Home > All articles
Article · 5 minute read

Reduce the Risk of a Bad Hire: Shining a Spotlight on Candidate Faking

Richard Williams

When it comes to implementing a fair, scientific and effective approach to hiring, CVs have always been problematic:  

  • Lacking a future focus – CVs showcase what someone has done in the past, not what they are capable of doing in the future. 40% of employers on LinkedIn are now using a skills-first approach, ditching many of the academic or work experience requirements a CV would typically focus on.

  • Inviting bias – CVs facilitate both conscious and unconscious bias in decision making by recruiters based on information relating to gender, age, ethnicity and social mobility. In a study run out of the US, two versions of the same CV were sent out to over 80 law firms.  Version A included hobbies that were considered ‘highbrow’, like golf, whilst version B included hobbies that were considered ‘lowbrow’, like soccer. The ‘highbrow’ CVs got significantly more call backs.

  • Vulnerability to Generative AI – The Times recently reported that hiring managers were so concerned about the use of generative AI tools such as ChatGPT to write CVs and cover letters, they were calling on the government to issue advice to employers.

This doesn’t mean CVs and online applications forms should be ditched completely, but it is more important than ever for them to be used in parallel with a robust, objective and scientific solution.

At Saville Assessment, we partner with clients to help reduce the risks associated with hiring, and help ensure the right people are identified for the right roles. The science behind our Wave® personality questionnaires helps reduce the odds of a bad hire from 1 in 5 to 1 in 50.

In this short article, we highlight some of the unique Wave features which help you spot potential red flags in a candidate profile and tips for preventing candidate faking.

Are they too confident?

Our flagship assessment Wave, the most scientifically-proven predictor of work performance and potential, includes some unique scoring aspects and powerful safeguards to prevent and control distortion. The first place to look for insight into how an individual has responded is the Response Summary.   

In this section, very high scores on acquiescence can be accompanied by good or reasonable consistency (see below). High acquiescence with consistency can be linked to strong performance in the workplace and should also be associated with a track record of achievement and progress.

Wave response summary excerpt

In selection scenarios, high acquiescence and the associated splits on the profile should be considered as areas for further exploration. Cross-referencing these areas is essential. Conversely, very low acquiescence can be a sign of self-criticism or a lack of motivation to complete.

Low consistency is a warning, but it is not an indication of faking as such. Users need to explore this further before coming to conclusions.

Where acquiescence is high or low, we tend to see lower normative-ipsative agreement in the response summary and we can then hone in on potential areas of distortion or mistruth with the normative-ipsative splits (more on these below) in the individual’s profile itself.

Are they trying too hard to tell you what they think you want to hear?

For each area of the Wave report, there is a rating score (normative) on a nine-point scale and a score based on the ranked data (ipsative). Normative-Ipsative (N-I) splits occur when there is a difference of three or more Stens between the two scores on the same dimension, and these are flagged for further investigation. 

If an individual is ‘faking good’, or overplaying their capabilities, this is likely to be illustrated by a number of splits where the normative (N) scores are higher than the ipsative (I). These dimensions need further probing at interview to see whether the individual can provide evidence and examples to back up their profile. 

On the flip side, where there is low acquiescence and splits where the I is higher than the N, the individual may be being self-depreciating or in some exceptional circumstances even ‘faking bad’. A large number of splits by five Stens or more should be considered a red flag and should raise questions for those interpreting the report, and require further exploration.

Are they being genuine in their responses?

The Wave profile also looks at an individual’s motives and talent; in simple terms, how much they enjoy doing something compared with how good they are at it. The average number of Motive-Talent splits (where there’s a mismatch between these two aspects) you’d expect to see on a Wave profile is four. Where there are more, this this could be an indication of someone whose career has plateaued or is in the wrong role, but it could also be a clue towards non-genuine responses, especially when combined with a large number of normative-ipsative splits.    

Can they walk the walk?

We also encourage cross-referencing the data generated by Wave.

Some examples of what to look out for are given below (in line with the different dimensions of Wave):

Different dimensions of Wave

5 tips for preventing faking

  • In all briefings, make clear that there are no ‘right’ and ‘wrong’ answers, and encourage them to answer all questions as honestly and straightforwardly as possible.
  • Let candidates know prior to the completion of an assessment that you will be cross-referencing the results with other information and assessment methods.
  • Explore other data collected in conjunction with the Wave results and fully explore any areas of concern.
  • Explore other data collected in conjunction with the Wave results and fully explore any areas of concern.
  • Inform candidates that the results of the Wave assessment will power the questions used at interview where they may be asked to produce behavioral examples to support their responses.
  • Advise candidates that there are intelligent mechanisms within the questionnaire which observe the manner in which they respond.