Glossary of Psychometric Terms
Ability describes the degree to which someone can carry out certain types of mental reasoning operations.
Ability Test refers to a test designed to measure either aptitude or achievement. While aptitudes are defined as inherent abilities for learning and raw talents, they are sometimes measured through a person's achievements because the tests are designed incorrectly. The many types of ability tests include verbal, numeric, literacy and abstract reasoning tests.
The acceptability of a test refers to how acceptable it is the test taker. This depends on things like what the test appears to be measuring and the test taker’s faith in it. If a test is not acceptable then test takers may not cooperate in taking it.
An adaptive test is taken online and questions are selected from a database depending on the test takers previous answers. This is done so that the questions selected provide the most information about that particular person's level of ability.
An administrator is the person who administers a psychometric test.
A series of aptitude tests which are used to provide an indication of general and specific aptitudes. For example, verbal reasoning, numeric reasoning and abstract reasoning.
Standardized tests measuring specific intellectual capabilities or other characteristics.
Aptitude is a potential to succeed at something and aptitude tests are designed to measure those mental abilities which affect the likelihood of someone acquiring some particular skill.
There is no widely accepted definition of the difference between ability and aptitude. Most psychologists would agree that to some extent the two terms refer to the same thing: aptitude referring to specific ability, and ability referring to general aptitude.
An attainment test measures knowledge and proficiency rather than the ability to learn. These tests specifically assess what people have learnt and the skills they have acquired, for example, word processing tests.
Authentication ensures that the individual is who he or she claims to be. One function of test administration is to authenticate the identity of test takers and ensure that the person who presents themselves for testing is actually who they say they are.
Whenever test takers responses vary in some predictable way due to factors that the test was not intended to measure this is known as bias. This can be due to cultural and educational background as well as sex and age.
The conscious process of knowing or being aware of thoughts or perceptions, including understanding and reasoning.
Areas of personal capability that enable people to perform successfully in their jobs by achieving outcomes or completing tasks effectively. A competency can be knowledge, skills, attitudes, values, or personal characteristics. The competencies required for successful performance may or may not be expressed formally by an employer.
Composite Test Score
A score produced by adding together scores of two or more tests which may be differentially weighted before they are added. Composite scores derived from batteries of tests are often produced to provide a general measure of suitability in a selection situation.
Identifies units of competency within a competency standard that an industry has agreed are essential to be achieved if a person is to be accepted as competent at a particular level. Core competencies are normally those central to work in a particular industry.
A test taker’s score is used to predict how they will perform on types of task not directly sampled by the test but which have been shown to be correlated with performance in that test.
A percentile-based scoring system where the raw scores are divided into ten categories each containing 10% of the distribution.
The distribution of test scores will usually follow the classic bell shaped curve and produce a normal distribution.
A domain refers to the class of ability or achievement. For example, verbal reasoning, numerical reasoning, spatial ability, etc.
Performance on a particular test is related to the level of performance required in a job by using a common standard. Judgments are then made about what level of performance on the test would be required for adequate performance of the job.
An attribute or need of a person which is considered to cause them to act in a certain way or motivate them to action. A drive, such as hunger or thirst, motivates the person to act in ways that will reduce the tension. So, for example, when you become hungry (tension caused by need for food) you are motivated to eat (method of reducing the tension).
A test is said to be fair if it is not biased with respect to the groups with which it is used and if it can be shown to be valid. For example, the test must be appropriate to the group being tested and must be an unbiased test.
Five Point Grading
A percentile-based scoring system where the top 10% of scores are classed as grade A; the next 20% as grade B; the next 40% as grade C; the next 20% as grade D and the lowest 10% as grade E.
A tabulation of scores from high to low, or low to high, showing the number of individuals who obtain each score or whose scores fall in each score interval. Frequency distributions are used to determine tables of percentile ranks.
General Ability Tests
These tests vary from those designed to give an overall measure of general intellectual ability through those designed to assess broad areas of ability (for example, verbal reasoning or numerical reasoning) to those focusing on specific mental operations (for example, three-dimensional spatial rotations). The latter tend to be used for aptitude assessment. General ability tests, in order to properly cover the full range of mental operations, tend to include items dealing with each of the main areas of ability.
General norms are intended to be representative of a large and diverse population. For example, USA adult males or UK graduates.
The gaining of agreement from a test taker to the terms and conditions under which testing is to take place. This agreement must be obtained having first clearly informed the test taker of their rights and responsibilities, the reasons for testing, the type of tests to be used and what will be done with the results of the tests.
An ipsative test compares a person's score on one scale with their own score on other scale(s).
The systematic, formal study of the duties and responsibilities that comprise job content. The process seeks to obtain important and relevant information about the nature and level of the work performed and the specifications required for an incumbent to perform the job at a competent level.
