Intelligence test

from Wikipedia, the free encyclopedia
Which figure in the bottom row makes sense to continue the top row? A typical example of a matrix test, correct solution: d

An intelligence test is a psychological diagnostic tool used to measure a person's intelligence . Since intelligence is defined differently, there are very different types of intelligence tests. It is assumed that differences in performance in intelligence tests also reflect differences in cognitive performance in everyday life. Psychological tests are used as an aid to find the optimal decision in the context of various diagnostic questions. The recording of intelligence can be helpful, for example, for predicting professional success or professional suitability (selection of personnel or career advice) or for recommendations on school education or a choice of studies. Various clinical questions (presence of diseases such as dementia or intellectual disabilities or the use of occupational rehabilitation measures) can also make it necessary to record intelligence.

The best known is the result of some of these tests, the intelligence quotient (IQ). Colloquially, intelligence tests are therefore also called IQ tests . Because of the risk of the IQ becoming absolute as a label for a person (such as height or weight - i.e. regardless of a specific question), this size is no longer used in technical terminology and other standard scales are also used to describe intelligence performance.

Intelligence tests are sometimes controversial. This is due, among other things, to the multitude of factors (environment, genetic makeup) that influence intelligence, a possible link with eugenics and doubts about objectivity.


Intelligence as an object of measurement in these tests is a construct that cannot be measured directly . In contrast to body size, for example, the determination of intelligence using intelligence tests is always subject to a measurement error. There is no intelligence test that covers all areas of intelligence. According to the Federal Ministry of Education and Research , the various tests “depending on the underlying theory and the set of tasks, cover more or less different areas of intelligence. In some tests, for example, the performance [...] is more dependent on previous knowledge, in others this is more educationally independent. Some tests only cover a partial ability of intelligence (e.g. abstract logical thinking), others cover a large number of different abilities [...]. Nevertheless, the result is expressed as the intelligence quotient (IQ) in almost all tests. On the face of it, one might think that these are the same abilities recorded. But beware: IQ is not the same IQ, and it also does not give the IQ test "!

According to Jens Asendorpf , the intelligence construct on which classic intelligence tests are based can be characterized as follows: "Intelligence is what intelligence tests measure that are designed to predict the level of education as well as possible, or in short: intelligence tests measure the ability to be highly educated."

Other essential personality dimensions, such as emotional intelligence , are not recorded by the intelligence test.

Standard scales for measuring intelligence

Intelligence tests are developed with appropriately large groups of test persons in such a way that a normal distribution is created. According to Rosemann, the normal distribution of intelligence is an ideal-typical model. Accordingly, tests that show a different distribution of results are considered poorly designed.

The intelligence quotient (IQ) is given a mean value of 100, with the standard deviation being 15:

(HAWIE and HAWIK Hamburg Wechsler intelligence test)

Another common quantity is the Standard Values ​​(SW), which are sometimes confused with the IQ. Here the mean is also 100, but the standard deviation is 10.

( IS intelligence structure test according to Rudolf Amthauer )

When comparing the values ​​of different intelligence tests , it should be noted that different performance requirements (see validity of intelligence measurements) determine the significance of the tests as well as the use of different standardization samples (e.g. different representativeness for the general population) on which the measured value for intelligence is based relates.

Types of intelligence tests

Intelligence tests are each related to an intelligence theory that must be taken into account when interpreting a result. Accordingly, there are tests for recording general intelligence or components (factors, dimensions) of intelligence.

The Binet-Simon test

Alfred Binet and Théodore Simon developed the first usable intelligence test in 1905, the basic ideas of which can be found in all modern tests. The Binet-Simon test consisted of a series of individual but different tasks (subtests). The tasks had an increasing degree of difficulty in order to be able to be solved as clearly as possible for the respective age group. The number of solved tasks was added to a point value. Some of the tasks themselves represent simple everyday questions and problems. Some are logical or mathematical tasks (for example adding a series of numbers ). The first tests also included measuring the memory span . In order to carry out the test, it was necessary to understand the spoken instructions. This gives rise to the justified methodological criticism that people who do not understand this instruction or only understand it inadequately do worse in an intelligence test for this reason alone.

