Fair Selection - who is it fair for?
"Fair Selection" is the most common name for a system of job advertising, short-listing, and selection by interview which purports to be entirely fair, by way of giving every applicant an equal chance of being successful at getting the job. It purports to rule out racism, sexism, ageism, etc. However, is it really fair to all? This hub explores the issue.
How does it work in theory?
The concept behind Fair Selection is that you start by writing a Job Specification which contains as many of the duties of the role as possible. Most Job Specs give themselves some latitude by stating something like "This Job Spec does not contain a comprehensive list of duties and responsibilities, but serves to list the main duties and responsibilities of the role." That get-out clause allows a company to change your Job Spec once you are in a role, without you being able to claim a salary increase, or other compensation. It also allows the Job Spec writer some latitude if they forgot to include something.
It will also include a "Person Specification" or list of "Requirements" which is a list of the skills and experience that the successful applicant should have. Some may be listed as "Essential" and others as "Useful", "Optional", "Advantageous", or some other term that suggests that having those skills or that experience will be an advantage.
Then the Job Advert is written, and, in terms of describing the role, it will not include anything which is not in the Job Spec.
Short-listing is done against the Job Spec on the basis of whether or not the candidate can demonstrate from their CV and Cover Letter that they meet all the requirements of the Job Spec. In the current climate of many good applicants for very few jobs, you have to meet every requirement, or you get rejected. In times of fewer applicants for more jobs, meeting 75% would have been enough.
Finally, a Job Interview "script" is written, which only asks questions directly related to the content of the Job Spec - i.e. questions are asked which directly test if the applicant has the skills and experience required, and is capable of carrying out the duties specified. Some specifics in terms of words, phrases, etc. may also be specified in the script to prompt the interviewer as to what would be good things for the interviewee to mention in the interview in answering each question.
In the interview, candidates are scored per question as to how well they demonstrate the capability to carry out the duties being asked about, or how well they demonstrate previous experience at carrying out similar duties, etc. and marks are awarded per question, out of a total allowance. Not all questions may necessarily carry the same weight, and this can be achieved by varying the total number of marks which can be awarded for each question asked.
A total score is calculated for each candidate, and the candidate with the most points is the one who gets the job. If there are two jobs, then the top two get the jobs, etc.
This is thus deemed totally fair to all, as everyone competes on an equal basis, and the process is totally transparent if audited after the event. What could possibly go wrong?
Can it be fiddled?
Let's imagine that a hiring manager, working for a company that declares that it uses a Fair Selection process, wants to promote a certain person, because they have worked with them for years, know that they are right for the more senior role, and that person works well with the rest of the team. Can that manager "fiddle" the Fair Selection process to ensure that their favourite candidate comes out top?
Yes they can. For a start, if it is a new job, then they can write the Job Spec specifically to favour the skills and experience of the person they want to be successful. Sometimes you read a Job Spec where this has clearly been done without too much subtlety, and you know that the hiring manager already has someone lined up for the role. An example might be "Required: 15 years recent experience as <something very specific>". This tends to give away that they want to narrow the field of possible candidates to favour someone they have in mind who exactly fits that "requirement". In reality, who needs 15 years of experience at something to be good at it? Five years maximum, I would say, and even that may be pushing it. If you see a list of similar "ridiculous" stipulations on a Job Spec, then you know that it has been written to select someone specific, who the hiring manager wants to give the job to.
Then, if the hiring manager wants to, there is nothing stopping him/her from coaching his/her preferred candidate to ensure that they will give great answers to the questions that will be asked, and thus score high marks. An unscrupulous manager might even give their favourite candidate the actual questions and model answers. That's clearly cheating, but if they do it discretely and neither party tells anyone else, then who is to know?
Marking itself can be quite subjective on many questions, and a few marks more or fewer for one candidate or another across a series of questions, can easily look fair and reasonable, but could be a deliberate skewing of the overall marks by the interviewer.
Thus, the hiring manager's preferred candidate comes out with the top score, and is seen to have been awarded the role on a completely fair basis, as the best performing candidate on the day, through a process which cannot be challenged by those who were not successful.
Can it go wrong by accident?
Yes it can. As mentioned in the previous section, marking on many questions can be quite subjective. As an interviewer, you need to be very careful about recording how much help you gave the candidate to get to the desired answer. Did they come out with everything without any prompting, did they require minimal prompting, etc., right down to you virtually telling them the whole answer? The scoring needs to reflect the amount of prompting.
Then there is the "halo" or "horns" factor. A confident, well-dressed, smiling, articulate, friendly candidate, may come across so well that you don't pay enough attention to the actual content of what they are saying, but give them the benefit of the doubt on ambiguous statements, because they seem like the kind of person who knows what they are talking about.
Contrast this with a scruffy, ugly, shy, quiet person who cannot make eye-contact, or an over-confident, obnoxious person, and you may start, inadvertently, to assume that everything they say is wrong without really listening to the content of what they are saying, or assume that you are always prompting them and giving away the answers, when actually you are just trying to keep them talking because they are shy and keep stopping before they have said enough.
Can this really happen? Yes it can. I have interviewed dozens of people using this process, and I can assure you that it happens more easily than you think. The solution is to have two or three interviewees, who all take notes throughout, and who all monitor each other for succumbing to the "halo" or "horns" effect.
