Validity of intelligence measurements

from Wikipedia, the free encyclopedia

The connection between intelligence and numerous other characteristics of people is examined in the context of studies on the validity of intelligence tests. The measurement of a person's intelligence using intelligence tests is used in differential psychology for basic research (e.g. uncovering connections between intelligence and other personality traits). In psychological diagnostics , this serves as a basis for diagnostic decisions (e.g. suitability for educational pathways or career choice, presence of a disorder). A prerequisite for this use is that based on the measured intelligence performance, well-founded prognoses with regard to the further characteristics (criteria) are possible and such relationships have been empirically proven within the framework of basic research.

It should be noted that correlation studies initially do not reveal any statements about causality . The external or prognostic criterion validity is usually estimated using these correlation coefficients .

Intercorrelations of the results of various intelligence tests

In psychology, intelligence is not understood as a one-dimensional construct (see also intelligence theory ). The intelligence test used also decides which aspects of intelligence have been measured. Therefore, the correlations with external criteria are also dependent on the test and the choice of the right procedure is one of the important tasks of psychological diagnostics .

For a German sample, for example, the correlation coefficients of the intelligence test CFT 20 with other intelligence tests (e.g. PSB, HAWIK, CPM, WIP) are on average r = 0.64 and range from r = 0.57 to r = 0.73. The Raven Advanced Progressive Matrices (APM), which are often used for the speech-free recording of the general factor of intelligence according to the Spearman model , correlate, for example, with the number connection test (ZVT) only with 0.33, although the ZVT should also record this general factor . This low correlation can be explained, among other things, by the fact that individual tests do not represent a sufficient operationalization of the general factor of intelligence . However, if the connection between comparable intelligence test performances is examined, almost perfect correlations emerge.

Intelligence and school / professional success in general

In his book Personality Psychology, Asendorpf summarizes various studies and states that the measured IQ correlates with a variety of criteria for successful integration into a performance society in which success is measured in terms of achievements in training and work. For example, it correlates with school grades, university grades, long-term educational level and professional prestige.

The following table provides an overview of typical mean correlations between intelligence test results and these variables .

Performed correlation
Medium grade in elementary school 0.5
Abitur grade 0.3
Final grade of a university degree 0.35
Success in the job as assessed by the superior 0.25
Highest educational qualification achieved at the age of 40 0.7
Professional prestige at the age of 40 (men) 0.7

Intelligence and academic success

The correlations between measured intelligence and the teacher's judgment about the intellectual level of the student are extremely high compared to the correlations otherwise common in psychology and amount to about 0.60. In large part, this phenomenon is due to the fact that this teacher's judgment is one of the two main criteria against which intelligence tests are validated (the second is school grades). From the entire spectrum of human talents, intelligence tests primarily capture those that enable an individual to be successful at school.

Since intelligence tests are validated on school grades, among other things, they are good predictors of school performance. The correlations between intelligence and school success are among the highest in psychological diagnostics .

A high correlation does not mean that school performance is 100% determined by the IQ. Using the example of the IST 2000 intelligence test, it can be shown how high the IQ correlates with the various school performance. The example relates to German and Swiss students. However, similar results can also be found in other countries:

grade Correlation between grade and test performance in the IST-2000
German 0.46 (a) (Corr. Relates only to scales for measuring verbal intelligence)
geography 0.38 (b)
physics 0.36 (b)
mathematics 0.35 (a) (Corr. Relates only to scales for measuring numerical intelligence)
history 0.34 (b)
English 0.20 (b)

In the PISA studies , in addition to performance, “problem-solving skills” were also examined. Problem-solving competence is an interdisciplinary competence. It describes the ability to think conclusively, to analyze, to recognize connections and to weigh ideas against each other. It is very similar to intelligence. Problem-solving skills are required for all subjects. Adolescents from the lower social classes generally have less problem-solving skills (intelligence) than adolescents from the upper classes. But even with the same problem-solving skills, young people from lower social classes have less professional skills and go to high school less often. See also: Analysis of the PISA studies: influence of social background

In 1990 it was also possible to prove that girls' intelligence correlates more strongly with school grades than boys. This is apparently because boys are less integrated into school operations and pursue more interests outside of school.

According to Jens Asendorpf , the level of education (the highest school-leaving qualification achieved at the age of 40, from a special school certificate to a doctorate) correlates to 0.70 with the IQ. The fact that intelligence correlates more strongly with the level of education than with individual grades is attributed to the fact that a single grade is strongly individual teacher is determined. However, the level of education is determined by all certificates, i.e. the grades of all teachers. This means that the data is more highly aggregated here.

