It's probably just for data collection purposes. Once they have enough data, they'll run it through SPSS and amend the test to make it valid. Or not. Not sure how reliable the test is, either. A reliable test on the same object (same person here) should yield about the same results every time the object is subjected to the test.
One major problem that I can see right off the bat with this test is people checking words they are familiar with, but whose meaning they don't necessarily know. Plus, I'm sure there will be many people checking off words randomly just to obtain a higher score thus rendering the test invalid.
Bednar took the test twice and obtained significantly different results. Granted, he didn't follow instructions :)
Given that we only have a silly email address for contact information, I doubt that this is a serious research project. So, use it responsibly: for fun only.