General intelligence

The term general intelligence refers to the general factor model by Charles Spearman . For example, the number connection test (ZVT) by Wolf D. Oswald and Erwin Roth or the matrix tests by John C. Raven are used.

Crystallized (crystalline) and fluid intelligence

This comparison goes back to the British-American psychologist RB Cattell and means the distinction between problem-solving ability (ability to solve problems and to adapt to new problems) - as fluid intelligence and knowledge (the previously acquired knowledge and learning processes as education-dependent component) - as crystallized intelligence (Original: crystallized ). If a person shows different characteristics in both areas, important conclusions can be drawn in both clinical psychology (e.g. degradation diagnostics) and developmental psychology (e.g. is there a suitable level of education?). These skills are particularly measured in the tests developed by Alan S. Kaufman . It enables a differentiation of the skills, acquired knowledge and the basic, fluid skills in problem solving. The well-known test for children from two and a half to twelve and a half years is the Kaufman Assessment Battery for Children (K-ABC). Here a distinction between fluid intelligence, crystallized intelligence and skills (such as spelling or acquired environmental knowledge) is possible. The K-ABC also includes a scale for non-language-related subtests, which provide value when language skills are impaired. The Kaufman test for measuring intelligence for adolescents and adults (K-TIM) has been available for adolescents and adults since 1997 .

Ravens Progressive Matrices

In 1938, John C. Raven developed a culture-independent, language-free method, called progressive matrices , which was supposed to exclude bias for test persons from foreign cultures. However, this hope for completely culture-independent tests has not been fulfilled, as the critics of IQ tests gave good reasons for the fact that culturally different thinking styles and cultural experiences cannot be reduced to just linguistic differences. It turned out that a remnant of “culture confusion” remains even with largely language-free tests.

Verbal and practical (action) intelligence

This subdivision is based on David Wechsler's concept of intelligence, who describes intelligence as "the compound or global ability of the individual to act purposefully, to think rationally and to deal effectively with his environment". The subdivision of intelligence into verbal and action intelligence made by Wechsler in the original version is only one of the many possible structures of this complex construct. It has no theoretically guided basis, but is rather pragmatically motivated; alternative concepts are therefore by no means excluded.

Wechsler's test series, developed in the 1950s, examines general knowledge , vocabulary , arithmetical thinking, audio-visual receptivity and abstraction skills in eleven partial tests . The Hamburg-Wechsler intelligence test for adults (WIE, formerly HAWIE ), modified by the Hamburg psychologist Curt Bondy , is now a common test for test subjects between the ages of 16 and 74. Similarly, there is a test for children between 6 and 16 years of age (WISC, formerly HAWIK ). Another revision from 1983 is indicated by an appended "-R". The child test is now available in a further revised version (WISC-IV, last edition 2010). The Wechsler Preschool and Primary Scale of Intelligence (WPPSI-III, last edition 2011), formerly available under the name Hannover-Wechsler Intelligenztest für die preschool age (HAWIVA), also enables the testing of children between 3 years and 7 years, 2 months .

In contrast to the current versions of the tests (WIE - Wechsler Intelligenztest für Adults; WISC-IV - Wechsler Intelligence Scale for Children, German version) there has been a strict separation of the sub-test tasks into a so-called verbal vs. Plot part. For WPPSI-III, this structure was also retained in the revision. In the two latest editions of the WIE and the WISC, however, this dichotomy was abandoned in favor of a differentiation into four so-called indices, which now describe the performance in the four areas of speech understanding, perceptual logical thinking, working memory and processing speed.

The following assignments apply within the framework of WISC-IV:

  • Understanding of language
    • Find common ground
    • Vocabulary test
    • General understanding
    • optional general knowledge
  • Perceptual logical thinking
    • Mosaic test
    • Image concepts
    • Die test
    • optionally add pictures
  • Working memory
    • Repeat numbers
    • Letters-numbers sequences
    • optional mathematical thinking
  • Processing speed
    • Numbers symbol test
    • Symbol search
    • optional strikethrough test

Multi-factor concepts

Multi-factor concepts are based on models by Louis Leon Thurstone or Adolf Otto Jäger , among others, and capture different sub-components of intelligence. Intelligence tests of this type known in German-speaking countries are:

