Construct valdity relates to whether a particular psychometric assessment...
Lead consultant at Test Partnership, Ben Schwencke, explains what is a sten score.
Sten scores (short for “Standard Ten”) are standardised 1-10 scores commonly used in psychometric testing. Sten scores allow psychometricians to convert rather abstract standardised scores into more interpretable numbers for a lay audience. As with other standardised scores, such as Z-scores and T-scores, they are based on the average and distribution of scores for a specific population.
Sten scores are calculated by multiplying the Z-score by 2, and then adding 5.5, giving the following formula:
Here is a worked example:
When using Sten scores, scores between 1-3 are typically considered to be “low” scores. Scores between 4-7 are often considered the “average” range, and scores 8-10 a considered “high” scores. Technically however, Sten scores theoretically can range from negative infinity to positive infinity, but the vast majority of data points in a normally distributed dataset will fall between 1-10. Traditionally, Sten scores are reported as whole numbers rather than decimals, but this is only by convention, as initial Sten score calculations usually require truncation or rounding.
By definition, the average Sten score is 5.5, which is analogous to a percentile rank of 50, or a Z-score of 0. Sten scores are similar to Stanine scores, which are short for “Standard Nine”. As with Stanine scores, Sten scores represent a simpler alternative to more granular standardised scores, which can be useful for setting broad pass marks or cut-scores.