Take the 2-minute tour ×
Writers Stack Exchange is a question and answer site for authors, editors, reviewers, professional writers, and aspiring writers. It's 100% free, no registration required.

For my research I produced random normally distributed data. The data is necessary for a statistical analysis which I describe in my paper. Is it recommended to state in the paper what tool I used to create the data? I am not sure if so, since there are many ways how to produce these data. The output is always the same and in my case only the output (the produced data) matter.

share|improve this question
4  
One sentence to describe how you produced the random set is usually both necessary and sufficient. It is necessary precisely because of the many ways random data can be created. –  KitFox Nov 8 '13 at 0:22
add comment

1 Answer 1

up vote 4 down vote accepted

If you used a common method which is known to produce completely random and normally distributed data, just mention its name or describe it in one short sentence.

It is often that various methods of random data generation are found to possess some bias, pattern, lesser than maximum possible entropy etc. - their randomness is not perfect. This may affect results, and for that eventuality you should at least name the method, so that in case the method is found vulnerable, your research could be verified, whether it's affected by the vulnerability or not.

share|improve this answer
    
Exactly. Most common software methods don't produce true randomness at all. –  what Nov 11 '13 at 9:09
    
@what: although there are good ways of obtaining truly random data, e.g. lowest bits of values from temperature sensors, which are essentially white noise. –  SF. Nov 12 '13 at 8:22
    
Yes, maybe, but (a) most users don't know how to access them and (b) CPU temperature apears to be constant in many computers (see answer to unix.stackexchange.com/questions/71285/…). But there are web services that offer true randomness, though I have never read an article using such services. –  what Nov 12 '13 at 8:54
    
@what: If the temperature is cropped to 2 significant digits, yes. If it measures temperature to a thousandth of a degree, the lowest bits are really a white noise. –  SF. Nov 12 '13 at 9:03
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.