Hi everyone,
I want to report results obtained from a number of measurements with the average value plus minus the standard deviation.
During my studies I was always thaught to use only one signifying digit in the error term and to adapt your average value in such way that it has the same precision as the error term (for example 5.4576 +- 0.0457 becomes 5.46 +- 0.05).
The reasoning behind this is that if you are already uncertain about the first signifying digit in the error term, it doesn't make a lot of sense to be more precice on this term. I tend to agree with this logic, yet in many many scientific papers this rule is not applied and I don't seem to find some kind of guidelines how to cope with these cases.
Are there guidelines or what are your opinions?
I think it depends whom are you presenting the results to.
The results have to support a message and the SD represent the error of your mean of the population you are describing. SD and p values in my opinion are useless, but a 95% confidence interval says to your audience that your data says that the actual value of the thing you measure is with 95% confidence within this band if the data is normally distributed. Otherwise the data is normally distributed, you must present either interquartile ranges or 5/95 percentile instead
However, if you are going to submit to a journal or for some other academic purpose, you must follow the guidelines in order to be published or get your academic task.
- Badges
- Science topic
- Similar topics
- Biological Science
- Microbiology
- Virology
- Virus