If you've ever opened a psychometric report and found yourself staring at a number with no idea what to do with it, you're in good company. These scoring systems aren't complicated once someone explains them, but they're almost never explained properly on the report itself.
Here's a quick guide to the three you're most likely to encounter: Z scores, sten scores, and stanine scores. I'll hold off on the heavy mathematical side of things.
All three scoring systems share one underlying logic
Whatever system you're looking at, the number on the page isn't telling you what a candidate scored in isolation. It's telling you where they sit relative to everyone else who has taken the same test.
A raw score on its own is meaningless. If someone scores 10 out of 16 on a numerical test, you can't do anything useful with that number unless you know how other people performed on the same test. Did most people score 7? Then 10 is impressive. Did most people score 12? Then 10 is on the lower end.
All three systems solve this by translating raw scores into something relative, using normal distribution. You don't need to remember anything from statistics class, but just think of those bell shaped graphs. The main thing to know is that most people cluster around the middle on any given measure, with fewer people at either extreme. That's what these scales are built around, where the middle represents typical performance.
Z scores are the statistical engine, not the number on your report
A Z score expresses how far a candidate's raw score is from the average, measured in standard deviations. A Z score of 0 means they scored exactly at the mean. A Z score of +1 means they're one standard deviation above it. A Z score of -1 means they're one standard deviation below.
They're not usually what you see on a psychometric report, though. They run from roughly -3 to +3 (including decimals). Telling someone their conscientiousness score is -0.74 is technically accurate and practically useless.
What Z scores actually do is power everything behind the scenes. They're what we test providers use to convert raw scores into the friendlier formats below.
Sten scores appear on most personality questionnaires
Sten scores run from 1 to 10, with a mean of 5.5 and a standard deviation of 2. The scale was introduced by Canfield in 1951 and became the standard format in UK and Irish occupational testing.
Each sten band covers a range of raw scores, so two candidates can have different raw scores and still land on the same sten. That's intentional. Small raw score differences don't get over-interpreted.
One thing worth knowing: a sten 5 or 6 looks underwhelming but isn't. It's squarely average, and average on a personality trait is exactly where most of the population sits.
Stanine scores follow the same logic on a nine-point scale
Stanine scores run from 1 to 9, with a mean of 5 and a standard deviation of approximately 2. The name is a contraction of "standard nine". Like stens, each point represents a band of raw scores rather than a precise value.
When using Stanine scores, scores between 1-3 are typically considered to be "low" scores, scores between 4-6 are often considered the "average" range, and scores 7-9 are considered "high" scores.
Operationally they work the same as sten scores, but instead of going up to 10, they stop at 9. Simple.
Conclusion and next steps
All three systems are doing the same fundamental job: showing you where a candidate sits relative to a reference group. Z scores are the underlying mechanism most people never see. Sten scores and Stanine scores work the same way but on a slightly different scale.
Find out more about the science behind our Test Partnership assessments.
Ben Schwencke
Chief psychologist at Test Partnership. MSc in Organisational Psychology with over ten years experience in psychometric testing.
