Glossary of Technical Terms

Aptitude: Aptitudes are specific cognitive abilities which focus on one form of reasoning only, i.e. numerical reasoning, verbal reasoning, inductive reasoning etc. Academic research has found aptitudes to be strong predictors of performance at work. Similarly, when scores among aptitude tests are combined they measure general cognitive ability, the strongest predictor of performance at work.

General Cognitive Ability: General cognitive ability (also known as general mental ability or the g factor) is the very broad capability involving the ability to reason, plan, solve problems and comprehend complex ideas. Academic research has consistently found general cognitive ability to be the best predictor of performance at work. Measuring general cognitive ability involves combining many different aptitude tests, forming an overall measure of cognitive performance. Averaging the sten scores of Test Partnerships three aptitude test scores provides a measure of general cognitive ability.

Inductive Reasoning: In a testing context, inductive reasoning is the ability to think logically, rationally solve problems and identify complex patterns. Inductive reasoning, abstract reasoning and logical reasoning are frequently used synonymously with one another and refer to the same type of ability.

Normal Distribution: A normal distribution (also known as a bell curve) is the commonly occurring observation that the frequency of a measurable variable tends to cluster around the average, and become less common further away from the average. Norm groups are normally distributed data, and form the basis of norm referenced tests.

Norm Group: Norm groups (also known as comparison groups) are datasets containing previous test takers scores, which are used for benchmarking purposes. Norm groups refer to a specific population, such as graduates, professionals or managers. The average performance of each population is used as a benchmark in which candidates scores are compared. Norm groups typically contain hundreds or even thousands of previous test takers, in order to ensure the accuracy of the average score.

Numerical Reasoning: Numerical reasoning is the ability to work with, understand and interpret numerical information. Numerical reasoning questions incorporate problems applicable to commercially relevant situations, and involve calculations frequently used in the world of work.

Percentile: Percentile scores are raw scores which have been transformed into more useful data using norm groups. Percentile scores range from 1-99, with higher scores indicating a higher performance. A score at the 50th percentile indicates an average performance, a score at the 90th percentile indicates a high performance and a score at the 10th percentile indicates a low performance, with respect to the norm group.  

Raw Score: A raw score is the number of correct answers a candidate has scored on a test. Without a norm group for bench marking, a raw score is merely an arbitrary number. With a norm group, raw scores can be transformed into percentiles, sten scores and other useful statistics.

Sten Scores: This is an abbreviation of “standard ten” and shows the approximate position of a candidate, in regard to other candidates within that population or “norm group”. The Sten score effectively represents where the candidate would place on a scale of 1-10, compared to all other candidates had they been arranged by performance.

Verbal reasoning: Verbal reasoning is the ability to understand, interpret and draw logical conclusions from written information.Verbal reasoning tests are particularly useful for selecting candidates working in roles which involve the interpretation of written information.


Article is closed for comments.