If you get unexpected results in any research, is it legal/ethical to repeat data collection or to increase the size of the sample? your opinions are highly appreciated.
There is nothing illegal about collecting additional data. It is very likely that based on the original sample size a level of significance in statistical testing was not achieved. Thus collecting more data under the same conditions and combining the two data sets into a larger data set is perfectly valid which can later be analyzed. As always follow the rigorous principles of data collection including randomness, no systematic bias, etc. and being aware at all times of reliability and the internal and external threats to data validity.
Replicating research is always a good thing, and especially when there are reasons to doubt the initial findings. Maybe what you are asking is whether you can keep collecting data until you find the results you want. It would be unethical to perform 5 studies, and only report on the fifth because that is the one that shows what you want. But if you report on all five or combine the data together from all five, then the repetition of the research should increase the reliability of your findings.
Bradford Chaney
thanks for your valuable comments, what I'm talking about when you find extreme findings in research, not collecting until finding the results I want.
Particularly when the findings are extreme, replication is worthwhile. It may also be good to check for errors in the data or analytic approach. It's all too easy to make up explanations to fit the data when there actually is a problem with the data.
Replication may be indicated in this case. It will increase the sample size.
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude