I used the Data from ssc2018public.xlsx (as available at: http://slatestarcodex.com/2018/01/03/ssc-survey-results-2018/). The data is not clean and before further analysis can be conducted text responses and obvously false data was removed.
some statistical key characteristics of the data:
| Gender | Mean | Max | Min | Median | SD | n |
|---|---|---|---|---|---|---|
| F (cisgender) | 137.3588 | 177 | 105 | 135.0 | 11.812918 | 170 |
| F (transgender m -> f) | 136.9677 | 172 | 108 | 135.0 | 13.202232 | 31 |
| M (cisgender) | 138.2355 | 190 | 90 | 138.0 | 12.654659 | 1762 |
| M (transgender f -> m) | 143.2500 | 150 | 136 | 143.5 | 6.396614 | 4 |
| No Answer | 130.5714 | 141 | 120 | 130.0 | 8.638232 | 7 |
| Other | 139.2245 | 180 | 114 | 140.0 | 12.191639 | 49 |
A density plot visualizes the distribution for each gender:
In adddition a combined plot of density and histograms was produced seperate for each gender to vizualize how the data is distributed in more detail:
An additional histogram plot shows the number of subject in each bin instead of the density:
In order to increase the number of datapoints the IQ equivalent of the Sat scores was calculated using the following formulas: \[\begin{align}IQ_{sat1600}& =26.8+0.085\cdot S_{sat1600}\\ IQ_{sat2400}& =28.142 + 0.05327\cdot S_{sat2400}\\ IQ_{combined}& =mean(IQ_{selfReported}, IQ_{sat1600}, IQ_{sat2400})\\ \end{align}\\ \text{The mean function is configured to omit NA values (e.g. empty responses).}\]
| Gender | Mean | Max | Min | Median | SD | n |
|---|---|---|---|---|---|---|
| F (cisgender) | 146.1823 | 177.000 | 105.00000 | 148.4911 | 10.468892 | 422 |
| F (transgender m -> f) | 143.9140 | 165.275 | 120.00000 | 144.8250 | 10.797465 | 58 |
| M (cisgender) | 145.5830 | 180.025 | 87.42332 | 147.0000 | 11.144422 | 3928 |
| M (transgender f -> m) | 147.9019 | 161.950 | 136.00000 | 147.4111 | 7.939274 | 12 |
| No Answer | 137.2735 | 162.800 | 118.60000 | 136.1625 | 13.918666 | 12 |
| Other | 144.9718 | 167.575 | 114.00000 | 146.0125 | 10.857336 | 78 |
The SAT score might be more reliable than the self reported IQ therefore the same analyses as above were conducted on IQ scores derived from SAT scores without self reported IQ.
| Gender | Mean | Max | Min | Median | SD | n |
|---|---|---|---|---|---|---|
| F (cisgender) | 149.8776 | 162.80 | 111.8000 | 151.7284 | 8.084869 | 346 |
| F (transgender m -> f) | 148.5903 | 161.95 | 131.2025 | 148.2659 | 8.367304 | 44 |
| M (cisgender) | 149.6573 | 162.80 | 94.7295 | 150.9000 | 8.689050 | 3083 |
| M (transgender f -> m) | 149.5260 | 161.95 | 140.0090 | 150.6630 | 8.196164 | 9 |
| No Answer | 143.2394 | 162.80 | 118.6000 | 143.6750 | 13.724685 | 8 |
| Other | 151.1167 | 162.80 | 129.3550 | 152.7938 | 7.283427 | 51 |