I am a young researcher studying the depolymerization of plastics. I still do not have great insights about how to identify great research.
I often read highly cited research in elite journals such as Nature or Science. There, I see that many chemical reactions are also reported with very few number of replications (a lot of times, just one!). How is it that this research becomes high impact of it is not sufficiently reproduced for a large number of samples?
There are two ways of looking at sample size, and they are related. Your question is related to the number of repeated measurements that are analyzed to provide a final result. The importance of this was beaten into us in quantitative analysis courses in college. We had to provide a statistical analysis of the results to justify their use. One measurment is just a rough guess subject to nany potential problems. Two measurents in agreement indicates consistancy. Three in agreement provides confimation - and that could be that you made the same error consistantly.... and I need to continue
Continuing.. the second aspect of sample size is the amount of material ussd in the analysis. Too small and you may not have a representative sample that could be adversely affected by a lack of homogeneity. This would show up in a wide variation in results in replicate analysis. Both aspects of sample can be contributing factors to the problem of inter- and intra-lab reproducibility.
Sorry about the two part answer - the app on the phone limited text lenght.
- Badges
- Science topic
- Similar topics
- Chemistry