Genius Association Test statistics
© March 2015 Paul Cooijmans
Norms
Scores on Genius Association Test
Contents type: Verbal. Period: 2003-present
16 | * |
19 | *** |
20 | * |
21 | * |
22 | ** |
23 | * |
24 | * |
25 | ** |
26 | ** |
26.5 | * |
27 | ***** |
28 | ** |
30 | * |
31 | ****** |
32 | ********* |
33 | **** |
34 | ******* |
34.5 | * |
35 | *** |
36 | *** |
37 | *** |
37.5 | * |
38 | ***** |
39 | * |
40 | **** |
41 | ** |
42 | ** |
43 | * |
44 | *** |
44.5 | * |
46 | * |
47 | **** |
48 | * |
49.5 | ** |
Scores by males
n = 82
16 | * |
19 | ** |
20 | * |
21 | * |
22 | ** |
23 | * |
24 | * |
25 | ** |
26 | ** |
26.5 | * |
27 | ***** |
28 | ** |
31 | ****** |
32 | ********* |
33 | **** |
34 | ****** |
34.5 | * |
35 | *** |
36 | *** |
37 | *** |
37.5 | * |
38 | ***** |
39 | * |
40 | **** |
41 | * |
42 | ** |
43 | * |
44 | ** |
44.5 | * |
46 | * |
47 | **** |
48 | * |
49.5 | ** |
Scores by females
n = 5
Correlation of Genius Association Test with other tests by Paul Cooijmans
(Test index) Test name | n | r |
(44) Associative LIMIT | 42 | 0.93 |
(85) Cooijmans Intelligence Test - Form 1 | 9 | 0.85 |
(53) Qoymans Multiple-Choice #3 | 18 | 0.83 |
(52) Qoymans Multiple-Choice #2 | 7 | 0.81 |
(56) Short Test For Genius | 6 | 0.72 |
(26) Verbal section of Test For Genius - Revision 2004 | 36 | 0.72 |
(75) Analogies of Long Test For Genius | 22 | 0.71 |
(63) Long Test For Genius | 20 | 0.71 |
(83) KIT Intelligence Test - first attempts | 7 | 0.70 |
(23) Gliaweb Riddled Intelligence Test - Revision 2011 | 10 | 0.69 |
(43) Test For Genius - Revision 2010 | 10 | 0.67 |
(7) The Final Test | 32 | 0.67 |
(18) The Nemesis Test | 9 | 0.66 |
(69) Odds | 8 | 0.64 |
(35) Intelligence Quantifier by assessment | 23 | 0.64 |
(87) Cooijmans Intelligence Test - Form 2 | 18 | 0.64 |
(21) Psychometric Qrosswords | 5 | 0.62 |
(66) Test For Genius - Revision 2004 | 27 | 0.60 |
(28) The Test To End All Tests | 17 | 0.59 |
(62) Reason Behind Multiple-Choice | 13 | 0.59 |
(12) Cooijmans On-Line Test (Two-barrelled) | 7 | 0.57 |
(57) Space, Time, and Hyperspace | 24 | 0.57 |
(40) Reason Behind Multiple-Choice - Revision 2008 | 19 | 0.55 |
(104) The Final Test - Revision 2013 | 5 | 0.54 |
(55) Spatial Insight Test | 9 | 0.52 |
(0) Test of the Beheaded Man | 9 | 0.52 |
(79) Association subtest of Long Test For Genius | 26 | 0.52 |
(16) Lieshout International Mesospheric Intelligence Test | 42 | 0.51 |
(106) Cooijmans Intelligence Test - Form 4 | 8 | 0.51 |
(103) Problems In Gentle Slopes of the second degree | 6 | 0.51 |
(24) Reason - Revision 2008 | 19 | 0.49 |
(2) Cooijmans Intelligence Test - Form 3 | 19 | 0.48 |
(3) Qoymans Multiple-Choice #5 | 19 | 0.45 |
(80) Qoymans Multiple-Choice #4 | 33 | 0.43 |
(82) Reason | 13 | 0.39 |
(4) A Paranoiac's Torture: Intelligence Test Utilizing Diabolic Exactitude | 5 | 0.37 |
(25) The Sargasso Test | 13 | 0.37 |
(48) Narcissus' last stand | 5 | 0.36 |
(27) Spatial section of Test For Genius - Revision 2004 | 36 | 0.36 |
(11) Isis Test | 13 | 0.32 |
(54) Test of Shock and Awe | 14 | 0.31 |
(68) Numbers | 17 | 0.31 |
(84) Bonsai Test | 8 | 0.29 |
(31) Numerical section of The Marathon Test | 5 | 0.21 |
(42) The Marathon Test | 5 | 0.21 |
(15) Letters | 8 | 0.21 |
(19) Numerical section of Test For Genius - Revision 2010 | 12 | 0.20 |
(45) Numerical and spatial sections of The Marathon Test | 5 | 0.18 |
(30) Verbal section of The Marathon Test | 7 | 0.17 |
