I'm taking a psychology class online. We had to do an experimental design, I did mine at the last minute and this is it.
The experiment I am designing is a variation of the conformity experiment testing to see if autistic adults are less likely, more likely, or approximately equally likely to conform their answers to match those of peers rather than trusting their own abilities. The independent variable is if the subject is on the autism spectrum or not (language note: There is not one agreed-upon term for autistic people, with preferences including person with autism, person with Aspergers, person with ASD, Aspie, Aspergian, Autie, A/autistic person, on the spectrum, and in some cases, Autistic as a noun. Autistic person will be used here.) The dependent variables will be likelihood of conforming to the incorrect answer 1) at least once over the course of the experiment and 2) on any given trial, with the groups divided into typical and autistic subjects.
The subject would be given a test alongside other "subjects," who are actually part of the experiment. The test would involve being given one line segment and a set of four line segments, one of which is the same length as the single segment. The subject would be asked to write their answer down prior to hearing the answers given by the fellow "subjects" so that it can be determined if the subject had the initial correct answer or not, after which the other subjects would begin to give their answer verbally. After this, the subject is asked to provide their answer, also verbally (this experiment excludes non-speaking autistic people due to this design constraint. A method allowing the person to type their answer in on an AAC device or choose a communication card to hand to the experimenter after hearing the other "subjects" answer may be usable, but it is uncertain and this will not be included in this experimental design.) Both initially written answers and verbal answers given to the room at large are recorded.
An initial incorrect answer with second answer correct and in alignment with the room, an initial incorrect answer with second answer differently incorrect in alignment with the room, and an initial correct answer with second answer incorrect in alignment with the room will all be recorded as conforming, unchanged answers where both are in alignment with the room (initially incorrect and second answer is the same incorrect answer in alignment in the room and correct answers for both when the other "subjects" give correct answers) and changed answers where both the initial and final answers are different from the room are neither conforming nor nonconforming, and any case not already covered where the final answer is different from that of the room is considered nonconforming.
This experiment would determine if autistic people are more, less, or equally likely to conform as typical peers, with it being necessary to match subjects so that gender and age do not affect the results. The experiment would not determine the reason for any differences which may (or may not) be found, only if there is such a difference.
This experiment design was inspired by the Asch conformity experiment and by the recent study "Children with autism do not overimitate" found in Current Biology and reported in Science Daily.