Ajit gave good advice. I would like to make explicit, as I shall discuss further, that statistical analysis may have to account for the schools that dropped out.
The schools that dropped out present one issue and their replacement another. The former issue is how school dropout and girl dropout, the outcome, may be associated with each other in the population of schools. Unless there is good reason to judge any such association to be negligible, statistical analysis must allow for it. And, since non-negligible association may have affected assignment of replacement schools to intervention, it might not be practicable to include them in the analysis.
The latter issue is how knowledge of the data may have affected school replacement. The consequences are mostly parallel to those of association between school dropout and the outcome. The main difference is that such an effect on school replacement would not affect analysis based on only the 48 schools originally selected, without the replacement schools. But, unless there is good reason to judge effect on school replacement to be negligible, statistical analysis based on all 50 selected schools would have to allow for such an effect. And it might not be practicable to do so.
Finally, if it is unknown whether both association between school dropout and outcome and dependence of school replacement on knowledge of the data are negligible, secondary analysis under an assumption of negligibility might be appended.
------------------------------
Thomas M. Davis
------------------------------
Original Message:
Sent: 07-05-2019 11:20
From: Ajit Thakur
Subject: Randomised controlled trials
As Dr. Leroy Thacker has mentioned, it is indeed a nested design where the schools are randomized and the students are nested within the school. Your problem is the drop-out (two schools being dropped out and then replaced with two different schools). For regulatory purposes, it is better to evaluate the data without the two replacement schools so that your randomization scheme is not affected. Most programs for mixed model analyses now-a-days can handle such situations without any significant impact. Hope this gives you some idea. Good luck.
Ajit K. Thakur, Ph.D.
Retired Statistician
------------------------------
Ajit K. Thakur, Ph.D.
Retired Statistician
Original Message:
Sent: 07-04-2019 11:37
From: Leroy Thacker
Subject: Randomised controlled trials
This sounds like a randomized design, but I believe it is a cluster randomization. That is the schools are randomized and the subjects are nested within the schools.
------------------------------
Leroy Thacker
Associate Professor
Virginia Commonwealth University
Original Message:
Sent: 07-03-2019 02:37
From: Shongkour Roy
Subject: Randomised controlled trials
Dear All,
I am working on school girl drop out project and this study was a RCT design. The total number of school was 54, from these school we randomly selected 48 school among them 24 for intervention and 24 for comparison areas. After initial assessment we found two schools drop out in intervention areas due to environmental disaster and we replaced two schools from rest of 6 schools.
I want to know that whether this study design existed as RCT ? or could we tell it quasi-experimental design?
If anyone give explanation/ provide methodological documents that would grateful.
Thank you.
------------------------------
[Shongkour] [Roy]
[Assistant Program Officer]
[Population Council]
------------------------------