  • Berlin intelligence structure test ( BIS -4) based on Adolf Otto Jäger's Berlin intelligence structure model
    • Processing speed
    • Retention
    • Ingenuity
    • Processing capacity
    • Linguistic thinking
    • Number-based thinking
    • Outcome thinking
  • Intelligence Structure Test ( IST 2000R )
    • Verbal competence (sentence completion, analogies, similarities)
    • Numerical competence (arithmetic problems, series of numbers, arithmetic symbols)
    • Figurative competence (selection of figures, dice tasks, matrices)
    • Retention , reasoning
    • Extension module: fluid and crystallized intelligence (general factors)
  • Wild Intelligence Test Version 2 ( WIT-2 )
    • Linguistic thinking
    • Computational thinking
    • Spatial thinking
    • Inferential thinking (integral of linguistic, computational and spatial thinking)
    • Retention
    • Work efficiency
    • Knowledge economy
    • Knowledge information technology

Information psychology

Since intelligence tests are performance tests, there is usually a time limit for processing the tests. From the observation that even the stringing together of simple tasks ( elementary cognitive tasks ), for example the task of connecting numbers randomly arranged on a sheet of paper at the highest possible speed in an ascending sequence, results in a measure of intelligence, new theoretical ones have been developing since around 1970 and practical approaches and further developments, for example the number connection test (ZVT) by Oswald and Roth from the task mentioned . By measuring the cognitive information processing speed and the memory span with the KAI test, the short storage capacity could now be calculated, whereby the IQ term was supplemented and challenged by information psychology .

Short tests

Especially in clinical examinations (in Alzheimer's disease, for example) there is a need to estimate the patient's level of intelligence approximately. Since an intelligence test of one or two hours duration is completely impractical in such cases, one strives for short IQ tests in the clinical field. Such a tried and tested short test is the multiple choice vocabulary intelligence test MWT by Siegfried Lehrl .
With the help of the statistical sub-test selection, the WIP, reduced Wechsler intelligence test , was developed from the Wechsler intelligence test (HAWIE) . The WIP is standardized for psychiatric patients. Its presentation time to measure general intelligence is around 15 minutes. The test statistical quality criteria are documented in detail

Intelligence tests and educational evaluation

According to the results of Rindermann (2006), the test procedures used in the PISA studies , TIMSS and IGLU should also be understood as intelligence tests - this is, however, controversial (Baumert et al. 2007). The study regards the validity and reliability of the PISA tests as a measure of the general factor of intelligence as being equivalent to or even higher than the tests listed above. However, PISA tests are not carried out under sufficiently reliable conditions, so that the results can usually not be used for an intelligence statement about the individual students.

Non-language intelligence tests

There are a number of conditions that, if neglected, lead to intuitively false results in IQ tests. Language-heavy IQ tests, for example, require that the person has a good command of the language in which the questions are presented: children of certain groups of guest workers and immigrants often do poorly.

For these cases there are language-free IQ tests such as SPM ( Raven Standard Progressive Matrices , see Progressive Matrices ), the Labyrinth Test according to Porteus and the Culture Fair Intelligence Test (CFIT) (or in the German-speaking world Culture Fair Test CFT ) . The CFT can also be used by people with limited cultural skills. The latest edition of the CFT-20 R also contains an optional (i.e. optional) vocabulary test and a number sequence test, which are intended to measure the ability to make logical conclusions in the linguistic and arithmetical fields. The SON-R 5 1 / 2-17 and the Bochum matrix test (BOMAT) are also examples of speech-free intelligence tests. However, success in non-language intelligence tests is also culture-dependent.


The authors of intelligence tests construct their tests according to implicit assumptions or on the basis of existing intelligence theories. It can be assumed that different intelligence tests capture different segments of intelligence.

Some methods use very different subtests / different item forms (IST2000, BIS), while others use very homogeneous, but particularly representative item forms ( mental speed , matrix tests ).

Some methods are specially designed for the recording of general intelligence ( model of the general factor of intelligence ), while others (WIT, LPS ) are more interested in the specific characteristics of the individual factors of intelligence.

