For hedge funds and big banks, this motive was to offload subprime mortgages that nobody wanted to buy. For an HR person their goal is to reduce as many potential qualified applicants as possible by reading too much into things.

The dilemma for them is that they post a job opening and they get flooded with 90 resumes in a week. Easily 70 of these are clearly underqualified or pure garbage. A software process can weed out potentially 5-10 more that just aren't as good a fit, but that still leaves as much as 15 really awesome applicants that need to be interviewed.

Nobody wants to just draw names from a hat, but essentially when you are disqualified for something small and meaningless like this, then that is exactly what is happening. They are formulating a pseudo science so that they don't have to admit to you or themselves that they are ruling out qualified applicants randomly.

• The first half of your post sounds like a political rant from circa 3 years ago. Just and FYI, bundling mortgages with different risk envelopes into one security makes for a lot of sense to a lot of people who need to build hedges. The reasons for the financial crisis were much more complicated than 'bad risk models'. – MrFox Feb 4 '13 at 17:00
• @suslik It is only political if you make it be. I was trying to use a famous example that faith in scientific models for highly complex problems with infinite variables is akin to rolling bones. Would you have perhaps changed your downvote if I said it was also caused by government forcing banks to provide loans to low income housing? – maple_shaft Feb 4 '13 at 17:08
• My understanding is that models in quantum physics also keep running into trouble, and the math there just goes way over my head. That doesn't make it any less of a science, right? My other problems with this answer are: a) Just because the problem is complicated, the study of it does not make it a pseudo-science. b) The last paragraph is a bit of a stretch, I mean you just outlined an entire conspiracy theory about HR at large. On an emotional level, I want to agree, but I think it's a bit far out. – MrFox Feb 4 '13 at 18:20