What else can go wrong?
This system does not test what the candidate knows overall, it only tests whether or not they know the answers to the questions asked. There is often no opportunity for the candidate to say anything else in the interview which counts towards the total marks. Good questions asked at the end by the candidate are often not marked, and thus are of no use to the candidate in increasing their chances of getting the job. Of course, there is nothing to stop the interviewer specifically allocating marks to every candidate for asking relevant questions at the end which demonstrate their ability to do the job, but usually they don't, because it is hard to mark.
The defence of this is that if the process if followed well, then the interview will determine each applicant's ability to do the the job, and that is the whole point of the process, but, in practise, it is impossible to cover every aspect of the job in a 45-minute or 1-hour interview, which they usually are, so there is an element of pot-luck for the candidates as to which particular aspects of the role are probed, and whether that plays to their particular strengths or weaknesses.
So why use this flawed process?
All processes are flawed in some respect. At least this one has a clear structure, and can be demonstrated to be fair, on the basis of every candidate being asked exactly the same questions, and all of the questions being relevant to the job role. However, it is a better tool for ensuring that the result of the interview process cannot be successfully challenged by an unsuccessful candidate, and the hiring company cannot be sued, than it is for ensuring that the very best candidate for the job was the one actually chosen. This is one major (but rarely openly admitted to) reason why many large companies, and particularly those with a significant public profile, choose to use such a system. They can prove that it is fair in law, even if a philosophical discussion might cast doubt on the fact that the best candidate is always chosen through this process.
Additionally, the more rigorously and carefully the process is adhered to, the better it performs, and if the interviewers are skilled and experienced in the process, and plan their way through it carefully, then it can, more reliably than not, select the best candidate, or, at the very least, the best candidate on the day.
So what are the alternatives?
The alternatives are many. "Fair Selection" is a policy of choice for a company. There is no legal requirement to select people for jobs on a "fair" basis. You may have to prove that you did not discriminate against someone because of their race, age, religion, etc. but you don't have to prove that the person you selected for the job was definitely the best person to do it. (The law adresses the people who did not get the job, not the person who did get it).
Anyone can offer anyone else a job for whatever reason they wish to. I know of someone (and I'm sure there are many such similar examples) who built a team under him of women selected on the basis of how much he fancied them, and how skimpy their clothes were. Going to the extreme of only recruiting women fitting this profile can start to open him up to a legal challenge of sexual discrimination against men, or discrimination against less-attractive, or more modestly dressed women. However, if someone only employs one or two people on that basis, and employs others who are different, then it is almost impossible to prove discrimination.
There are many other interview techniques, involving casual chats, tasks, group exercises, meals in expensive restaurants to test the ability to conduct oneself properly in such an environment, and aggressive interviews, designed to test a person's resolve under extreme pressure. Alternatively, jobs can just be given to the hiring manager's friends. I know someone who protected his own position by having a team immediately below him consisting entirely of his close friends, knowing that they would support him and not challenge him, so that he could focus on managing upwards, and getting his next promotion. The pros and cons of these techniques are subjects for other hubs, but a case for and against can be made for each and every one.
At the end of the day, an employer wants people working for their business who will help to make that business successful, or help to make them personally successful. How they choose them is up to them. What I find is a shame is that the process is often quite short - it might be one 45 minute interview, and that's it - success or failure for the candidate in less than one hour. Can anyone really assess someone's ability in such a short time, by whatever method they use? So can any process that lasts only 45 minutes truly be fair to every candidate?
There may be two or three stages, and that is potentially better, but they tend to be elimination stages, whereby a silly mistake, or mental block, in the first stage, eliminates someone who may be an excellent candidate, and may have come out the best if all candidates had gone through all stages and were assessed overall.
However, we know why such rigour doesn't happen. Managers do not have the time to go through such a robust and time-consuming process. This is no doubt why there are so many recruiters and head-hunters, who do have the time to do at least the first stage selection and short-listing process, and then put the best through to the company for the other stages. The companies do not then necessarily need to select the best overall candidate - just the one who does the best on the day (or days of interviews), as they know that all the candidates they are seeing are capable of doing the job to a high standard. That's what they paid the recruiter for.
Indeed, this is where "informal chat" techniques might be used to see which candidates fit well with the team, otherwise known as "does their face fit"? It's a legitimate selection criteria.
My conclusion is that there is no totally fair way to select the successful candidate for a particular job. Fairness is not actually the issue, and rarely, in reality, the aim. The issue is finding someone who fits in with the rest of the team and is easily capable of doing what is required of them, and thus who is a benefit to the company who is employing them, making their business more successful.
Whether that person is found by a system which aims to be fair to all, or makes no pretence of being fair to all, doesn't matter, as long as a person who fits is found and employed in the role.
This can be quite demoralising for strong candidates who are not good at selling themselves in all of the various types of interviews and selection processes, who are trying to get back into employment after being made redundant, or others who are desperate to improve their careers, or those trying to get their first ever job - indeed for anyone applying for a job; but that's the way of the world, and all you can do is accept it, don't get hung up about it, and work out how you can make sure that you are the one that is selected, fairly or unfairly!