Have just finished a neat update to the site. On the correlations page you can now see the matrix of cross-test sample sizes:
You can click any of the green cells to look at the scatterplots. For instance, French vocabulary (0-10 part of the 15-language test) and US-centric general knowledge test:
You can also discover weird stuff, like maybe Greek speakers aren’t very irritable (self-report):
I am trying to encourage users to take more than 1 test, so that the data can be linked. Many of these scales have never been studied together before, so new analyses become possible this way.
Have fun!
Hi Emil, I think your website is very interesting! However I think a lot of the knowledge-based surveys are going to have skewed results because a large percentage of people taking them are likely to be cheating. As an example, the average scores of the Pew Science Test when actually conducted by a representative sample of the public (i.e. the results of the original study) are considerably lower than the average of online test-takers. Unless everyone who takes the test online is multiple standard-deviations more knowledgeable about science than a representative sample of the public, it's clear that people are cheating. My score on that quiz was about 97 percentile compared to the results found by Pew, but only 63 percentile compared to the online test-takers. Is it possible to control for this, perhaps by making the questions timed so people cannot Google the answers?
I liked the 8 axes political test, tho focus on spending failed to capture issues of fraud. It also seemed to make sense.
The multi-lingual test was long and didn’t seem to make much sense, so I stopped.
A comparison of politics with OCEAN Big 5, and MBTI would make more sense.