Take the 2-minute tour ×
Academia Stack Exchange is a question and answer site for academics and those enrolled in higher education. It's 100% free, no registration required.

It recently occurred to me that probably more often than it should happen, while reading someones paper, they see a reference to a different paper, and just use the reference without looking. In my area, a common example is in referencing data, such as average human height, etc.

I came across this as I was looking for data on human dimensions from a somewhat smaller country that does not use English. I found 9 papers that all referenced the same data sheet. I then contacted the author of that data sheet and was informed that there is no online version (never has been), and the only print copy exists in that country. The author herself does not have this book. I'm guessing the only reason the other authors referenced the data was because the author of the referenced data wrote a paper that referenced her own data.

I am not interested in how to format these citations. I would like to know if it is appropriate to cite this information at all, or cite that someone else cited it. I am curious as I am not sure about the accuracy of the information. If I do cite someone else's citation (which i am 99% confident they never saw the data), it seems like I can just make something up myself. Is there a way to express that maybe this data is accurate, but that I'm not sure, since it is the only available source about that data? Or should I ignore it all together and pretend as if there is no data?

share|improve this question
2  
I agree with ff524's answer, but I want to add that I think this is a really terrible situation. Just yesterday I found a study with over 400 citations which, once I obtained a copy, was extremely sloppy/borderline fraudulent. –  Alex Becker Apr 7 at 3:15
    
Interesting related paper on arXiv: arxiv.org/abs/1109.2272 From the abstract: "We develop a stochastic model of the citation process, which [...] shows that about 70-90% of scientific citations are copied from the lists of references used in other papers." (People who copy the reference line may still obtain and read the cited paper, but still...) –  cbeleites Apr 7 at 16:04
    

2 Answers 2

up vote 19 down vote accepted

Is there a way to express this (not as a citation format), that maybe this data is accurate, but not sure, since it is the only available source about that data.

How about something like: "X, measured by Jones in 1950, is commonly given as the value for Y (as cited in Smith, 1989 and Cutler, 1995). However, the original manuscript by Jones describing this result is not generally available."

Or should I ignore it all together and pretend as if there is no data.

Don't do this. If X is cited often in the field, then by pretending as if there is no data, you give the appearance of not knowing your field.

Whether or not you should use X (knowing that it is unreliable) depends on your purposes.

If your work hinges on having an accurate value for X, then you definitely should not use X if you cannot track down a reliable source for it. (Nor should you make something up.)

If you are using X just as a "sane" value for something, and it isn't central to your work, on the other hand:

If you think you have a better value than X, and can justify why you are using a different value than everyone else in your field, go ahead.

Otherwise, if you need a value, you should just use X (unreliable as it is). A bit of unreliable data that is accepted in your field is still better than a bit of unreliable data that you just made up with no justification.

share|improve this answer
    
I agree that you shouldn't ignore the existing data, and that it is appropriate to cite it. But you should not use if you aren't convinced that it is reliable. Of course, if you have no reliable data, and can't produce your own, then you might not be able to proceed with this project; that's just the way it is. –  Nate Eldredge Apr 7 at 16:05
    
@NateEldredge I would say it depends how much that value impacts the rest of the work. If you're using X as a "sane default" for something tangential, then an accepted but unreliable X is not such a concern; if your whole study hinges on X, then you have a major problem. –  ff524 Apr 7 at 16:13
    
Sure. I guess by "reliable" I meant "reliable enough for your purposes". –  Nate Eldredge Apr 7 at 16:15
    
@NateEldredge agreed, clarified in an edit –  ff524 Apr 7 at 16:21
    
yea in this situation it is anthropomorphic measurements, so theres no way I can sample a large population size to get my own estimate, but it does data weighs heavily on the project –  user1938107 Apr 7 at 23:39

Actually, "there is no online version (never has been), and the only print copy exists in that country" shouldn't be an obstacle - that's what inter-library agreements are for; univeristy libraries generally are able to get [a copy of] that book for you even it's not online and the physical copies are all located abroad.

It's not conventient and does take time for it to arrive, but if it's widely cited but suspect, then taking a look at the original (and publishing what you find out) would be a Good Thing.

share|improve this answer
3  
Some sources are just not retrievable - ILL only goes so far. But if your library can track down the source, that's definitely the way to go! –  ff524 Apr 7 at 13:54
    
as I mentioned, there is only one copy, so ILL is not an option. Its also at a national library, not a university –  user1938107 Apr 7 at 23:38

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.