Impossible Interviews

Whilst not impossible, on the topic of "think outside the box questions" I once got asked if I could add all the number from 1-10, on the spot, in the interview.  Apparently the answer "Yes" was not what they were looking for, then after fumbling around with some finger counting I answered it but was quite flustered.  The man explained they asked the candidates this to see how they think on thier feet, thankfully I didn't get the job as I landed a job at the IET just after.


The only impossible question I was asked was if I was going to have children soon - there is really no good way to answer this question which is probably why you're not supposed to ask it in the first place, needless to say I didn't get the job when I tried to tell them something along those lines. 

Kathryn,

Being asked to add up all the numbers one to ten is an interesting one to me as answering it can show the difference between mathematics and arithmetic, the difference between 'smart' and 'slog'.


Slog is adding one to two, getting three and adding that to three etc. Smart is realising that adding one to ten is the same as adding ten to one then realising that adding numbers from each of those 'sums' in turn is the same, i.e. 1 + 10, 2 + 9, 3 + 8 = 11 and that there are ten of those identical sums, so that is 10 x 11 = 110, which is twice the original sum, giving 55. Such a brilliant piece of 'lateral thinking'! If you knew this 'trick' it would have been nice to have bounced back with, "Yes but can you add all the numbers between 1 and 1000?" (half of 1001 x 1000).


The 'pregnancy' issue might once have been just seen as a case of common humanity, the interviewer attempting to establish a degree of empathy with the interviewee. Now it is 'assume nothing, don't ask!', Another reason why perhaps the interview has had its day?

Hi James,


I would not have thought of doing it that way, which is an interesting approach.  I did go through the slog, 3,6,10 etc...


I tend to prefer the competence based approach to interviews for this reason, although you are right that particulary for engineering a lot of companies will arrange things like assessment days or workshops to see skills like teamwork etc in action (at least for graduate or similar roles).

Kathryn,

I deduced from the way you described the question and answer that neither your interviewer nor yourself had been schooled in the Arithmetic Series and just hoped that I was right!

Anyone who could come up with that method from scratch under interview conditions would be pretty smart! Actually it is a good example of why the interviewer shouldn't use these 'clever' questions, how does one know if candidate 'A' already knew the smart answer while 'smarter' candidate 'B' had never met that sort of question before?


What would the reaction of the interviewer have been if you had responded 'instantly' with 55? This candidate is too smart? No-one ever knows if an interview gets the right result, the selection process and subsequent employment 'experiment' can't be run again.


P.S. The arithmetic series isn't just an 'interesting approach', it is an excellent demonstration of what mathematics can do! Try adding all the number one to a million, it will only take about six days. Now multiply a million by a million and one, then halve it. Ten seconds at the most and the answer will be right. It's a good party trick if nothing else!

James Shaw:

Kathryn,

Being asked to add up all the numbers one to ten is an interesting one to me as answering it can show the difference between mathematics and arithmetic, the difference between 'smart' and 'slog'.


Slog is adding one to two, getting three and adding that to three etc. Smart is realising that adding one to ten is the same as adding ten to one then realising that adding numbers from each of those 'sums' in turn is the same, i.e. 1 + 10, 2 + 9, 3 + 8 = 11 and that there are ten of those identical sums, so that is 10 x 11 = 110, which is twice the original sum, giving 55. Such a brilliant piece of 'lateral thinking'! If you knew this 'trick' it would have been nice to have bounced back with, "Yes but can you add all the numbers between 1 and 1000?" (half of 1001 x 1000).


The 'pregnancy' issue might once have been just seen as a case of common humanity, the interviewer attempting to establish a degree of empathy with the interviewee. Now it is 'assume nothing, don't ask!', Another reason why perhaps the interview has had its day?

I would have gone for 10 + (9 + 1) + (8 + 2) + (7 + 3) + (6 + 4) + 5 = 55.  I have played a few card games where your score at the end is the sum of all the cards left in your hand (low scores being better).  My mental arithmetic is terrible, so I find it easiest to find groups of cards that add up to ten, then worry about any odd ones left over.

