It is hard to select research design approach being followed in my study. Since I did not randomly select(draw) sample from a population but I randomly assigned the 100 students equally to the control and treatment groups. Is it still random sampling? As random sampling means the selection of sample from population. I did this the other way around. Since it was convenience for me to work with a certain teacher so I took her classes for further experimentation. As far as, random sampling or selection is concerned, I can not use Quasi Exp design instead I can employ Experimental Design. Since the research was conducted in natural setting(school) and the independent variables(various instruction types ; conventional and an edu app instructions) were isolated, controlled and then manipulated which are against the Experimental Design and supports only the Quasi Exp Design. So what sampling technique should I mention in methodology section? Since I have employed Quasi Exp research Framework where I did Random assignment and I worked with experimental and control group. This is a bit confusing.
Citations:
Campbell, D.T., Stanley, J.C. (1966). Experimental and Quasi-Experimental Designs for Research. Skokie, Il: Rand McNally.
Cohen and Manion's Book
"Random Selection & Assignment", retrieved from
http://www.socialresearchmethods.net/kb/random.php
Dear Amna: it is confusing, indeed. In a few words, which is your main hypothesis?.
I used Quasi Exp research design as it best fits in school setting and students from two classes of 8th grade were randomly assigned to the control and treatment groups. Please note that girls' and boys' classes were segregated. I assigned half of the girls and half of the boys to the treatment group. and other halves to the control group. Now Does this pose a problem? Random assignment and sampling employed in quasi experimental approach?
I understand as "quasi experimental" instances when a given intervention -mostly those you cannot use experimentally- affects a certain group of persons accidentally or unintentionally (regarding the resulting effects). That would be the classic Loperamide case. In such a situation, "randomization" does not exist, although a very small probability that both unintentional and randomization could occur. Besides, randomization is usually made following random numbers generated by a software program. I do not know how you made it. Nevertheless, any answer could fall wrong without knowing what you want to prove (or expect as result of your eventual analysis).
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis