Scribd
Upload
791 views6 pages

Scores of High Range Tests

This document provides statistics on test performance of high-scoring individuals on standard psychometric tests and tests created by Dr. Xavier Jouve. It analyzes scores on (1) standard tests, (2) Dr. Jouve's tests, and (3) both, finding mean scores around 138-148 and a normal distribution of scores. It also shows high correlations between theoretical IQs from hriqtests.com and scores on standard tests and Dr. Jouve's tests, indicating the quality of tests on that site.

Uploaded by

Remik Buczek
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
791 views6 pages

Scores of High Range Tests

This document provides statistics on test performance of high-scoring individuals on standard psychometric tests and tests created by Dr. Xavier Jouve. It analyzes scores on (1) standard tests, (2) Dr. Jouve's tests, and (3) both, finding mean scores around 138-148 and a normal distribution of scores. It also shows high correlations between theoretical IQs from hriqtests.com and scores on standard tests and Dr. Jouve's tests, indicating the quality of tests on that site.

Uploaded by

Remik Buczek
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Scores of high range tests’ candidates on standard psychometric batteries

& tests by Dr. Jouve.

hriqtests.com, July 2019

Aim of this document is to provide information regarding the “statistical profile” of


high range tests’ candidates, according to their reported performance on standard
psychometric batteries & tests created by Dr. Xavier Jouve.
Although consisting of mostly simple, descriptive, statistics, it’s more than valuable in
drawing certain conclusions regarding mean (and expected) performance on - and
norming of - high range tests, as some of them may suffer from inflated/deflated
scores and insufficient statistical processing. Plain language is used, in order this
report to be approachable to anyone interested, regardless special knowledge on
statistics.
At this point, it has to be mentioned that tests of Dr. X. Jouve were chosen due to the
fact that they have been normed using quite large testing samples (most of them on
N>300, consisting mainly of high achieving individuals with mean score >131,
M=100, SD=15 on standard psychometric batteries).
In addition, one may notice that both r (Pearson correlation coefficient) and ρ
(Spearman’s rho, Spearman’s rank correlation coefficient) are presented at Part 5 of
this report. Although there are certain indications that scores of high-range tests’
candidates distribute almost “normally”, a non-parametric measure as the
Spearman’s Rho (ρ) will be more reliable until it’s proven otherwise.
Histograms, some parameters and further discussion, follow.
In addition, an overall correlation of theoretical IQs provided by tests on
hriqtests.com with standard psychometric batteries and tests by X. Jouve is provided.

Tests used (Scores’ N=350) :


A) Standard psychometric batteries (alphabetically): BADyG-M (2), BLS4-2T (1),
CCFIT III (17), CFT20-R (2), FRT-A (12), FRT-B (9), GAMA (1), IBF-S (3), IDF (1),
IST 2000R (2), IST 70 (3), MAT (10), NGCT (1), OLSAT (2), PCAT (1), RAIT (1),
RAPM (27), RIAS (1), RSPM (5), SBIS (5), TOGRA (1), WAIS (R,III,IV - 70), WISC
(9), WPT (4), Unknown Mensa Entrance Test (Probably RAPM or FRT - 21).

B) Tests by Dr. X. Jouve (alphabetically): C-09 (17), C-10 (14), C-12 (1), CCAT (2),
CSE (abbreviated NVCPE-R form, granted after C-09 to high-scoring contestants - 3),
IAW (2), JCCES (9), JCTI (23), NVCPE (3), NVCPE-R (59), TLAP-R (3), TRI (2).

Parts
1. Scores on standard psychometric batteries (N=212).
2. Scores on standard psychometric batteries (N=172, low-ceiling & extrapolated
scores excluded).
3. Scores on Dr. Jouve’s tests (N=138).
4. Scores on standard psychometric batteries & Dr. Jouve’s tests (N=310).
5. Overall correlation of hriqtests.com tests with standard psychometric batteries &
Dr. Jouve’s tests.

(Proceed to next page)


1. Scores on standard psychometric batteries.

All reported scores are included (fixed scores, low-ceiling scores (scores >x are
presented as x), extrapolated scores). Lowest score is 85 sd15 (1, WAIS) and highest
scores are 185 sd15 (2 extrapolated, WAIS & SBIS).

Mean Score : 140,34.


Median : 137,5.
Standard Deviation : 15,23.

As a lot of extrapolated and ceiling scores are included (N=40), no significant


conclusions can be made.