The issue of how the question is supposed to differentiate between candidates is tricky. Any questions (trick or straight) should give the interviewer some sort of insight into how well or badly the candidate will do the job. Had I been asked to add all the numbers from 1 to 10 in an interview I could have immediately responded with 55 as this is something I know (without having to go through the arithmetic series method, which I also know), but this does not affect how well I do my job either way. I could also point out that being pretty good at mental arithmetic, if I had to do the sum '1+2+3....' I could probably complete it faster than some could do the "(1+10)*10/2" calculation, so how could the interviewer know how I had arrived at the answer? James's sum of one to one million is a better differentiator, but why not make it 1 to 37 or some other random number? I therefore can't see that this question has any bearing on differentiating between candidates.

When interviewing my method is to ask questions to probe the candidates knowledge around the job to be done. Some of them may be 'impossible questions' on the basis that it is exceptionally unlikely that anyone will be coming into the job with 100% of the needed knowledge and they will therefore be unable to answer all of the questions, but the knowledge of the candidates limitations gained at the interview will identify how much training/on-the-job learning will be necessary before the candidate is fully functional. It also identifies candidates such as Andy who admit they don't know the answer but can show they know where to look to find out. This does not mean it is an impossible interview, as I only ever had one candidate who was able to provide all the answers, though given his background I would have been surprised if he couldn't. I went ahead and asked the same questions as the other candidates had been asked to make the process fair (and I had two vacancies so I still needed to consider the best of the other candidates).

Alasdair

I am sure there is a whole topic on useful mental arithmetic tricks that some have been taught and others have developed for themselves. My mental arithmetic has never been very good and I used to calculate approximate logarithms in my head during tube journeys when scoping an idea so as to avoid multiplication. The other trick was to round numbers off say by 20% and compensate at the end of the calculation. Bear in mind that this was for engineering not banking!


It occurs to me that people may have knowledge but don't use it. They have been examination trained, they have the certificate, they can churn it out at an interview but 'on the job' it is never applied. In real life problems don't come well-defined and certainly not with a neat list of tabulated parameters.


When I was studying A-level physics we had a new teacher who thought it was a good idea to set questions in a practical style. We were given a Boyle's Law problem involving a bicycle tyre. The 'number-pluggers' were able to solve it, the 'thinkers' were stumped. "It's a constant volume problem", he explained in frustration. "But it is a rubber tyre, it stretches but we don't know how much." "Not very much.", says he. Academic precision changed to real-world approximation and nothing to tell us that the rules had changed makes for an 'Impossible' question!

An interesting thread that I have only just picked up on,

 

Andy’s quote from Max Eggert largely nails it for me and his other comments are aslo very insightful. Interviews in their various forms have been studied extensively by various types of behavioural scientists , but still remain an something of an enigma. We fall in love, we instinctively like some people and dislike others, we use intuition etc.

 

HR professionals and industrial psychologists developed practices to make selection decisions more structured and “objective”. These include psychometric tests and/or structured “competency based” interview questions. However the reality of most people’s experience of a job selection process is different and probably seems quite “amateur” at times, which it often is.

As in many fields, something  “fit for purpose” applying rules of thumb, might achieve similar results to a costly “professionally optimised” approach at the cost of slightly higher risk.     

 

Prospective external recruits often first come into contact with recruitment organisations e.g. https://www.rec.uk.com/ which at the front end are often sales led and commission incentivised. At this point they are “matchmaking” not selecting, trying to generate a credible pool for their client to select from. If they exclude you it is because they don’t think that their customer will “buy” you. At the next stage they are typically acting as an “outsourced HR function”. Some specialist Recruitment Consultants are highly professional experts in selecting for their field and may interview on behalf of clients. Assuming that they don’t actually  make the final selection, then the responsible manager will interview, although a panel of two or three is common. In some sectors the “agencies” are themselves the employer, renting people to the client.    

 

Internal selection, because the prospective candidate is known and understood should be simpler, but in my experience can be quite difficult. In multi-division organisations candidates and interviewers may not be known to each other, avoiding a political dimension. But in an a local situation there is usually someone in “pole position” and interviews can feel like something of a sham to unsuccessful candidates. “Promotion Boards”, PhD vivas etc. are usually designed to offer a challenge rather than to “fail” people. IET Professional Registration Interviews fall into that category, some candidates are unsuccessful because they weren’t able to fully the satisfy the required criteria, but everyone can in principle succeed.  The success rate of interviews is an important indicator of overall process health.

