The score provided in an individual’s report – their standardised score – is determined by comparing their raw score to that of a reference population. This reference population comprises a large group of individuals whose scores serve as a basis for comparison. Essentially, it acts as a benchmark that reflects the broader society. Therefore, the standardised score reflects the individual’s position within the general population.

The calculation of standardised scores involves several steps. For each of our assessments we have created a reference group comprising thousands of participants. For each of these groups, we compute the mean and standard deviation. Next, we use these values to construct a normal distribution curve. By comparing an individual’s raw score to this distribution, we determine how many standard deviations their score deviates from the mean. This comparison yields the individual’s position within the normal distribution, which is their standardised score. 

For example, let’s consider a reference group with a mean of 5 and a standard deviation of 2. If an individual obtains a raw score of 7, we calculate their deviation from the mean using the formula (raw score – mean) / sd. Substituting the values, we get (7 – 5) / 2 = 1. This indicates that the person is 1 standard deviation removed from the mean. In a normal distribution, this deviation corresponds to a standardised score of 84.

Psychometric tests, like the AEM-Cube, typically use standardised scores. This is because standardised scores provide a common metric for interpreting individual outcomes, enabling practitioners to assess an individual’s contribution and perspective relative to a baseline. This facilitates more accurate and reliable evaluations. 

Suppose, for instance, that a group of students completes a French test, and they all score around 95%. Does this mean that everybody in that group speaks nearly perfect French or was the test too easy? The answer lies in comparing the raw scores of this group to that of a reference population, and thereby gaining a much better understanding of the implications behind these outcomes.

The use of standardised scores has the effect that even individuals with relatively neutral responses might still end up with seemingly extreme standardised scores.

For instance, suppose the reference population has a mean exploration score of 65 and a standard deviation of 10 on a scale from 1 to 100. In this case, a person with a score of 60, while closer to the explorative end in absolute terms, is actually less explorative than the average person. Standardised scores reflect this nuance.

See the image below for an illustration of this concept.

In other words: Are people really more explorative, or do they answer as if they are because these positions can be considered more attractive? Whilst social desirability does introduce a bias to a certain degree – as is common for psychometric tests, after all nobody is entirely objective – this bias is partly corrected by using standardised scores. This is because, in the event of a bias, the average of the reference population will also be influenced. Therefore, by comparing the individual to the reference population, the impact of potential social desirability bias on the standardised score will be mitigated. Furthermore, statistical testing has validated the reliability of the assessments in measuring the intended constructs. For a detailed description on the statistical validation of our assessments, please refer to the scientific research published on our website: https://human-insight.com/scientific-research/

The normal distribution curve (see image below), also known as the bell curve because of its shape, serves as a visual representation of a prevalent phenomenon observed
in various aspects of the world. It’s called “normal” because lots of things in nature, such as human heights, follow this distinct pattern. Namely, this curve illustrates how data tends to cluster around a central value, or mean. 

For instance, the heights of a large group of people. If you plot their heights, you would notice that the data clusters around a central value – the mean. And fewer and fewer people having heights significantly far removed from that mean. In other words, on any given day, you see a lot of average sized people, and only a small number of really tall or really small people.

Please find the size of our reference populations below. 1

For the AEM-Cube:
Self-images: 28.145
Feedback-images: 51.369

For the Qi:
Current: 1292
Desired: 941