Proportion of female high-range candidates outscored by I.Q.

© August 2008 Paul Cooijmans

Fewer females than males take high-range tests, and their median score is roughly 10 I.Q. points lower. Therefore it makes sense to provide the female score distribution with proportions outscored for informational purposes, especially to enable female candidates to learn how they did compared to other females.

To obtain a representative sample, the female scores of a large number of tests are combined, converted to I.Q. (S.D. = 15) using the current norms. The tests are: Test of the Beheaded Man, Cooijmans Intelligence Test (Forms 1 and 2), De Laatste Test, The Final Test, Genius Association Test, Lieshout International Mesospheric Intelligence Test, Numbers, Qoymans Multiple-Choice #4, The Test To End All Tests, Spatial Insight Test, "Space, Time and Hyperspace", A-10, Long Test For Genius (English, Netherlandic and French versions), Test of Shock and Awe, Short Test For Genius, Bonsai Test, Giga-test, and <Cooijmans On-Line Test>. The following statistics result:

I.Q. # Prop. outsc.
95 * .005
96   .006
97   .007
98   .008
99   .010
100   .011
101   .012
102   .014
103   .015
104   .016
105 * .017
106 * .027
107 * .038
108   .044
109 ** .055
110 ** .077
111   .088
112 ** .099
113 ** .121
114 * .137
115 * .148
116   .154
117 *** .170
118   .187
119 ** .198
120 * .214
121 **** .242
122 *** .289
123 **** .319
124 *** .357
125 * .379
126 *** .401
127   .417
128   .419
129 **** .440
130 * .476
131 *** .489
132 **** .527
133 * .555
134 ** .571
135 * .588
136 ** .604
137   .615
138 ***** .643
139   .669
140   .671
141 * .676
142 **** .703
143   .724
144   .726
145 ** .736
146   .154
147   .779
148   .781
149 **** .802
150 *** .841
151   .856
152   .858
153 ** .868
154 ** .890
155 * .907
156 ** .923
157 ** .945
158   .956
159 * .962
160   .967
161 * .973
162   .976
163   .977
164   .979
165   .980
166 ** .989
167   1
168   1
169   1

Below a histogram by 5-point classes to show the shape of the distribution better, revealing that the modal range is 120-124. This shape is roughly comparable to that of the sex-combined (but largely male) high-range distribution, but shifted downward by 10 points: An asymmetrical wedge with the mode about 5 to 10 points below the median.

I.Q. class# scores
95-99*
100-104 
105-109*****
110-114*******
115-119******
120-124***************
125-129********
130-134***********
135-139********
140-144*****
145-149*********
150-154*******
155-159******
160-164*
165-169**

Noteworthy is also the apparent "dip" at 140-144, above which it picks up again, reminding of the similar "dip" in the high 150s of the sex-combined (largely male) distribution. If two distributions are so similar but shifted by such distance, this makes one wonder whether it would be better to separate them entirely and have within-sex norms as the principal norms, instead of the current sex-combined norms that are to the great disadvantage of females. That would be similar to having chess tournaments for women and men separately. A drastical step that should not be taken overnight, but it might become more rewarding for females to take the tests then, especially if society admission would change to using within-sex norms as well.

The within-sex norms meant in the previous paragraph would be the "proportion outscored" norms, not the I.Q.s which should always remain the same for a given performance regardless of sex. It would then have to be decided on which population to base the I.Q. norms, and the logical choice would be to base them on the male score distribution (the only one that covers the full range).

The most recent "Proportion outscored" information within females can be found in the Protonorms table.