Lewis M. Terman followed the lives of 1,528 gifted people . He found that gifted people usually develop very positively (more on this under Terman study ). Joel. N. Shurkin has criticized the results of Terman in his book "Terman's Kids: The Groundbreaking Study of How the Gifted Grow Up". The Marburg gifted project also dealt with this population group. The MHP researchers came to the following conclusion: "In summary, the highly gifted can be characterized as being well integrated in the school system, successful at school, socially inconspicuous, psychologically particularly stable and self-confident."

Intelligence and success at university and in on-the-job training

In a meta-study, which summarized six other studies, it was found that there is a connection between IQ and training / study success. Smart people are more successful. The average correlation was 0.44.

Intelligence and professional success

Intelligence is significantly related to professional success. However, other factors must also be taken into account - such as expert knowledge .

According to somewhat older American studies, professional success (measured by professional status) is determined most strongly by motivation to perform (correlation: 0.43), second most by IQ (correlation: 0.33) and third by social class (correlation: 0.23).

It should be noted that these variables can correlate with one another to different degrees. For example, in a study of children, there was a significantly stronger correlation between intelligence and achievement motivation in working-class boys than in boys from the middle class (0.4 vs. 0.17). One explanation for this could be that teachers generally turn more to intelligent working-class children than less competent working-class children. Furthermore, the difference experienced between low social status and high competence could provide further impetus. Overall, however, it is to be expected that intelligent working-class children have a higher level of achievement motivation than similarly intelligent children from the middle class. These differences may persist into adulthood.

Other studies come to the conclusion that the prestige of the occupation carried out correlates with the IQ with around 0.57 to 0.71.

It could be shown that occupational status (regardless of whether it is defined by income or occupational prestige) is highly correlated with IQ; but it would be premature to conclude that a high IQ leads directly to a steep career. It is possible that it is not the IQ itself, but rather the success at school and university that is the ticket to successful professional life. The IQ would then (because it correlates with school success) predict the level of the entry position only indirectly. After starting a career, it depends on the type of activity carried out whether the IQ correlates with professional success. Relatively high correlations of 0.47 are only available for academics. In the trades, there are low correlations of 0.00 to 0.19. There is even a negative correlation between IQ test performance and those professional performances that primarily require simple repetition of work processes. This can be interpreted as an indication that a high IQ is a handicap when performing simple routine activities. The more academic demands a profession has, the higher the correlations. This is why IQ values ​​correlate more closely with performance in vocational training programs than later with performance in the job itself.

There are certain minimum IQ levels below which certain high status jobs cannot be exercised. Conversely, however, there are people with a high IQ in occupations with low status. The reasons for this are lack of motivation to perform, family relationships, different goals or individual living conditions.

A study by Stephen J. Ceci (1996) with the question “What is better, to be rich or to be smart?” Showed that, at least in the USA, social origin had a much stronger influence on later income than did Intelligence.

Intelligence and "assortative pairing" (choice of partner)

It can be seen that people tend to choose a partner with similar academic skills and intelligence. In one study, for example , the ability of spouses in reading, spelling and arithmetic, the size of the vocabulary of the spouses and the verbal IQ and total IQ of the two spouses correlated with one another (see also Assortative pairing ).

Intelligence and hereditary diseases

Certain hereditary diseases are particularly common in families of intelligent people:

The Tay-Sachs disease is an example. The syndrome is caused by a mutation on chromosome 15 , locus 15q23-24. If you have two defective copies of this gene, you get Tay-Sachs syndrome. As the disease progresses, there is therefore a progressive decline in cognitive abilities (decline in intelligence) and a loss of mobility and vision. The disease leads to death in childhood. It is surprising that such a disease has not yet died out in the course of human history. The reason seems to be that people with only one defective gene are more intelligent than people with two healthy genes.

The literature on the relationship between IQ and schizophrenia is inconsistent. Schizophrenia seems to be rather rare in people of average intelligence. On the one hand, people with below-average intelligence and school difficulties have an increased risk of developing schizophrenia. On the other hand, the disease seems to affect mathematically gifted people and their families particularly often. The COMT gene is considered to be a jointly responsible gene in schizophrenia. At the same time, connections between polymorphisms of the Comt gene rs 165599 and performance on IQ tests could be demonstrated.

Research on this topic is still in its infancy and the scientific results are by no means certain.

Intelligence and myopia

Various studies have shown a correlation between IQ and myopia. On average, nearsighted people achieve about 7 to 9 points more in the IQ test than normal sighted people. Gifted people are particularly often nearsighted. A study of 157,748 Israeli recruits found that 8% of the people with the lowest IQ were nearsighted, compared with 27.3% of the group with the highest IQ.

