The difficulty about repeating educational research is that one has to control for the variability that is inherent in the subjects/respondents. On also has to have strict controls to eliminate the variability introduced by time, because each year the cohort of subjects/respondents is different. International replicability is also problematic because of cultural and curricular differences. As a consequence we must not expect the same tolerances on our educational measurements as we would expect in research on objects.
The difficulty about repeating educational research is that one has to control for the variability that is inherent in the subjects/respondents. On also has to have strict controls to eliminate the variability introduced by time, because each year the cohort of subjects/respondents is different. International replicability is also problematic because of cultural and curricular differences. As a consequence we must not expect the same tolerances on our educational measurements as we would expect in research on objects.
For research trying to establish or negate causal relationships, replication is important to ascertain external validity and generalizability (Hill's criteria)...
Since individual variations are more prominent in education research as suggested by Ian, it is worthwhile exploring applicability of a certain intervention, tool or phenomenon in a different cultural environment or a system of education...
but this has to be very well thought out, and the replication has to be justified, not just for the sake of repeating an observation...
The inability to replicate the results of scientific studies is a BIG problem. See the attached news story.
In some cases, particularly in social science research, there are uncontrolled variables not accounted for in the research design.
In other cases, the problem is that the methodology is so poorly described in the published paper that it cannot be replicated with accuracy.
And of course in some cases, the data are manipulated and the conclusions are essentially falsified, in which case of course it would be impossible to replicate by a legitimate scientist.
Thanks Michael. Your opinion was useful in order to identify some relevant weaknesses for replicating a research not only due to aspects like culture or political laws which regulate scientific projects even owing to the fact of manipulation in results and a poor description of methodology.