Quality criteria of the tests

Common intelligence tests usually achieve good reliability . The criterion validity and the predictive validity related to school grades are also mostly good. This is because the validity of the tests is designed from the outset according to the differences that can be found between students with higher or lower general education at different types of schools. In the long term, a test has therefore proven to be particularly effective if the success in school correlates with the performance in IQ tests. The underlying assumption that intelligence is a general skill that exists separately from test constructions and can be found, which, from this point of view, is ultimately reflected in school success, is problematized by the test constructors themselves. The solution to the question of whether intelligence tests measure what they are supposed to measure and how well at the same time answers the question of the conceptual content of intelligence, which is no longer up for debate - in the sense of the recognition of intelligence tests.

The predictive power of intelligence tests is often not completely independent of the level of the test value. In the case of very low or very high IQ values, the predictive power (e.g. for professional success) is usually somewhat greater than in the middle range.


Intelligence tests are standardized according to age groups and, if necessary, according to other characteristics such as school leaving certificates. Some authors speak of the so-called calibration sample. This standardization is usually carried out separately for individual countries (also in the German-speaking area). Cross-country comparisons are therefore only possible to a limited extent.

Intelligence tests need to be recalibrated regularly, as the average measured intelligence changes over time. The Flynn effect describes the fact that average IQ scores increased in industrialized countries until the 1990s. The IQ stagnated at the beginning of the 1990s and has even decreased again since the end of the 1990s.

Well-known intelligence tests

IQ test on admission to the US Army
for adults
  • Berlin intelligence structure test (BIS)
  • Intelligence Structure Analysis (ISA-L and ISA-S)
    ISA was first published in 1998, followed by the second edition in 2001, the special forms ISA-L (light) and ISA-S (heavy). This test was used by Mensa in Germany , among others . The ISA is based on the Berlin intelligence structure model, it has nine subtests that measure the four skill areas "verbal intelligence", "numerical intelligence", "figural-spatial intelligence" and "memory performance". The aggregation enables an assessment of the general intelligence. The norms refer to samples with a size of N = 3813 that were collected between 1997 and 1999.
  • Intelligence basis factors (IBF-L and IBF-S) This test is used by Mensa in Germany , among others .
  • Wechsler intelligence test for adults (WIE). This test procedure is often used in forensic reports . The result can provide indications of a reduced culpability or incompetence of a defendant or with regard to the credibility of witnesses who will be followed up in the context of the assessment.
  • WIP - Reduced Changeover Intelligence Test (WIP). The WIP was developed from the statistical sub-test selection of the HAWIE. The battery of four AW, GF, BE, MT has been statistically extensively analyzed and standardized in several clinical and normal groups. The WIP can be carried out in about 15 minutes, so that the time burden for the sick is limited. Multiple correlations of the WIP subtests show good agreement with the overall HAWIE scale and are between 0.89 and 0.97, with the intelligence structure test (IST) at 0.83.
for adults
for teenagers and adults
  • Analytical Intelligence Test (AIT); by R. Meili , published 1971
for teenagers
  • Berlin intelligence structure test for young people: Gifted and gifted diagnostics (BIS-HB)
for children and teenagers
for children
Age-independent / specific age versions


Critics have suggested that class and cultural influences have an impact on the test results. The so - called culture - free tests also target disadvantaged people from minority groups.

The psychologist and educationalist Howard Gardner criticized that there are many things that the IQ does not capture and explain. Gardner claimed the existence of numerous intelligences that the IQ test does not measure. Empirical evidence of such intelligences was not found, however, and Gardner's concept could therefore not gain acceptance in psychology (see main article Theory of Multiple Intelligences ).

According to the Marxist journalist Freerk Huisken, the logical operation for determining "intelligence" assumed in all IQ tests corresponds to the logic of force and expression in Hegel : Intelligence test constructors assumed that a general ability for intelligent performance was found in certain measurable test performances outer. This is justified by a tautological breakdown of the colloquial term intelligence into a self-contained cause-effect relationship: One doubles intelligent performances - out of the need for measurability, generalized as abstract intelligence performance - in their expression ( certain measurable test performances) and that The ability to do so allegedly underlying these statements - expressed in the IQ value. Through this arbitrary breakdown of intelligence into “ability and expression of the same”, the carefully thought-out criteria for the construct intelligence claimed by the test become a measuring instrument for intelligence. A causality is therefore only constituted through the mental separation or duplication. In this respect, Edwin Boring's famous statement from 1923, that intelligence is what the IQ test measures, applies to every intelligence measurement and therefore only allows intelligence to come true as an abstract object under the hand. Certain intelligent achievements are not abstractly comparable in terms of content - unless the will to find abstract intelligence produces them theoretically precisely as what one wants to measure out of certain practical interests (e.g. for fine-tuning selection). It also follows from such a criticism that the type and manner of testing cannot be relevant in detail. The personal taste of the tester and his ideological practical orientation determined the technique of the IQ value measurement ("language-free", "mathematics-free", "milieu-neutral" etc.).

