Posted on July 13, 2020 by P.K. Maric
Some users ask why so many candidates fail job screening tests. It’s normal for most candidates to fail; in fact —that’s the point. Unlike school tests, which are designed so that 90% of students will pass, pre-employment tests usually eliminate 90% of candidates. This is especially true when it comes to programming, where many people have noted that most applicants can’t write code at all. This seems bad, but it’s actually good news for you because you can quickly narrow down your selection of candidates. That way, you don’t have to waste time interviewing unqualified candidates.
Skill tests are often used for pre-interview screening, which is what we recommend because it’s a quick and effective way to filter out candidates at the start of your hiring process and narrow down your selection to the best candidates. This means that most candidates will fail. But why do they perform so poorly?
Candidates apply to jobs they are not qualified for
When someone is looking for a job, they usually apply to whatever job ad they think is interesting, regardless of how (un)qualified they are. And why shouldn’t they? They have absolutely nothing to lose— they already don’t have the job. So they’ll apply to your job opening (and many others) hoping they’ll slip through the cracks in your hiring process. Or maybe they honestly believe that they could learn on the job and be good employees, perhaps they’re even right.
But you need to know for sure, and for that, you need a robust hiring process and a good skill test. Consider this: There are about 250 candidates for each corporate job opening. Only one can get the job, which means 249 candidates need to fail at some point during the hiring process, and most of them will fail the skill test.
But there are things to consider if you want the testing process to be fair and select the best candidates:
The Score Gap
The gap between the candidate’s score and 100% may reveal some crucial information about their abilities. A candidate that scored 90% on the test definitely looks good, but it depends on which 10% of the test they got wrong. While the score gives you an excellent overview of how a candidate performed in the test, it doesn’t tell you everything you need to know. You should always look at their answers to gain more insight. If they just made a few simple mistakes, but otherwise performed well, they might still be worth considering, despite what the score says. However, this depends on how you construct the test and how important each question is to the job fit.
If you’re using a pre-interview test, one of the most common mistakes is making it too hard. Yes, most candidates should fail the test, but not because it’s difficult, but simply because they’re not suited for the job.
A pre-interview test should be relatively short and easy. This test’s goal isn’t an in-depth evaluation of candidate abilities, but to test for the necessary skills needed for the job. For example, if you are hiring a programmer, 2 or 3 questions that test basic programming skills are enough to filter out most candidates.
A harder, more in-depth evaluation of skills can come later, and you can use this to compare candidates to find the best fit for your role.
Is a 70% score on a test good or bad? It depends on how you construct the test, the job criteria, and what the candidate got correct and what they got wrong. It’s important to set criteria beforehand and stick to it in order to compare candidates objectively and fairly.
Biases and bad questions
Tests should be unbiased and designed in such a way that only relevant job skills are tested. Brain teasers and puzzles are shown to be useless and do not belong in the hiring process. You can save other non-skill related questions for the interview.
Can you pass your own test?
It’s unfair to expect a new hire to be able to do something that you or your existing employees in the same role cannot do. You should calibrate your test based on your current employees’ performance if you expect new hires to perform similarly. A screening test should not be just an out-of-sight, out-of-mind tool that’s used blindly, without understanding what you’re asking candidates to do or why.
Provide feedback and follow up
If a candidate made a few mistakes, but otherwise performed well on the test, follow up with them in the interview and ask what they think they could have done better. A skill test, or even multiple skill tests, don’t tell you everything you need to know about a candidate. They are a very useful tool, but you should combine them with other effective hiring methods to accurately predict job performance.
If you’re using a good skill test, most candidates will be filtered out so that you can focus on the best. However, the criteria for passing or failing the test are up to you. You decide what kind of test you’ll use. Our tests allow you to set a passing score, which gives you an estimate of how many candidates will pass the test.
You will also be able to see a breakdown of performance per skill, and even each question. The candidates’ full submissions are available with timestamps. If you need help with setting up a test or interpreting candidate results, you can always ask us.