(1) Cartoons of Shock | 20 | 0.15 |
(32) Spatial section of The Marathon Test | 5 | 0.14 |
(39) Combined Numerical and Spatial sections of Test For Genius - Revision 2010 | 11 | 0.10 |
(36) Reflections In Peroxide | 5 | 0.09 |
(41) The LAW - Letters And Words | 6 | 0.07 |
(51) Qoymans Multiple-Choice #1 | 8 | -0.03 |
(29) Words | 7 | -0.63 |
Weighted average of correlations: 0.512
Conservatively estimated minimum g loading: 0.72
Ranking in above table is based on the unrounded correlations. All available data is present in this table, no tests are left out except for those with less than 5 score pairs. All known pairs are used, including possible floor/ceiling scores or outliers.
Correlation of Genius Association Test with tests by others
(Test index) Test name | n | r |
(235) Nonverbal Cognitive Performance Examination | 11 | 0.79 |
(214) Epiq Tests | 5 | 0.75 |
(240) Strict Logic Spatial Exam 48 | 7 | 0.52 |
(234) Strict Logic Sequences Exam I | 14 | 0.51 |
(239) Titan Test | 9 | 0.47 |
(201) Wechsler Adult Intelligence Scales | 9 | 0.47 |
(242) Unknown and miscellaneous tests | 37 | 0.43 |
(224) T.R.I. | 5 | 0.42 |
(206) W-87 | 5 | 0.35 |
(238) 916 Test | 8 | 0.30 |
(236) International High IQ Society Miscellaneous tests | 15 | 0.26 |
(230) Omega Contemplative Items Pool | 5 | 0.25 |
(231) Mysterium Entrance Exam | 8 | 0.20 |
(225) Logima Strictica 36 | 15 | 0.13 |
(211) Culture Fair Numerical Spatial Examination - Final version | 18 | 0.03 |
(223) Strict Logic Sequences Exam II | 6 | -0.12 |
(248) Gliaweb Riddled Intelligence Test (old version) | 7 | -0.13 |
(229) Mega Test | 7 | -0.16 |
(218) Raven's Advanced Progressive Matrices (I.Q.) | 6 | -0.25 |
(246) Sequentia Numerica Form I | 5 | -0.36 |
(243) Scholastic Aptitude Test (old) | 5 | -0.43 |
(220) Cattell Culture Fair | 5 | -0.49 |
Weighted average of correlations: 0.248
Ranking in above table is based on the unrounded correlations. All available data is present in this table, no tests are left out except for those with less than 5 score pairs. All known pairs are used, including possible floor/ceiling scores or outliers.
Please be aware that correlations with these external tests are in most cases affected (depressed, typically) by one or more of the following: (1) Little overlap with the object test because of the much lower ceilings and inherent ceiling effects of the tests used in regular psychology; (2) Candidates reporting scores selectively, for instance only the higher ones while withholding lower ones; (3) Candidates reporting, or having been reported by psychometricians, incorrect scores.
Estimated loadings of Genius Association Test on particular item types
These are estimated g factor loadings, but against homogeneous tests (containing only particular item types) as opposed to non-compound heterogeneous tests. Although tending to surprise the lay person, it is not uncommon for tests to have high loadings on item types they do not actually contain themselves. Such loadings reflect the empirical fact that most tests for mental abilities measure primarily g, regardless of their contents; that the major part of test score variance is caused by g, and only a minor part by factors germane to particular item types. It is of key importance to understand that this is a fact of nature, a natural phenomenon, and not something that was built into the tests by the test constructors.
