Millisecond Forums

SC-IAT Methodical question

By Nighthawk - 5/18/2010

Hello everybody,

I have programmed a single category IAT using the sample script from milisecond software page :

I have a methodical question concerning this Single-Category IAT which came to my mind after making the first data collection trials with subjects. Let me explain:

You know in this SC-IAT there only exist one target-Category. The Stimuli (if they are words) are presented in greencolour font if they belong to the e.g. positive or negative category. These categories (positive and negative) are also presented in green colour font.

Now, in the Normal IAT (with 2 target categories) the target categories are written in a white colour font and also the stimuli which belong in either of these target categories are also written in white just like positive or negative stimuli are written in green.

In the script it says:"The green and white labels and items may help to identify the appropriate category. Use the E and I keys to categorize items into three groups left and right, and correct errors by hitting the other key."

So in the normal IAT the different colours are helpful. But now in the SC-IAT you could make the categorisation of stimuli words from the target category work without reading the  actual word because you can categorise it just by its colour (white), due to the fact that one colour belongs only to one category (in this example white is always the target category.

In the Normal IAT that doesnt work because one colour always stands for 2 categories. But in the SC-IAT subjects apply this rule which isnt wanted by the experimenter of course...

Do you understand what I am trying to express? I hope there is an error in my reasoning otherwise my data collection is more or less useless.

Thanks very much...

By IvoR - 5/19/2010

Hey Nighthawk,

That's a good point your making. The best way to test if your theory holds, is to analyze your data. If you have a difference between participants, than apparently the color does not make a difference. If you don't find a difference, than either they looked at the color of the words, or apparently they associated the target with both a positive and a negative attitude. Either way, these data aren't what you're hoping for. 

But this is of course something you can only do afterwards when you've already got the data. I agree that using a different color for the word isn't such a good idea, as there is a possibility that people are reacting on color. Because the good and bad stimuli are in green, people have to read these words, which automatically makes them read the white words... but of course that is an assumption you make. So I agree that using just one color is better.

Did you come up with this afterwards or was it something you also ran into before the experiment? If so, why did you decide not to change it?



By Nighthawk - 5/19/2010

Thanks for the reply Ivor.

This SC-IAT is the very first one  I ever programmed and tested on subjects. Fortunately, it was rather a pretest of the IAT. This pretest consisted of a sample of ten people. If there hadnt been those problems I would have continued with the data collection but after the first day a subject told me that he just used the colour for the categorisation task. So to answer your question, I came up with this afterwards. But given the facts I mentioned above it is not that bad. I will continue my research question with the standard IAT because now I think it fits better.



By TAO HU - 5/22/2013

Hello nighthawk

I know you had programed a sample about it.If you do not mind,please send me your SC-IAT sample by email,I need a SC-IAT sample for a research.

My email is

Thanks a lot!

Ps:I'm a Chinese student.My English is not good.Please forgive the error in my reply.

If I have offended, please forgive me!