Developing an aptitude test for recruitment - a Basic Primer

It’s fun to create an aptitude test for new hires. All that you have to do is to add some subjective, some ‘IQ’ types, some technical question and get those hapless candidates to take the test. Your organization will be thankful to you for the test that you have created and those hapless candidates will rarely complain if you, in an unlikely case, make any mistake!

Right? Wrong! Are you sure that your new test

1.       meets the requirement given in role or job description?

2.       weighs the different questions in scorecard appropriately based on their difficulty and criticality levels?

3.       conforms to equal opportunity employment requirement?

4.       does not give undue importance to a skill or attitude that are not that critical in the job, and vice versa?

5.       is not too easy or too difficult to make results useless in screening candidates?

6.       has adequate, not more nor less, number and variety of questions to assess the desired skills and behavior.

Creating an aptitude test for new hires is not really difficult but it’s easy to make mistakes and a wrong hiring decision can be very expensive. The above check list was useful when I created or administered tests to thousands of employees for my employers.

Supervised versus Unsupervised Online test

Unsupervised or unproctored online tests are cost effective and saves time for assessed as well as assessor. If you are looking to hone your skills, these unsupervised tests provide you all the flexibility that you need at a very little cost as I believe that you will be honest to yourself. For behavioral assessments, where there is no clear right or wrong answer, unsupervised tests provide very reliable results.

However, as you would have guessed easily, unsupervised tests expecting specific answers are prone to identity fraud and can be compromised. Many organizations hesitate to use unsupervised test for this reason. But organizations can gain from the flexibility provided by unsupervised tests without losing the integrity of assessment by using one or more of the following methods.

1.       Post-test Validation:

Follow up with questions based on the test during the interview for the shortlisted candidates. This can be very effective if planned right.

2.       Unsupervised screening followed by supervised test for shortlisted candidates

By comparing the scores achieved in different modules of the two tests, one can filter out majority, nut not all, of identity frauds. Assessor also has a choice of using only the latest test score or a suitably weighted average of two tests.

3.        Access control

Allow only previously validated users (student or job candidates) to take the test at a specified time, possibly in conjunction with extra identity validation features like keylogging or webcam snapshot recording.