See also


  • Hans Jürgen Eysenck : Intelligence test . 1st edition. Rowohlt, Reinbek near Hamburg 1972, ISBN 3-498-01609-1 .
  • Stephen Jay Gould : The Wrongly Measured Man . 3. Edition. Suhrkamp, ​​Frankfurt am Main 1999, ISBN 3-518-28183-6 .
  • KJ Groffmann: The development of intelligence measurement . In: R. Heiss (Ed.): Psychologische Diagnostik (=  Handbuch der Psychologie . Volume 6 ). CJ Hogrefe, Göttingen 1964, p. 148-199 .
  • Peter Lauster : Test your intelligence. Humboldt-Taschenbuchverlag Jacobi, Munich 1974, ISBN 3-581-66225-6 .
  • Nicholas John MacKintosh: IQ and Human Intelligence . Oxford University Press, Oxford 1998, ISBN 0-19-852368-8 .
  • Franzis Preckel, Matthias Brüll: Intelligence tests . Reinhardt, Munich 2008, ISBN 978-3-8252-3027-2 .
  • Keith E. Stanovich: What Intelligence Tests Miss: The Psychology of Rational Thought . Yale University Press, 2009, ISBN 978-0-300-12385-2 .
  • Marc Wittmann, Andreas Eisenkolb, Christoph Perleth: New intelligence tests . A comprehensive test and exercise program . Augustus Verlag, Augsburg 1997, ISBN 3-8043-3055-X .
  • Rolf Brickenkamp: Manual of psychological and educational tests . Publishing house for psychology. Dr. CJ Hogrefe, Göttingen / Toronto / Zurich 1975, ISBN 3-8017-0092-5 , p. 168-173 .
  • Gerhard Dahl: WIP. Manual for the reduced Wechsler intelligence test, application, evaluation, statistical analysis. Standard values . 2., completely redesigned. and exp. Edition. Verlag Anton Hain, Meisenheim am Glan 1986, ISBN 3-445-02464-2 .

