Harvard IAT is TRASH.

What the IAT is: The IAT measures associations between concepts (e.g., White Americans and Black Americans) and evaluations (e.g., Harmless Objects, Weapons). People are quicker to respond when items that are more closely related in their mind share the same button. For example, an implicit preference for White Americans relative to Black Americans means that you are faster to sort words when ‘White Americans’ and ‘Harmless Objects’ share a button relative to when ‘Black Americans’ and ‘Harmless Objects’ share a button.

My results: Your responses suggested a slight automatic association for Harmless Objects with White Americans and Weapons with Black Americans.

Before the test, I was advised that I may not enjoy the results of the test — they were correct. This test may be interesting, but nothing more. I would are that i would get different results if the test were flipped in the beginning, thus suggesting I would have an automatic association for harmless objects with Black Americans and weapons with White Americans. If anything, the test was built for someone to fail… perhaps with hopes of making you question yourself and build an internal desire to learn more.