- This topic has 5 replies, 3 voices, and was last updated 9 years, 1 month ago by Anonymous.
8th November 2011 at 10:41 am #2967
I am wondering if I conducted a study to evaluate an intervention that was NOT conducted by me, is it still an experimental study? Or it is just an evaluative study? For example, there is a laptop program at a secondary school that I want to focus on. The program was initiated by someone else a couple of years back. There are control groups available that I can compare the treatment group to. Is it an experimental study?
I will appreciate it if you can shed some lights on this.
Lisa8th November 2011 at 10:06 pm #2972AnonymousInactive
can you randomly allocate people in the control and treatment groups? or is it that a certain secondary school was allocated already to that lap-top program by whoever did this study a couple of years back, and now you want to compare those students to ones who are similar but who did not participate in the same program?9th November 2011 at 1:10 am #2971
You are right. I want to compare the the classes that is already in the laptop program with those that are not. I can’t randomly allocate students in the control and treatment groups. So that makes it a quasi-experiment or what?9th November 2011 at 5:56 am #2970AnonymousInactive
you are most certainly right. if you cannot randomise across conditions then it cannot be a true experiment (at least not in the sense they are referred to in statistics).4th December 2011 at 12:58 pm #2969Ingo RohlfingMember
Well, some would say that it is not a classic field experiment because you did not assign the treatment (the program) to pupils. However, your analysis qualifies as a natural experiment if you can argue that you are dealing with as-if randomization. Someone else installed the programs, but if the reason why some pupils can and cannot use the program is unrelated to this, one could take it as an instance of as-if randomization. As-if randomization is given when, for example, pupils with higher grades can work with the program and others not. Finally, if you can argue that the pupils who can and cannot use the program are similar in all respects that my be related to your research interest, you might have a quasi-experiment at hand. In particular, it might qualify as a differences-in-differences design or an interrupted time-series design (see Cook, Thomas D. and Vivian C. Wong (2008): Better Quasi-Experimental Practice. Alasuutari, Pertti, Leonard Bickman and Julia Brannen (eds.): The Sage Handbook of Social Research Methods. Thousand Oaks: Sage: 134-165.)5th December 2011 at 12:48 am #2968
Thanks a lot for the reply. This is really helpful!! I will definitely check out the book you recommend.
- You must be logged in to reply to this topic.