A summary of the most important features of a job, including the general nature of the work performed (duties and responsibilities) and level (ie, skill, effort, responsibility and working conditions) of the work performed. It typically includes job specifications that include employee characteristics required for competent performance of the job. A job description should describe and focus on the job itself and not on any specific individual who might fill the job.
Job Simulation exercises are often used in the procedures that come under the general heading of the Assessment Center Method. Job simulations may take the form of in-tray exercises, group problem-solving exercises and so on. They start from the assumption that the candidate does not yet possess the requisite knowledge or skill, but that the underlying ability will manifest itself when he or she works through an exercise that simulates the broad demands of the job in question.
Measures of maximum performance measure how well people can do things, how much they know and how great their potential is. Measures of maximum performance include aptitude and ability tests. These measures are usually distinguished from measures of typical performance which assess personality and levels of motivation.
This is the push of the mental forces to accomplish an action. Unsatisfied needs motivate. On the biological level basic human needs of food, shelter and survival are powerful motivators. On the psychological level people need to be understood, affirmed, validated and appreciated. On the business level motivation occurs when people perceive a clear business reason for pursuing a transfer of knowledge or practices.
Multiple Choice Format
Test takers have to select one of a number of possible answers.
The sample of people from whom norms are derived.
A term synonymous with the standard normal distribution. The normal distribution (a bell-shaped curve) represents a theoretical frequency distribution of measurements. In a normal distribution, scores are concentrated near the mean and decrease in frequency as the distance from the mean increases. The mean, mode and median are all equal to each other; the proportion of the values falling between any interval along the scale is known from the mathematical properties of the distribution. There will always be, for example, 68% of the values between -1 and +1 standard deviations
Norm Referenced Measure
This defines where a test taker's raw score lies in relation to the scores obtained by the norm group.
Information usually in the form of a table, which enables raw scores to be converted into percentile scores or standard scores.
The value on the raw score scale below which a given percentage of the sample's scores lie. For example, if the 75th percentile rank is 40, then 75% of the sample will have scored less than 40.
Personality Questionnaires that assess preferred or typical ways of acting or thinking. Personality inventories attempt to measure how much or how little a person possesses of a specified trait or set of traits.
This contains all the people who conform to some specification. For example, US graduates aged 21-30, etc. Psychometrics involves making inferences about people who come from some population on the basis of information known about the behavior of a representative sample from that population.
A capacity to perform or acquire the skills to perform some class of actions.
The focus of a power test is on how many items a person is able to answer correctly. The time limit is designed to allow most people to complete all of the test items. If a person's score is mainly affected by their ability to answer the questions correctly, rather than their speed then the test can be described as a power test.
A percentile-based scoring system where the raw scores are divided up into four categories each containing 25% of the distribution.
This is the total number of correct answers a person obtains. A raw score is the total score a person gets on a test.
Relationship Test Items
where a test taker has to identify the relationship between two or more things and then use that relationship to select from a set of alternatives.
The extent to which one can rely on the score being an accurate measure of a person's ability or aptitude, rather than a measure of incidental random factors.
The numerical scores attributed to a test taker's answers to individual test items are added up to provide a single measure called a raw scale score. Scores obtained on a test are generally referred to as scale scores.
Self Report Tests
These tests ask the respondent to answer a structured set of questions about themselves. Most personality questionnaires are self-report.
An ability or aptitude test which contains relatively easy questions but which have a strict time limit. The measure of performance stresses the number of items attempted within the fixed time.
The 'standard' scale developed for measuring psychological characteristics is called the z-score or sometimes simply the standard score. A z-score is a measure equal to one standard deviation of a distribution. The most commonly used standard score scales are z-scores (one SD), T-scores (one tenth of an SD), Stens (half and SD).
The procedure of establishing the initial set of norms for a test, defining the conditions under which it should be used, and of assessing its reliability and validity.
A source of measurement error which is predictable and can lead to possible unfair bias.
A level of awareness and knowledge of tests or testing without which a person's scores may be negatively biased. Test sophistication may arise from prior exposure, the process of testing, or through the use of practice tests and information describing testing procedures.
Personality traits are those relatively stable and enduring characteristics of people that make them predictable.
Measures of typical performance, for example, personality questionnaires are designed to assess disposition, values, interests and to measure motivation. Measures of typical performance are usually distinguished from measures of maximum performance which are designed to assess how well people can do things and measure ability and aptitudes.
The process of building up evidence about what can and cannot be inferred from test scores.
Information on the validity of a test tells the user what inferences can be drawn about the person who has produced the score on a test and what is being measured by a test.
A work sample test is one in which the task has been taken from a job. The task is done under standardized assessment conditions.
A standard score scale with a mean of zero and a standard deviation of one.