Further findings

There are many other studies that identify relationships between intelligence and other characteristics of a person . For example, people with a low IQ smoke and are more likely to be delinquent than people with a higher IQ. Less intelligent people have a shorter life expectancy than intelligent people. One of the reasons for a shorter life expectancy could be a longer response time; This could, for example, lead to less intelligent people having road accidents more often. In addition, less intelligent people are more likely to commit suicide . Furthermore, different studies independently of one another showed that people with a pronounced tendency towards left-liberal and atheistic views have an IQ that is more than 10 points higher than people who rate themselves as distinctly conservative .

Intelligence correlates positively with resilience .

See also

literature

  • Jens Asendorpf : Personality Psychology . Springer Medizin Verlag, Heidelberg 2009, Chapter 4.3.1.

Individual evidence

  1. ↑ Rudolf H. Weiß : Basic intelligence test scale 2 (CFT 20) with vocabulary test (WS) and number sequence test (ZF) . 4th edition
  2. ↑ Test label APM (PDF; 62 kB) Specialist group Diagnostics of the SDBB
  3. ↑ Jensen, AR, & Wang, L.-J. (1994). What is a good g? Intelligence, 18, 231-258.
  4. ^ Reeve, CL, & Blacksmith, N. (2009). Identifying g: A review of current factor analytic practices in the science of mental abilities. Intelligence, 37 (5), 487-494. doi : 10.1016 / j.intell.2009.06.002
  5. ↑ Gignac, GE (2015). Raven's is not a pure measure of general intelligence: Implications for g factor theory and the brief measurement of g. Intelligence, 52, 71-79. doi : 10.1016 / j.intell.2015.07.006
  6. ↑ Johnson, W., Nijenhuis, J. te, & Bouchard, TJ (2008). Still just 1 g: Consistent results from five test batteries. Intelligence, 36 (1), 81-95. doi : 10.1016 / j.intell.2007.06.001
  7. ↑ Valerius, S., & Sparfeldt, JR (2014). Consistent g- as well as consistent verbal, numerical and figural factors in nested factor models? Confirmatory factor analyzes using three test batteries. Intelligence, 44, 120-133. doi : 10.1016 / j.intell.2014.04.003
  8. ^ Jens B. Asendorpf: Personality Psychology. Springer medical publisher. Heidelberg. 2009. p. 80
  9. ↑ Joachim Funke , Bianca Vaterrodt-Plünnecke: What is intelligence? 2nd, revised edition. Verlag CH Beck, Munich 2004, p. 80
  10. ↑ Elsbeth Stern, Ilonca Hardy: Differential Psychology of Learning in School and Training. In: Birbaumer et al .: Encyclopedia of Psychology - Subject Area C: Theory and Research - Series VIII: Differential Psychology and Personality Research - Volume 5 Theories and Fields of Application. Hogrefe Verlag, 2004, ISBN 3-8017-0534-X , p. 595
  11. ^ Psychological review. 2/2007, p. 132
  12. ↑ Swiss calibration for the intelligence structure test 2000. (PDF; 67 kB) Spectrum career advice; Retrieved October 9, 2012 (in the case of the correlations marked with (a); test calibrated on a Swiss sample) and label of the IST 2000, SVB Diagnostic Commission (PDF; 60 kB): IST 2000: The German norms for adults are also in the Switzerland applicable. accessed October 9, 2012 (in the case of the correlations marked with (b), test calibrated on German samples)
  13. ↑ Manfred Amelang: Differential Psychology and Personality Research , 3rd revised and expanded edition. Kohlhammer, Stuttgart / Berlin / Cologne 1990, ISBN 3-17-010747-X , p. 234
  14. ^ Jens B. Asendorpf: Personality Psychology . Springer medical publisher. Heidelberg. 2009. p. 80
  15. ↑ Jens B. Asendorpf: Psychology of Personality. 2nd revised and updated edition. Springer, Berlin / Heidelberg / New York 1999, ISBN 3-540-66230-8 , p. 171
  16. ↑ JN Shurkin: Terman's kids: The groundbreaking study of how the gifted grow up. Little Brown, Boston, 1992
  17. ↑ Detlef H. Rost (ed.): Highly gifted and high-performing young people. New results from the Marburg gifted project. Waxmann, Münster 2000, ISBN 3-89325-685-7
