I am currently doing a research proposal for one of my modules, investigating whether there are differences in attitudes towards death between hindu's and the non religious.
As I would be assessing intrinsic and extrinsic orientations against the control group of non-religious individuals, would intrinsic and extrinsic be separate experimental groups?
Please help
Thank you.
Yes but my advice is to design a real experiment. You can always relax requirements later if needed. Best, David Booth
I could not see an experiment here. It appears an analytical study where you are going to assess attitudinal differences between two distinct groups. How you are going to sample individuals into your study groups is a different question altogether. If there is an intervention you are administering as a part of your study, please elaborate on that.
If you can't see an experiment how can you see a quasi experiment. I am always told by my colleagues that a quasi experiment is what you do if you can't randomize. Well why then can't you design a randomized experiment and just ignore the randomization. Obviously I am not an expert in quasi designs so I defer to those are. Best, David Booth.
For me, there as to be some kind of experimental intervention for a research project to be either a true experiment (with random assignment to groups) or a quasi-experiment (using naturally occurring groups). All I see here are two independent variables and no intervention.
David L Morgan David Eugene Booth Viswa Chandu
I was advised by my tutor to use a quasi-experimental design. The study is hypothetical, meaning I will be marked solely on the research proposal and the study will not be carried out. If I were to continue using a quasi-experimental design, could an 'intervention' be something along the lines of images that signify death?
Not sure if this makes total sense, apologies I'm an amateur at these types of experiments.
David Eugene Booth , I'm curious if you believe your IRB would allow people to be randomly allocated into a Hindu or non-religious group? I would say this is an example where calling something a quasi-experiment is a little silly since it is not like the non-religious are really any sort of control group.
If this is just for design purposes, then showing images that signify death to half the sample would be an experimental design -- although might also raise IRB issues.
Daniel Wright
I can see your point, although regarding my proposal as 'silly' isn't actually a positive contribution to my question. Please treat me with some respect, I am not a world-renowned scientist I am a student.
Meg lucy Cardwell , I wasn't saying your proposal is silly. I was saying the phrase quasi-experiment is over-used. Read the comment: "calling something a quasi-experiment is a little silly."
@Meg lucy Cardwell please look into ex post facto designs. That could be relevant in your context. Little detail is provided below.
An ex post facto research design is a method in which groups with qualities that already exist are compared on some dependent variable. ... Although differing groups are analyzed and compared in regards to independent and dependent variables it is not a true experiment because it lacks random assignment.
@Danial wright. I didn't get your point. I thought I was proposing exactly what you said. Design your experiment and just don't randomly assign. Isn't that a quasi design?
As I said I am pretty much a Fisherian when it comes to experimental design. Best, David Booth.
David Eugene Booth , I guess I was confused by the ignore the randomization part. Are you proposing just acting as if randomization was used?
- Badges
- Science topic
- Similar topics
- Social Science
- Media
- Journalism