(Proceed to next page)


2. Scores on standard psychometric batteries (low-ceiling & extrapolated
scores excluded).

Lowest score is 85 sd15 (1, WAIS) and highest is 182 sd15 (1, CCFIT III).

Mean Score : 138,63.


Median : 137,5.
Standard Deviation : 14,04.

Distribution of scores is quite close to normal. A possible anomaly is easily noticed at


131-140 range and there are certain reasons for this. According to my experience and
in my humble opinion, it’s indeed an anomaly occurring at the aforementioned range
and we are not talking about a distribution different than the normal one; there is a
certain reason for this : A lot of assessment tools (like FRTs) provide scores up to that
level or close to that level. So, there might be either not that reliable results in case of
people reaching a test’s ceiling or some ceiling scores may have unwittingly been
reported as fixed ones.
As one may notice in one of the next histograms, in which scores on standard
psychometric batteries are combined with the ones on tests by Dr. Jouve, such an
anomaly is not present. And do notice that Dr. Jouve’s tests have significantly higher
ceilings than 131-140.

(Proceed to next page)


3. Scores on tests by Dr. Jouve

Lowest score is 120 sd15 (1, JCTI) and highest 198 sd15 (1, NVCPE-R, Rasch-analysis
equated. Fluid Intelligence Index was reported as 181 sd15). That was the highest
score ever reported, until quite recently, when a score of 204 sd15 was announced.

Mean score : 148,02.


Median : 147.
Standard Deviation : 11,45.

(Proceed to next page)


4. Scores on standard psychometric batteries & tests by Dr. Jouve.

Lowest score is 85 sd15 (1, WAIS) and highest 198 sd15 (1, NVCPE-R).

Mean score : 142,81.


Median : 142.
Standard Deviation : 13,75.

(Proceed to next page)


5. Overall correlation of hriqtests.com tests with standard psychometric
batteries & Dr. Jouve’s tests.

Correlations between theoretical IQs provided (until late July, 2019) on


hriqtests.com’s tests (CPE-N, VRA50, The Phi Test, SEE30) and : a) Standard
psychometric batteries (Pairs’ N=99, low-ceiling and extrapolated scores excluded), b)
Dr. Jouve’s tests (Pairs’ N=125), c) Combination of the aforementioned (N=224), are
provided.

r: Pearson correlation coefficient (Pearson’s r).


ρ: Spearman’s rank correlation coefficient (Spearman’s rho)
MIQ : Mean IQ on the mentioned test(s) (SD=15).
MIQ-HR : Mean theoretical IQ provided by tests of hriqtests.com (SD=15).

Test Pairs (IQ range) r ρ MIQ MIQ-HR P-Value


Total 224 (85-172) 0.73 0.77 140.72 141.03 <0.001
Supervised 99 (85-172) 0.74 0.79 136.82 137.53 <0.001
Jouve 125 (120-170) 0.71 0.73 143.8 143.81 <0.001
WAIS 32 (85-164) 0.86 0.84 135.66 135.38 <0.001
RAPM 20 (111-147) 0.52 0.67 133.6 136.15 0.0012
JCTI 34 (123-160) 0.69 0.67 143.29 144.91 <0.001
C-09 31 (120-153) 0.8 0.82 140.18 142.84 <0.001

Correlation of 0.7+ with a significant amount of scores on standard psychometric


batteries and especially that of 0.8+ with WAIS, that is the most used and up-to-date
intelligence assessment tool, shows that the work done in the past (tests’ revisions,
certain kind of items selection according to discrimination indexes etc.), was of
utmost importance and had to be done. New generation of tests (CPE-N, SEE30,
VRA50, The Phi Test), as well as the currently running contests (INSC & IVAC) have
proven themselves as ones of high quality.
It’s also proven that common work is needed, between authors and tests’ candidates.
Tests’ authors should thoroughly examine their tests statistically and DEFINITELY
revise them if needed (it is, until they reach some certain level of quality) and
candidates should report other scores (and, by this way, help in author’s statistical
“journey”).
To sum up, I will repeat something written at a former report: “Regarding test
designing, after 8 years of having created several of them, one thing can be said for
sure: There exist specific (and different, of several difficulty levels) patterns (in other
words, kinds of items) that tend to behave steadily and “healthy” – that is, keeping
their difficulty regardless the environment (whether they are part of one test or
another), showing high discrimination index. It’s only after 8 years of studying and
experimenting with tests that I can now say that I ‘ve created a team of decent tests.
And in this point, one should realize that “attractiveness” or “originality” of a test do
not necessarily guarantee its quality. Of course, combination of the aforementioned is
the ultimate goal.”

You might also like