  18. ↑ a b Elsbeth Stern, Ilonca Hardy: Differential Psychology of Learning in School and Training. In: Birbaumer et al .: Encyclopedia of Psychology - Subject Area C: Theory and Research - Series VIII: Differential Psychology and Personality Research - Volume 5 Theories and Fields of Application. Hogrefe Verlag, 2004, ISBN 3-8017-0534-X , p. 596.
  19. ↑ Intelligence leads to success - meta-analysis confirms earlier studies.
  20. ↑ Elder: Earning a living . In: Ders .: Children of the Great Depression. University of Chicago Press, Chicago 1974, p. 175
  21. ↑ Elder: Earning a living . In: Ders .: Children of the Great Depression. University of Chicago Press, Chicago 1974, pp. 176f.
  22. ↑ Manfred Amelang: Differential Psychology and Personality Research , 3rd revised and expanded edition. Kohlhammer, Stuttgart / Berlin / Cologne 1990, ISBN 3-17-010747-X , p. 237
  23. ↑ Joachim Funke, Bianca Vaterrodt-Plünnecke: What is intelligence? 2nd, revised edition. Verlag CH Beck, Munich 2004, p. 84
  24. ↑ Eysenck: Intelligence test. Weltbildverlag, Augsburg 1996. Original edition: Eysenck: Know your Own IQ Pengiun Books, Hormondsworth / Middlesex 1962
  25. ^ Later Professor of Developmental Psychology
  26. ^ Yue-Pan, Ke-Sheng Wang: Spousal concordance in academic achievements and IQ. A principal component analysis. In: Open Journal of Psychiatry. Volume 1, 2011, pp. 14-19
  27. ↑ Robert J. Desnick (Ed.): Tay-Sachs disease. Acad. Press, San Diego [et al. a.] 2001
  28. ↑ Análise molecular das doenças de Gaucher e Tay-Sachs no Brasi.
  29. ^ F. Post: Creativity and psychopathology: a study of 291 world famous men . In: Bristish J. of Psychiatry , 165, 1994, pp. 22-24.
  30. ^ Jon L. Karlsson: Psychosis and academic performance. In: The British Journal of Psychiatry. Mar 2004, 184 (4) 327-329; DOI: 10.1192 / bjp.184.4.327
  31. ↑ HJ Möller: The influence of genetic variations in the COMT gene on cognitive phenotypes. (PDF; 1.3 MB) accessed on February 7, 2008
  32. ↑ Andrew A. Fingelkurts, Alexander A. Fingelkurts: Exploring Giftedness. ( Memento of the original from June 16, 2009 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice. (PDF; 233 kB) accessed on October 9, 2012 @1@ 2Template: Webachiv / IABot / www.bm-science.com
  33. ^ GC Ashton: Myopia and cognitive ability. In: Behav. Genetics. Volume 13, 1983, p. 526
  34. ↑ D. Lubinski, L. Humphreys: Some bodily and medical correlates of mathematical giftedness and commensurate levels of socioeconomic status. ( Memento of the original from August 19, 2012 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice. (PDF; 3.5 MB) In: Intelligence. Volume 16, 1992, pp. 99-115 @1@ 2Template: Webachiv / IABot / www.vanderbilt.edu
  35. ↑ Myopic people are more intelligent . In: Die Zeit , No. 42/1999
  36. ^ JJ Ray: Smoking and intelligence in Australia. In: Social Science and Medicine. Volume 20, 1985, pp. 1279-1280.
  37. ^ T. Hirschi, MJ Hindelang: Intelligence and Delinquency: A Revisionist Review. In: American Sociological Review. Volume 42, No. 4, 1977, pp. 571-587
  38. ^ Ian J. Deary, Geoff Der: Reaction Time Explains IQ's Association With Death. In: Psychological Science. Volume 16, No. 1, 2005, pp. 64-69, doi: 10.1111 / j.0956-7976.2005.00781.x
  39. ^ AS David, A. Malmberg, L. Brandt, P. Allebeck, G. Lewis: IQ and risk for schizophrenia: a population-based cohort study . In: Psychological Medicine . tape 27 , no. 6 , 1997, pp. 1311-1323 , PMID 9403903 .
  40. ↑ Satoshi Kanazawa: Why Liberals and Atheists Are More Intelligent. In: Social Psychology Quarterly. Volume 73, 2010, pp. 33-57, doi: 10.1177 / 0190272510361602
    Intelligent people have 'unnatural' preferences and values ​​that are novel in human evolution. Higher intelligence is associated with liberal political ideology, atheism, and men's (but not women's) preference for sexual exclusivity. eurekalert.org, February 24, 2010
    Conservatives are less intelligent . sueddeutsche.de, February 26, 2010
    see also: Conservatives have lower IQs . Spiegel Online , March 1, 2010