“”
News

Bidding Adieu to Multiple Choice Questions in 2024

published by
Beth Thouin
February 21, 2024
Find out more
Bidding Adieu to Multiple Choice Questions in 2024

It was in the second half of 2023 that the HiringBranch team started reconsidering multiple-choice questions. We knew then this question format failed to assess a candidate’s deep thinking skills, could create unreliable responses from candidates guessing, and could more easily lead to cheating since there is one known answer to the question.

We became convinced that open-ended questions were the only way to properly assess candidates to the point that we wrote a whole blog about it and put together this nifty battle card:

See the full battlecard here

Fast forward just a few months and we are thrilled to announce the retirement of our beloved Multiple Choice Questions (MCQ). HiringBranch has migrated to 100% open-ended speech and chat questions in our soft skills and communication assessment. 

As we usher in an era of AI-powered open-ended questioning, we ask you to join us in celebrating the contributions of MCQ!

The Future is Not A, B, or C

Multiple choice questions have been a pillar within pre-hire assessments for many years. This format has played a vital role in recruitment processes around the world, helping evaluate the psychometric behaviours and abilities of countless candidates. We owe much of the success of HiringBranch’s humble beginnings to MCQ.

AI has allowed us to approach the candidate assessment from a new, more dynamic perspective. With AI, we can now delve deeper into a candidate's thought process, creativity, and problem-solving skills, as well as the spontaneity of answers. AI gives us the gift of 1000X more data points to analyze than MCQ. Consider the following illustration demonstrating how our AI would read pieces of this sample response from a personal banking customer service scenario:

HiringBranch AI reading open-ended questions

Why Open-Ended Questions Are Everything

The additional data that hiring teams get from open-ended questions isn’t just providing more context about a candidate, it’s adding more validity to the outcomes of the assessment. Open-ended questions can demonstrate high relatedness and correlation, which leads to higher reliability scores overall. 4X more reliable results that is. 

HiringBranch has also observed differences in the average difficulty index, item-factor correlations, and consistency. Inconsistencies and randomness become prevalent in multiple-choice questions because candidates will simply guess the answer if they aren’t sure. Ultimately, this means that unfit candidates can pass through more easily than with an open-ended question assessment. 

Beyond that, candidates want to engage with assessments that are relevant to the position they’re applying for, and open-ended questions can provide scenarios that mimic the job. They also measure comprehension and deep-thinking skills, eliminate cheating, and so much more. Read all the additional benefits that open-ended questions can deliver next to MCQ by downloading our complete battle card of pros and cons.

Raise Your Glass

So, as we say goodbye to multiple-choice questions, we also celebrate. We ask you to raise your glass in honour of MCQ’s official retirement at HiringBranch. Thank you MCQ for the service and dedication MCQ has shown us over the years. As we turn our attention to open-ended questions, your achievements and contributions will not soon be forgotten. 

If you would like to share some memories and well wishes for MCQ, we encourage you to do so on our LinkedIn post. To test drive the new open-ended-questions-only assessment experience, contact HiringBranch today.

Image Credits

Feature Image & Image 1: Property of HiringBranch. Not to be reproduced without permission.

Does SHL guarantee a performance improvement?

We do. Gone are the days of guessing skills in an interview. Rely on technology do it for you with 99% accuracy and boost your hire quality, guranteed.

Book a demo
Some of our happy customers
New Report

Skills-Based Performance Report: AI Edition

Download report