The research problem: Developing effective teaching procedures for teaching the "Wh-" questions as mands.
Hypothesis: Teaching children diagnosed with autism to mand (request) using "Wh-" questions will reduce socially inappropriate behavior.
Population/Participants: Children diagnosed with autism (Ages 3-12)
Selection of Participants: My theoretical population would be all students with autism. Given that here in Richmond we have a 'School for Autism' and I happen to know the 'Director of CABAS Research' there I would use them as my 'sample'. I would use a proportional stratified random sampling with the 'strata' being ages to insure I had equal representation across age levels. I would choose this method over random sampling cluster sampling to make sure I could demonstrate effects across all ages evenly. I would have half of the children in each grade be the 'control' group to demonstrate effects.
You have a good start, but you need to be more specific with your research problem statement. Your independent and dependent variables should be clear, and you should state what you are trying to investigate. If a particular intervention is your independent variable, make sure you clearly explain that intervention. If your dependent variable is socially inappropriate behavior, you need to be clear about the operational definition that you are using for this outcome. Look back at the slides from week 2 and see if you can revise your problem statement. Sometimes it takes two or three sentences to make the problem or questions clear. Your hypothesis gets closer to what we are looking for in a good testable, quantitative problem.
ReplyDeleteWith respect to sampling, you are on the right track. I think stratified sampling would be appropriate. Just be careful how you use the words “random”, “proportional”, and “equal representation.” I’m not clear from your description if you want an equal number of students in each group for your sample, or if you want a sample with age group proportions that are similar to the study population. You can say that you will use a stratified sampling method where the sampling frame is divided into strata by age group, and participants are selected randomly from each age group, so that participants are representative of the population in terms of age. You can also say that you will randomly assign participants to treatment and control groups within your strata. But we don’t refer to that as “stratified random sampling.”