Individual evidence

  1. ^ Edwin Black: War against the weak. Eugenics and America's plan to create a master race. Basic Books, New York 2003.
  2. Stephen Jay Gould: The Mismeasure of Man. Norton, New York 1996.
  3. ^ David G. Myers: Psychology. Worth Publishers, 2008, pp. 405-406.
  4. Finding and promoting gifted children - a guide for parents and school. Information brochure from the Federal Ministry of Education and Research, October 2003, p. 36.
  5. Finding and promoting gifted children - a guide for parents and school. Information brochure from the Federal Ministry of Education and Research, October 2003, pp. 12–13.
  6. ^ Jens Asendorpf : Personality Psychology. Springer Medizin Verlag, Heidelberg 2009, pp. 80–81.
  7. Herrman Rosemann: Intelligence Theories. Rowohlt Verlag, 1979, pp. 53/54.
  8. David G. Myers: Psychology. Springer, Heidelberg 2005.
  9. JC Raven: Progressive Matrices. Lewis, London 1938.
  10. H.-M. Sweet: Culture fair. In: KD Kubinger. RS Jäger (Ed.): Keywords in psychological diagnostics. Beltz, Weinheim 2003.
  11. D. Wechsler: The measurement of the intelligence of adults. Verlag Hans Huber, Bern 1956, p. 13.
  12. ^ David Wechsler: WISC-IV - Manual 1. German version by Franz and Ulrike Petermann. Pearson Assessment, Frankfurt am Main 2010.
  13. ^ David Wechsler: WPPSI-III. German adaptation by Franz Petermann with the assistance of Monika Lipsius. 2., revised. and exp. Edition. Pearson Assessment, Frankfurt am Main 2011.
  14. Martin Kersting: Wilde Intelligenztest 2nd test manual . Hogrefe, Göttingen 2008. (Detailed description)
  15. G. Dahl: WIP. Manual for the reduced changer intelligence test. Application. Evaluation. Statistical analysis. Standard values. 2nd, completely revised and enlarged edition. Anton Hain, Meisenheim am Glan 1986.
  16. R. Brickenkamp: Handbook of psychological and educational tests. Psychology Publishing House Dr. CJ Hogrefe, Göttingen / Toronto / Zurich 1975, pp. 168–173.
  17. Heiner Rindermann: What do international school performance studies measure? School achievement, student skills, cognitive skills, knowledge, or general intelligence? In: Psychological Rundschau . tape 57 , no. 2 , 2006, ISSN  0033-3042 , p. 69-86 . (PDF)
  18. Jürgen Baumert, Martin Brunner, Oliver Lüdtke, Ulrich Trautwein: What do international school performance studies measure? - Results of cumulative knowledge acquisition processes . In: Psychological Rundschau . tape 58 , no. 2 , 2007, ISSN  0033-3042 , p. 118-128 , doi : 10.1026 / 0033-3042.58.2.118 .
  19. Glenn Wilson, Diana Grylls: How Smart My Child is: Intelligence Test for Children . Bastei Lübbe, Bergisch Gladbach 1980, ISBN 3-404-01356-5 , p. 21 .
  21. SO Ortiz, AM Dynda: Use of intelligence tests with culturally and linguistically diverse populations. In: DP Flanagan, PL Harrison (Ed.): Contemporary intellectual assessment. 2nd Edition. Guilford Press, New York 2005, pp. 545-556; see also Raven's Progressive Matrices
  22. Tieze / Tewes 1994, p. 57 ff.
  23. Susanne Donner: Researchers are sounding the alarm: The IQ is on the decline in industrialized countries. In: May 16, 2005, accessed July 25, 2019 .
  24. ^ TW Teasdale, DR Owen: A long-term rise and recent decline in intelligence test performance: The Flynn Effect in reverse. In: Personality and Individual Differences. 39, 2005, pp. 837-843.
  25. JM Sundet, DG Barlaug, TM Torjussen: The end of the Flynn effect? A study of secular trends in mean intelligence test scores of Norwegian conscripts during half a century. In: Intelligence. 32, 2004, pp. 349-362.
  26. ^ Institute for Test and Talent Research (ITB) and G. Gittler: Intelligence Structure Analysis. A test to measure intelligence. Swets Test Services, Frankfurt am Main 1998.
  27. E. Fay, G. Trost, G. Gittler: Intelligence Structure Analysis (ISA). 2., ext. Edition. Swets Test Services, Frankfurt am Main 2001.
  28. N. Ibrahimovic, p Bulheller, R. Horn, G. Gittler, Institute for Test Development and Talent Research GmbH: IBF. Intelligence base factors. Harcourt, Frankfurt am Main 2006.
  29. talentlens: IBF
  30. Analytical Intelligence Test (AIT). In: Dorsch: Lexicon of Psychology. 17th edition.
  31. ^ Anita Woolfolk: Educational Psychology. 10th edition. revised and translated by Prof. Dr. Ute Schönpflug. Pearson Studium, 2008, ISBN 978-3-8273-7279-6 , p. 149 (box).
  32. Freerk Huisken : The science of education. Introduction to the fundamentals of pedagogy. Hamburg 1991, pp. 107-120.
  33. Power and Expression. In: GWF Hegel: Logic. II, pp. 172-179.
  34. Ingeborg Titze, Uwe Tewes: Measurement of intelligence in children with the HAWIK-R. Bern / Göttingen / Toronto / Seattle 1984/1994, pp. 13-21.
  35. ^ Edwin G. Boring: Intelligence as the Tests Test It. In: New Republic. Volume 36, 1923, pp. 35-37.
  36. Helmut Heid, Gabi Fink: Talent. In: Dietrich Benner, Jürgen Oelkers (Hrsg.): Historical dictionary of pedagogy. Weinheim / Basel 2004, pp. 146–152.