Type | g loading of Genius Association Test on that type |
Verbal | 0.72 |
Numerical | 0.57 |
Spatial | 0.68 |
Logical | 0.67 |
Heterogeneous | 0.68 |
Compound tests have been left out of this table to avoid overlap.
Balanced g loading = 0.66
National medians for Genius Association Test
Country | n | median score |
Belgium | 2 | 40.8 |
United_Kingdom | 6 | 40.8 |
Germany | 5 | 36.0 |
Australia | 2 | 35.0 |
Finland | 5 | 34.0 |
United_States | 29 | 34.0 |
Canada | 5 | 32.0 |
Spain | 3 | 32.0 |
Sweden | 7 | 31.0 |
Greece | 3 | 27.0 |
Norway | 2 | 27.0 |
South_Africa | 3 | 26.5 |
For reasons of privacy, only countries with 2 or more candidates are included in this table. Ranking is based on the medians, and then alphabetic.
Correlation with national I.Q.'s of Genius Association Test
Correlation of this test with national average I.Q.'s published by Lynn and Vanhanen:
Correlation of Genius Association Test with personal details
Personalia | n | r |
P.S.I.A. Ethics factor | 54 | 0.35 |
P.S.I.A. True | 39 | 0.35 |
Observed associative horizon | 10 | 0.32 |
Observed behaviour | 21 | 0.19 |
P.S.I.A. Rational | 39 | 0.19 |
Father's educational level | 74 | 0.11 |
Disorders (own) | 83 | 0.09 |
P.S.I.A. Deviance factor | 54 | 0.08 |
P.S.I.A. System factor | 33 | 0.08 |
P.S.I.A. Orderly | 39 | 0.03 |
Educational level | 83 | 0.02 |
P.S.I.A. Extreme | 39 | 0.01 |
Sex | 87 | 0.01 |
Mother's educational level | 76 | 0.01 |
Disorders (parents and siblings) | 83 | -0.01 |
P.S.I.A. Introverted | 39 | -0.06 |
P.S.I.A. Cold | 39 | -0.06 |
P.S.I.A. Aspergoid | 39 | -0.07 |
P.S.I.A. Rare | 39 | -0.08 |
P.S.I.A. Neurotic | 39 | -0.09 |
Candidate's self-estimated I.Q. | 9 | -0.11 |
Cooijmans Inventory of Neo-Marxist Attitudes | 7 | -0.19 |
Year of birth | 86 | -0.23 |
P.S.I.A. Cruel | 39 | -0.28 |
P.S.I.A. Just | 39 | -0.32 |
P.S.I.A. Antisocial | 39 | -0.33 |
Gifted Adult's Inventory of Aspergerisms | 22 | -0.43 |
Correlation with personal details of Genius Association Test - within females
Personalia | n | r |
P.S.I.A. Deviance factor | 3 | 0.96 |
P.S.I.A. Ethics factor | 3 | 0.94 |
Educational level | 5 | 0.63 |
P.S.I.A. System factor | 3 | 0.43 |
Father's educational level | 5 | 0.22 |
Mother's educational level | 5 | 0.02 |
Disorders (own) | 5 | -0.15 |
Disorders (parents and siblings) | 5 | -0.15 |
Year of birth | 5 | -0.92 |
Correlation with personal details of Genius Association Test - within males
Personalia | n | r |
Observed associative horizon | 8 | 0.47 |
P.S.I.A. True | 39 | 0.35 |
P.S.I.A. Ethics factor | 51 | 0.30 |
P.S.I.A. Rational | 39 | 0.19 |
Observed behaviour | 18 | 0.13 |
Disorders (own) | 78 | 0.10 |
Father's educational level | 69 | 0.10 |
P.S.I.A. Deviance factor | 51 | 0.06 |
P.S.I.A. Orderly | 39 | 0.03 |
P.S.I.A. Extreme | 39 | 0.01 |
Mother's educational level | 71 | 0.00 |
Disorders (parents and siblings) | 78 | 0.00 |
