The third part in our Re-thinking Recruitment
series, which explores various aspects of the selection process, by Dexter Davies Smith.
Welcome to Part 3 of our journey into recruitment, taking an academic approach to the selection processes. In this entry we will examine the traditional approach to recruitment – using application forms and CVs to find and filter applicants before interviewing the chosen candidates to find our eventual employee.
As a quick re-cap, in Part 1
we evaluated the financial impact of getting recruitment right or wrong. In Part 2
we looked at some of the psychological principals behind reviewing selection methods, namely the concepts face validity, predictive validity and adverse impact - if you haven’t read that yet, I would recommend doing so to fully understand the evaluations we will make in this entry
. We concluded the last entry by looking at why human judgment is so often off the mark when predicting how someone will perform as an employee.
Before we dive in to analysing these selection methods, I want to clarify that the objective of this series is not to provide a one-size-fits-all prescription for how to recruit. These reviews are written with the aim of stimulating thought and discussion around why we use the selection methods we do and will hopefully provide some insight that may otherwise be lost in academic texts. With that covered, let’s begin our evaluations with two of the most common methods, often referred to as ‘traditional recruitment’.
Application Forms & CVs
Submitting a CV (or increasingly there seems to be a movement towards online application forms) is likely to be the starting point for most recruitment processes. This is a process that I have no doubt you will have experienced yourself. A typical CV or application form will contain information on the candidate’s educational background, occupational experience and sometimes additional information to give an insight to their character.
: The high level of use and acceptance of application forms and CVs support that they are perceived to have high face validity. We intuitively believe that asking candidates to summarise and present information on themselves will allow us to effectively differentiate between weak and strong candidates.
Within this, each company may be looking at different criteria to judge these applications by (a CV which illustrates a strong candidate for a Waiter may not be so strong for being a Company Secretary) so a second facet of this face validity is that we intuitively believe that the criteria we use to judge these applications will be good filters for predicting which candidates will perform well.
: There is little research into the predictive validity of application forms. There are a two keys reasons for this:
Firstly, because the criteria used to judge these applications varies between companies and roles, we would need studies validating each individual application form and selection criteria which, from an academic perspective, is un-realistic and wouldn’t provide substantial insight. However, this is a process that would be really simple for individual companies to do, and a process that long-term would lead to better recruitment (if anyone wants information on how to do that, leave a comment and I’ll write up a short guide).
The second reason, which is an issue for application forms as a whole, is that the applicants can lie! Studies have shown that up to 57% candidates will falsify information around their previous employment and that people are most likely to lie about matters which are opinion, such as why they are applying for this role.
The level and nature of adverse impact from using application forms/CVs will depend on the selection criteria you use to filter between applicants.
The obvious strength of application forms is that they are a way for candidates to put their name in the hat for the role and provide a snap-shot of voluntary information. The costs around application forms/CVs is one that is easily miscalculated.
Online applications & CV sent in by e-mail give the impression that it costs companies very little to receive these applications. However, the sifting process is extremely time consuming. Someone has to take the time to read all these applications and make a decision on which candidates are in or out – this may be a member of the HR team, the line manager responsible for the position being fulfilled, or an external consultant.
My concluding advice for using this method, is that without keeping a record of the criteria used to filter and then recording how that employee goes on to perform, we have no indication of how good these criteria are at identifying the best talent from our candidate pool. It is a simple practice to implement and can quickly make a big difference.
A case example: a business-to-business sales team recruited purely graduate candidates for their entry level sales roles. One time, due to a lack of suitable candidates they interviewed and hired a non-graduate that went on to out-perform the rest of the team. When hiring the next open position, they returned to their graduate only policy, on the basis that it had always been the criteria, regardless of evidence that it wasn’t a good predictor of performance.
The Selection Interview
In the last entry, we looked at several reasons why human judgment is usually a poor predictor of how someone will perform, which would suggest it is unfortunate that interviews are predominately the main method used in recruitment. However, the idea of not having an interview seems unfathomable, but is there any evidence it is a useful process?
A distinction is made between an unstructured interview in which the candidate and interviewer have a conversation that has no ridged set course and structured interviews, in which there are set questions, a set scoring system for rating answers and little room for divergence.
Unstructured interviews score pretty poorly, a meta-analysis on the research suggest that around 10% of a candidate’s future job performance can be predicted by unstructured interview.
In stark contrast, structured interviews perform very well, a meta-analysis on the research suggests that around 30 –35% of future job performance can be predicted by a structure interview.
Due to the unstandardised nature between interviews it is impossible to give them an overall adverse impact score. The more structured the interview process, the less room for human bias (hence the improved predictive validity) so the less likely personal biases will cause adverse impact. It is worth reviewing whether the questions and criteria used in your recruitment processes are likely to be adversely bias against any social groups.
I would be amazed if there is ever a recruitment process that doesn’t include an interview as out of all the methods we will review, it is the only one that has a social interaction between the candidate and the employer. However, what is clear, is that this social interaction isn’t by nature a strong method for identifying top performers. The more this method can be standardised, the stronger it is likely to be for identifying the best talent from the candidate pool. As a final thought for interviews, there is a significant amount of evidence supporting the theory that they are geared towards rewarding extroverts, who are more comfortable with the intense social interaction and draw their energy from it. The biggest issue with interviews is we make the false assumption between confidence (or lack of it) in the interview and competence later on it the role.
In conclusion, the traditional selection methods are certainly not without merit and the high-scoring predictive validity for structured interviews are really promising. However, both methods looked at are labour intensive, taking up a significant amount of employee time to achieve pretty unreliable differentiation between candidates’ future performance as employees. It is easy to see why recruitment can be a frustrating process that often feels more like astrology than science.
In the next entry we will look at two psychometric methods of recruitment, considering their strengths and weaknesses in adding to the recruitment process.