 

Much of my career was spent with an emphasis on “development” rather than “selection” although obviously the fields overlap. For example an “assessment centre”, commonly used for graduate recruitment could be have similarities to a “development centre” for managers or even “select and develop”.  The key difference as I see it is the psychology of "rejection" for the individual and the implications for others (e.g. the organisation).  

 

Returning to selection interview questions, a structured interview would ensure that each candidate was asked the same obviously relevant questions, with the answers evaluated and recorded in the same way. For example “tell me about a project that you were involved in” with supplementary questions if needed  to tease out relevant issues. “Off the wall” or “outside the box” questions are risky, because no one, often including the questioner, may understand what answers they are actually seeking and how to evaluate them.  If mathematical fluency is relevant to performance of the role, then that is reasonable. But as James correctly pointed out this could easily become a smokescreen for “selecting in our own image”, or at worst forms of illegal discrimination.  

 

James also highlighted the problem of changing relationships. I experienced myself a situation (in a senior role) of going from “hero to zero” under a “new broom” leader. Reorganisation also seems to have become normal rather than exceptional and tenure in companies shorter.   

 

There are plenty of myths and folklore around, including the brilliant “detective” or “legal advocate” model using clever questions or The Oxbridge Professor's intellectual jousting questions. There are proven sales techniques using types of questions.  There are also people who are more naturally insightful about others for different reasons, but many who overate themselves.

 

Ideally the question of “can the candidate do the job” should mostly be answered before the interview by their track record , although we might be verifying or trying to evaluate the extent of any gap. The questions of “will they do it and fit in” are two-way. On that basis my focus would be on trying to develop a good rapport , helping me to understand and “warm” to this candidate. I would like ideally at the end of the interview for both parties to be willing to enter the relationship and if not, know why not. Choice would come after reflecting on the options. A problem with this type of approach is that it is probably best served by being 1 to 1 or at least relatively informal and intimate, which has obvious risks. Perhaps this is the “final” or “second” interview approach?

 

Probably the commonest problem is the “horns and halo effect”, where sub-consciously the interviewer comes to a decision very quickly, often based on superficial impressions and then tends only to accept information that confirms this bias.  Even consciously there can be a rush to judgement and strong characters may persuade others in panel situations. Self-awareness by interviewers is important. For example if I feel instinctively negatively towards someone (which isn’t often, but it happens) I try consciously to find reasons to like or at least to respect them. 

 

I hope to have offered some additional insight without being too long-winded or self-indulgent.  James perhaps you should apply for Countdown if Rachel Riley moves on

 

Roy,

Thank you for your comprehensive summary of the interview process in all its forms. Just picking up on one item out of the many, I certainly recognise the internal interview situation where it can often be a formal cover for 'Buggin's Turn'. The first part of my career was essentially spent 'doing', I was 'the expert'. After gaining another degree I changed employer and became more a commissioner/manager of contracted works. At some point there was an internal vacancy to do a job very much related to what I had done in my 'doing' phase. I must surely be the ideal candidate, proven ability in the manager role and hands-on expertise in the subject! It soon became very clear that the interview was just a paper exercise, my experience was 'out of date' apparently, (good excuse). The interviewer went on to tell me how he was spending millions on a project that was obvious to me from my training and experience couldn't possibly work. I just bit my tongue and decided I wouldn't have fitted in anyway.


Actually that raises a "I wish I had known that earlier" point. One of our management courses covered 'Stages of Delegation', something that doesn't seem to be common knowledge. The theory is that at Level One the boss does everything, the subordinate watches and learns. Going up the levels the subordinate does more and the boss less, until the deadly Level Five. At this point the subordinate knows more than the boss. "You don't want to be there" said our management guru. He was right.


No chance of Countdown - I can't stand the jingle and arithmetic is for machines! Reading Simon's comment, I'm very impressed if he can mentally pair out numbers in the one to ten set!