P.S.I.A. System factor | 30 | -0.00 |
Educational level | 78 | -0.03 |
P.S.I.A. Introverted | 39 | -0.06 |
P.S.I.A. Cold | 39 | -0.06 |
P.S.I.A. Aspergoid | 39 | -0.07 |
P.S.I.A. Rare | 39 | -0.08 |
P.S.I.A. Neurotic | 39 | -0.09 |
Candidate's self-estimated I.Q. | 8 | -0.13 |
Cooijmans Inventory of Neo-Marxist Attitudes | 7 | -0.19 |
Year of birth | 81 | -0.19 |
P.S.I.A. Cruel | 39 | -0.28 |
P.S.I.A. Just | 39 | -0.32 |
P.S.I.A. Antisocial | 39 | -0.33 |
Gifted Adult's Inventory of Aspergerisms | 22 | -0.43 |
Estimated g factor loadings upward and downward of particular scores
In parentheses the number of score pairs on which that estimated g factor loading is based. The goal of this is to verify the hypothesis that g becomes less important, accounts for a smaller proportion of the variance, at higher I.Q. levels. The mere fact of restricting the range like this also depresses the g loading compared to computing it over the test's full range, so it would be normal for both values to be lower than the test's full-range g loading.
Raw score | Upward g (n) | Downward g (n) |
0 | 0.72 (812) | NaN (0) |
23 | 0.70 (758) | 0.42 (6) |
28.5 | 0.67 (574) | 0.42 (117) |
34 | 0.72 (301) | 0.60 (465) |
39.5 | 0.76 (101) | 0.64 (613) |
44 | 0.65 (40) | 0.66 (741) |
60 | NaN (0) | 0.72 (812) |
Reliability
Error
Scores by age
Age class | n | median score |
55 to 59 | 5 | 34.0 |
50 to 54 | 8 | 34.5 |
45 to 49 | 13 | 36.0 |
40 to 44 | 11 | 37.0 |
35 to 39 | 8 | 35.5 |
30 to 34 | 10 | 29.5 |
25 to 29 | 11 | 34.0 |
22 to 24 | 11 | 32.0 |
20 or 21 | 1 | 37.0 |
18 or 19 | 5 | 32.0 |
17 | 1 | 36.0 |
16 | 2 | 25.5 |
Scores by age - within females
Age class | n | median score |
50 to 54 | 2 | 37.0 |
45 to 49 | 1 | 41.0 |
40 to 44 | 1 | 34.0 |
16 | 1 | 19.0 |
Scores by age - within males
Age class | n | median score |
55 to 59 | 5 | 34.0 |
50 to 54 | 6 | 34.5 |
45 to 49 | 12 | 35.0 |
40 to 44 | 10 | 37.5 |
35 to 39 | 8 | 35.5 |
30 to 34 | 10 | 29.5 |
25 to 29 | 11 | 34.0 |
22 to 24 | 11 | 32.0 |
20 or 21 | 1 | 37.0 |
18 or 19 | 5 | 32.0 |
17 | 1 | 36.0 |
16 | 1 | 32.0 |
Scores by year taken
Year taken | n | median score |
2003 | 9 | 38.0 |
2004 | 27 | 31.0 |
2005 | 14 | 34.5 |
2006 | 3 | 41.0 |
2007 | 5 | 32.0 |
2008 | 6 | 36.0 |
2009 | 3 | 44.5 |
2010 | 4 | 29.3 |
2011 | 6 | 38.8 |
2012 | 2 | 31.3 |
2013 | 2 | 31.0 |
2014 | 3 | 36.0 |
2015 | 3 | 32.0 |
ryear taken × median score = -0.22 (n = 87)
Robustness and overall test quality
Item analysis
Item statistics are not published as that would help future candidates. To detect bad items, answers and comments from candidates are studied, as well as, for each problem, the correlation with total score and the proportion of candidates getting it wrong (hardness of the item). Possible bad items are removed or revised, resulting in a revised version of the test. So far, no bad items have been detected in this test.