By Julian - 2/2/2016
Hello,
For my experiment, I edited two copies of the Sexuality IAT script. I edited one copy of the Sexuality IAT script to only display images of straight couples and gay men couples, and used this as a gay men IAT. I edited the other copy of the Sexuality IAT script to only display images of straight couples and lesbian women couples, and used this as a lesbian women IAT. I administered both IATs, counterbalanced for order, to all of my participants.
Having collected all of my data, I am now looking at combining participants from both IATs collected from each participant to reflect their attitudes towards both gay men and lesbian women as one D_biep score. Could I do this by just adding the lesbian women IAT d_biep score and the gay men IAT d_biep score and dividing it by two? I do not see any reason why this would be methodologically inaccurate, but I would very much appreciate a second opinion on this. No one in my department is familiar with the IAT scoring procedure, and cannot advise me on this.
Thank you,
Julian
|
By Dave - 2/2/2016
First I would recommend reviewing the published literature re. whether anyone has ever done anything similar.
My personal gut feeling is this: Combining the scores *may* be valid if the underlying construct measured in the two IATs is the same (e.g. attitude towards members of the LGBTQ community). If it is not, I'm not sure how one could interpret such a combined score: Is someone's attitude towards homosexual men plus someone's attitude towards homosexual women divided by two *necessarily* equal to one's overall attitude towards gay people (of both sexes)? I'm not so sure.
|
|