The Nemesis Test - Statistics

Norms

# right	Protonorm
0	340 or lower
.5	360
1	379
1.5	381
2	393
2.5	400
3	413
3.5	433
4	447
4.5	487
5	500
5.5	507
6	520
6.5	527
7	533
7.5	540
8	545
8.5	546
9	548
9.5	549
10	553
10.5	560
11	567
11.5	573
12	587
12.5	593
13	607
13.5	620
14	640
14.5	647
15	660
15.5	667
16	680
16.5	693
17	707
17.5	720
18	733
18.5	747
19	760
19.5	773
20	787
20.5	800
21	820
21.5	840
22	860

Quality of norms: .84

Used for norming (with the usual norming method) are prior tests correlating .44 or higher with the Nemesis Test in below table.

Scores so far

# testees: 10
Mean: 3.3
SD: 3.09
Male mean: 3.67
Female mean: 0 (1 persons)

0	***
2	*
3	**
4	*
4.5	*
6	*
10.5	*

Correlation with other tests

Test	#	cor
Genius Association Test	3	0.99
Numbers	4	0.99
Short Test For Genius	3	0.99
Cooijmans Intelligence Test - Form 2	3	0.97
LIMIT	3	0.97
Cooijmans Intelligence Test	3	0.95
Long Test For Genius	5	0.86
Analogies of LTFG	5	0.85
KIT Intelligence Test	3	0.83
Association Subtest of LTFG	5	0.82
Spatial Insight Test	3	0.82
Test To End All Tests	7	0.77
Cartoons of Shock	3	0.76
Space, Time & Hyperspace	6	0.76
Reason	3	0.7
Final Test	5	0.63
Test of Shock and Awe	4	0.6
Spatial section of Test For Genius Revision 2004	6	0.44
Test For Genius - Revision 2004	6	0.37
Bonsai Test	4	0.33
Verbal section of Test For Genius Revision 2004	6	0.22
Logima Strictica 36	4	0.21
Qoymans Multiple-Choice #4	4	0.06

Weighted average of correlations: 0.659

Estimated minimum g loading: 0.81

Ranking in above table is based on the unrounded correlations.

National averages

Country	# scores	Average score
Netherlands	2	6.8
United_States	3	2.7

Correlation with personal details

Personalia	#	cor
PSIA Deviance	5	0.76
Disorders (own)	6	0.23
Disorders (parents and siblings)	6	0.05
Year of birth	9	-0.62
PSIA Ethics	5	-0.63
Father's eduational level	6	-0.68
Mother's educational level	6	-0.69
Educational level	6	-0.83

Correlation with national IQs

Correlation of this test with national average IQs published by Lynn and Vanhanen:

r = 0.31
n = 10

Scores by age

Age group	# testees	Average score
45 to 49	1	3
40 to 44	2	3
35 to 39	1	10.5
30 to 34	2	3.5
25 to 29	4	1.6

Scores by year taken

Year taken	# testees	Average score
1997	1	0
1998	1	10.5
2001	1	6
2002	2	2.5
2005	5	2.3

Correlation year taken vs average score = -0.21
n = 10

Additional statistics October 2008

Paul Cooijmans

Scores

# Candidates: 20
Median: 3.0
Quartile deviation: 2.6
Range: 16
Maximum possible: 22
Mean: 4.5
Standard deviation: 5.20
Male median: 3.0
Female median: 0.0 (# females: 1)
Hardness: 0.86
Resolution: 0.32

0	******
2	***
3	***
4	**
4.5	*
6	*
10.5	*
14	*
16	**

Correlation with other tests

Test	#	cor
Cooijmans Intelligence Test - Form 2	6	0.98
Test of the Beheaded Man	4	0.96
Cooijmans Intelligence Test - Form 1	4	0.93
Short Test For Genius	4	0.87
Association subtest of Long Test For Genius	6	0.86
Long Test For Genius	6	0.86
Space, Time and Hyperspace	7	0.84
Test of Shock and Awe	5	0.83
Lieshout International Mesospheric Intelligence Test	6	0.82
Culture Fair Numerical Spatial Examination - Final version	4	0.78
Numbers	5	0.75
Analogies of Long Test For Genius	6	0.69
The Final Test	11	0.65
Genius Association Test	4	0.64
The Test To End All Tests	11	0.63
Culture Fair Numerical Spatial Examination - C	4	0.60
The Sargasso Test	4	0.59
Isis Test	7	0.55
Reason	4	0.50
Qoymans Multiple-Choice #4	6	0.49
Test For Genius - Revision 2004	8	0.43
Verbal section of Test For Genius - Revision 2004	8	0.41
Spatial section of Test For Genius - Revision 2004	8	0.41
Reason Behind Multiple-Choice	4	0.40
Bonsai Test	4	0.33
Cartoons of Shock	7	0.29
Logima Strictica 36	5	0.21
Scholastic Aptitude Test (old)	4	0.05

Weighted average of correlations: 0.614

Conservatively estimated minimum g loading: 0.78

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 4 score pairs. All known pairs are used to obtain the true, honest statistics; Correlations are not artificially inflated by leaving out ceiling scores, outliers or other anomalies.

National medians

Country	# Scores	Median score
Netherlands	3	3.0
United_States	6	3.0
Finland	3	2.0

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 personal details

Personalia	#	cor
Cattell Culture Fair	3	0.97
Observed associative horizon is wide	3	0.90
PSIA Cruel	7	0.67
Observed behaviour is good	7	0.58
PSIA Antisocial	7	0.53
Is resident of the European Union	7	0.40
PSIA Cold	7	0.38
Educational level	15	0.33
PSIA Rational	7	0.30
PSIA Just	7	0.23
PSIA Aspergoid	7	0.20
Is male	20	0.20
PSIA Rare	7	0.20
PSIA Neurotic	7	0.10
PSIA Introverted	7	0.08
PSIA Deviance factor	9	0.06
Number of characters in candidate's name as registered	20	0.03
Mother's educational level	15	-0.04
Year of birth	20	-0.15
Disorders (parents and siblings)	14	-0.21
Father's educational level	15	-0.21
Disorders (own)	14	-0.26
PSIA Extreme	7	-0.39
PSIA True	7	-0.55
PSIA Orderly	7	-0.56
PSIA Ethics factor	9	-0.61

Correlation with national I.Q.s

Correlation of this test with national average I.Q.s published by Lynn and Vanhanen:

r = 0.15
n = 20

Reliability

Split-half = 0.91
Cronbach's alpha = 0.93

Error

Standard error = 1.5 raw score points
Quartile error = 0.8 raw score points
Range of error = 4.7 raw score points

Scores by age

Age group	# Scores	Median score
45 to 49	2	3.5
40 to 44	3	6.0
35 to 39	4	6.3
30 to 34	2	3.5
25 to 29	5	2.0
22 to 24	2	0.0
20 or 21	1	2.0
16	1	3.0

Scores by year taken

Year taken	# Candidates	Median score
1997	1	0.0
1998	1	10.5
2001	1	6.0
2002	2	2.5
2005	5	3.0
2006	2	0.0
2007	5	14.0
2008	3	3.0

Correlation year taken versus median score = 0.07
n = 20

Robustness and overall test quality

Rob = 0.78
Quality = 0.78

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 (validity of the item in this test) 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.