Subscribe To Our Blog

Psychometric Test   | 2 Min Read

Changing The Old Guard: Why You Should Rethink Your Psychometric Assessments

Written By Tonmoy Shingal

The world of Human Resources today is characterized by the constant need to venture into high stake investments or decisions. These are choices that considerably impact the bottom line of an organization. In consequence, with changing trends in technology, several tools – assessments included – emerged to assist HR in their evolutionary transformation into Business Partners.

However, with this change emerged a need for HR to understand the relevance of their tools, stay updated, and on par with the dynamic nature of a workplace. Needless to say, a tool is often subject to growing research and development as well. These factors in their entirety promised HR the data-driven accuracy and precision their decisions so required.

“The old order changeth, yielding place to new.” - Lord Alfred Tennyson, Poet Laureate of Great Britain Wise words by a wise man. This, for

Wise words by a wise man. This, for it is important to ensure the freshness and relevance of assessment tools to keep them culturally stable, thereby improving their psychometric properties.

Take for example Apple, a phenomenal company motivated to develop and release state-of-the -art products on an annual basis. Even Google with its Android operating system strives to fix bugs, upgrade interface and more. Such is the progressive nature of our world.

In similarity, assessment tools also require the same touch of nurture, to eliminate gaps and improve efficiency.

"The old order changeth, yielding place to new" - Lord Alfred Tennyson, Poet Laureate of Great Britain.


What Kind of Gaps?

Most psychometric tests are not mastery tests with criterion references that determine performance. In fact, these tests require an entirely different method to classify scores as low or high. To assess overall performance, test makers often employ something known as a standardization sample.

This allows for the creation of a normal distribution, which can later be used for comparison of any specific future test score.

Raw scores from the sample group is converted into percentiles to construct a normal distribution – used later to rank future test takers.

But enough of the science. After all, it is not our aim to bore. The bigger question here is, “Why is Norming so important?”

Well, norms are not standards of performance for one. They, in actuality, serve as a frame of reference for test score interpretation. Of course, the size may vary from a few hundred to a hundred thousand people.

More People = Better Approximation

To more important matters, and to answer the question asked, A sample must be representative.”

Simply put, test children if you’re developing a test for children; test adults if you’re developing a test for adults. The closer the match between your sample and your intended population of test takers, the more accurate the distribution as a ranking guide.

If we look at Stratified Sampling, which is by far the most accurate way of developing a norm group – the test maker considers every demographic variable imaginable to describe the population of interest. These include age, gender, socioeconomic status, geographic region and more.

In other words, using tests made on foreign soil is pointless – pointless being a generous word. Yes, it does not matter if the test maker is of world renown if the test is not built for the target audience.

This is applicable from an employment standpoint as well. Now, in terms of business there are other factors to consider as well, of course. Some of them include:

A Personality Test Designed in the Employment Context

We must certainly avert our eyes from intrusive, clinical or otherwise irrelevant personality tests. An ideal test here ought to be designed to include items and statements more reflective of work experiences or situations. There must be a conscious effort to eliminate the use of items that may come across as generic or irrelevant to the work context.


Topics: Psychometric Test

Originally published December 27 2018,updated June 29 2019

Would you like to comment?