Statistics of Bonsai Test

Remark: This test is no longer used in its own right but part of "Cartoons of Shock". These statistics are from the period before the test was incorporated therein.

Norms

Raw	Protonorm
0	207 or lower
1	242
2	272
3	297
4	320
5	327
6	340
7	360
8	367
9	380
10	387
11	433
12	487
13	544
14	569
15	594
16	619
17	649
18	684

Quality of norms: 0.96

Used for norming are all of the known scores from tests correlating 0.51 or higher with this test. The used scores are if needed converted to protonorms using the relevant conversion formula or table. The raw scores on this test that form pairs with the used scores are rank-equated to the used scores, and where needed there is interpolation or extrapolation.

Scores

# candidates: 25
Median score: 10.0
Quartile deviation: 2.0
Maximum possible score: 18
Mean: 8.8
Standard deviation: 2.81
Male mean: 9.0
Female mean: 3.0 (1 persons)
Hardness: 0.51
Resolution: 0.12

3	**
5	**
6	**
7	*
7.5	*
8	***
9	*
10	****
11	*****
12	***
13	*

Correlation with other tests

Test	#	cor
Analogies of Long Test For Genius	8	0.92
Strict Logic Sequences Exam I	5	0.90
Cartoons of Shock	10	0.79
Long Test For Genius	8	0.79
Lieshout International Mesospheric Intelligence Test	7	0.78
Association subtest of Long Test For Genius	8	0.67
KIT Intelligence Test	6	0.67
Reason	5	0.63
Qoymans Multiple-Choice #1	7	0.57
The Final Test	5	0.54
Test of Shock and Awe	9	0.54
Space, Time and Hyperspace	14	0.53
Spatial section of Test For Genius - Revision 2004	8	0.51
Qoymans Multiple-Choice #3	9	0.45
Logima Strictica 36	6	0.45
Test For Genius - Revision 2004	8	0.43
Qoymans Multiple-Choice #4	8	0.40
Odds	6	0.36
Cooijmans Intelligence Test - Form 1	10	0.36
Short Test For Genius	5	0.22
Verbal section of Test For Genius - Revision 2004	8	0.22
Numbers	13	0.17
Genius Association Test	7	0.15
Sigma Test	6	0.08
The Test To End All Tests	8	-0.03
Cooijmans Intelligence Test - Form 2	6	-0.28
Mysterium Entrance Exam	5	-0.44
NVCPE	5	-0.63
Spatial Insight Test	5	-0.74
Unknown tests	5	-0.78

Weighted average of correlations: 0.354

Conservatively estimated minimum g loading: 0.59

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 to obtain the true, honest statistics; correlations are not artificially inflated by leaving out ceiling scores, outliers or other anomalies.

National averages

Country	# scores	Average score
Finland	2	9.8
United_Kingdom	3	9.3
United_States	5	7.4

For reasons of privacy, only countries with 2 or more candidates are included in this table. Ranking is based on the unrounded averages.

Correlation with personal details

Personalia	#	cor
Observed behaviour is good	6	0.88
PSIA Rational	6	0.54
Is male	25	0.42
PSIA Orderly	6	0.36
PSIA Extreme	6	0.29
Mother's educational level	17	0.26
Year of birth	22	0.25
Disorders (parents and siblings)	17	0.24
PSIA Cruel	6	0.24
Number of characters in candidate's name as registered	25	0.19
PSIA System factor (phased out)	8	0.04
PSIA Cold	6	0.03
Disorders (own)	17	-0.02
Educational level	17	-0.03
PSIA Ethics factor	13	-0.08
PSIA Introverted	6	-0.15
Father's educational level	17	-0.17
PSIA Deviance factor	13	-0.22
PSIA Rare	6	-0.29
PSIA Antisocial	6	-0.31
PSIA Neurotic	6	-0.34
PSIA Aspergoid	6	-0.51
PSIA True	6	-0.64
PSIA Just	6	-0.78

Correlation with national IQs

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

r = 0.04
n = 22

Reliability and standard error

Using the "split-half" method:

Rel = 0.71
SE = 1.5 raw score points

Scores by age

Age group	# Candidates	Average score
55 to 59	1	7.5
50 to 54	3	4.3
45 to 49	2	10.0
40 to 44	3	9.7
35 to 39	3	10.7
30 to 34	3	9.0
25 to 29	5	9.8
22 to 24	1	12.0
20 or 21	1	6.0
18 or 19	1	9.0
17	1	7.0

Scores by year taken

Year taken	# Candidates	Average score
2002	7	9.7
2003	4	7.0
2004	8	7.5
2005	4	11.3
2006	2	9.3

Correlation year taken vs average score = 0.30
n = 25

Robustness and overall test quality

Rob = 0.93
Quality = 0.70

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.

In this period, one problem remains unsolved while all of the other correlate positively with total score.