20

So this is pretty much it. My co-author, who is also the main author, has hired an assistant through Elance (an online work outsourcing platform) for doing statistical analysis and also to collect and annotate some relevant literature about connected topic X. I was aware of this and was OK with it, knowing that his or her work is not significant enough to claim authorship but will be acknowledged properly. Now it turned out that, for reasons not entirely clear to me, the RA does not want their name to be disclosed in the acknowledgement section. Even funnier, we do not even know their real name or email address, just the nickname and the profile they use on Elance.

We have the following options:

  1. Do not acknowledge the contribution at all (at first, out of question; but see below why it should be considered)

  2. acknowledge the contribution without naming the contributor. (Like "The authors would like to thank <Main Author>'s research assistant for their help in preparing this manuscript...") As far as I know this would be nonstandard.

  3. acknowledge the contribution without naming the contributor but indicating that remaining anonymous was his explicit request.

  4. acknowledge the contribution and identify them by their Elance nickname (Like "The authors would like to thank to "FyI1978" from Elance for their help in preparing this manuscript...") This would make us seem unprofessional or laughable; if I have the right impression.

  5. redo the statistics and clean the manuscript from any elements that might bear the mark of the RA's contribution (so that we would not have to acknowledge them at all).

What is your advice, how should we proceed?

  • 4
    If #4 is what the RA wants, I think it's the morally correct to do so even if it sounds janky. Ask them. – Max Aug 12 '15 at 18:09
  • 12
    There are many good reasons for your freelancer to want anonymity. Perhaps the person is moonlighting and their day job has a non-compete clause. Also, I recall reading on a ASA email list that acknowledgments (vs. co-authorship) only gives liability without reward and the email poster advised avoiding it. Last, you've paid the person so they have already received their compensation. – Richard Erickson Aug 12 '15 at 19:55
  • 2, 3, and 4 are the most honest solutions, I would ask them which they prefer. They might consider their Elance nickname as too much information as well (or conversely might want their nickname mentioned). – Jonathon Aug 12 '15 at 22:43
  • 2
    @RichardErickson: I think it actually is a problem to have results and analyses in a paper, where the person/company who have produced them is not mentioned by name. Namely, because they are responsible for the validity of their work or product. Mentioning does not necessarily have to be through co-authorship or a formal acknowledgement, but if work has been outsourced to a company, it should at least be mentioned in the Material & Methods-section (or equivalent). – Gerhard Aug 12 '15 at 23:59
  • @Gerhard: I agree your concerns are valid as well. My perspective was from the perspective of the statistician doing the work, not the authors' or generally readership. Pharmaceutical companies have been known to ghost write favorable studies. On the slip slide, Gossett did all of his work under the pen name "Student". The ideal solution would be to discuss this when hiring the statistician, not in a post hoc setting. In this case, Massimo Ortolano's answer appears to be the best solution. – Richard Erickson Aug 13 '15 at 12:41
29

There are several reasons for which that person might want to remain anonymous (e.g. s/he has been a victim of stalking).

I'd then write something like:

The authors would like to thank an anonymous contributor from Elance for their help in statistical analysis and...

Edit: I amended the sentence above as suggested by Jonathon Wisnoski in a comment ('his help' -> 'their help') to reflect the fact that the gender of the contributor seems unknown to the OP.

  • 2
    Yes, great for not throwing suspicion on the contributor. If the paper gets indexed it may increase the online footprint after her name, and there are tons of legal and justifiable reasons for wanting to keep one's online footprint to a mininum. – Mindwin Aug 12 '15 at 20:59
  • 19
    Presumably: their help The poster seems unsure of the gender. – Jonathon Aug 12 '15 at 22:46
  • 2
    @JonathonWisnoski: Agreed. As a non-native speaker I haven't yet got used to this usage of their and them as gender-neutral singular pronouns. – Massimo Ortolano Aug 13 '15 at 7:50
  • 2
    I think there is one more aspect to consider: acknowledging who worked on your research is also a way to make any conflicts of interest explicit. As statistical analysis may affect the overall message of the paper, having someone anonymous on your team may raise suspicion, especially when there is something to gain from a biased result (e.g. global warming, pharmaceutics, ...). I don't think this is the case here, but in general this might be an issue. As in en.wikipedia.org/wiki/Academic_authorship#Ghost_authorship – Martin Modrák Aug 13 '15 at 12:07
  • 3
    @MassimoOrtolano There is no agreement that this is acceptable. Many people would insist that this usage remains simply wrong. In this particular instance, I'd just omit the pronoun altogether. Obviously it is his/her/their help and 'for help...' reads fine. – cfr Aug 14 '15 at 2:20
19

Your suggestion #2 is the best, I think. I've got a paper on my desk just now that acknowledges me (and a few colleagues) in this way - we're referred to as the staff of X group. It doesn't particularly imply that you're doing it to keep them anonymous, and you don't have to make a point of saying so - it just happens that for whatever reason you've not given the name(s).

If the journal objects, you can discuss it with them further.

  • 3
    I'm not sure that the two situations are really comparable. Acknowledging a group without listing its members still reflects well on those members: you can say, "Hey, I was a member of the famous X group, acknowledged in all those papers!" But acknowledging a single person without giving any indication of who they are seems weird, as if you're withholding that information for some reason. Better, I think, to explicitly say that they were an anonymous contributor: then people might wonder why they didn't want their name in the paper, rather than wondering why the authors didn't divulge it. – David Richerby Aug 12 '15 at 19:42
  • @DavidRicherby In this case, the group is "-- library/archives", but I take your point :-). This looks a lot more normal when it's referring to more than one person. – Andrew Aug 12 '15 at 20:11
14

I see nothing wrong with your suggestion #4, if Fyl1978 agrees. As a reviewer or editor, I would think this is interesting and unusual, but certainly not unprofessional -- you thank someone who deserves your thanks, and they chose to not use their real name.

Heck, people put all sorts of thanks into acknowledgments. I've thanked jet lag, one of my friends has thanked a beer company for providing inspiration. Thanking someone who doesn't want their name published doesn't strike me as something particularly out of the ordinary. As I said, certainly not as something unprofessional.

  • 1
    I also like suggestion #4, but thanking Fyl1978 is not exactly anonymous, #4 is fine it the person just doesn't want to share their real name, but is not ok if they want to be anonymous (and of course, by anonymous I don't mean untracable). – dtldarek Aug 13 '15 at 7:48
  • Right, that's why I said "..., if Fyl1978 agrees". – Wolfgang Bangerth Aug 13 '15 at 22:25
13

Write nothing.

Acknowledgements aren't obligatory. Just thank the person in person (aka, via email).

  • 15
    That's just poor style in my book. It's not obligatory to thank those who have contributed, but it's certainly good style. – Wolfgang Bangerth Aug 12 '15 at 16:28
2

The first 4 options are all okay, option 5 is (in general) problematic from a procedural point of view, because you are then not going to declare that you have re-done the statistical analysis. The problem is then that there is typically some freedom in the methods used, to avoid bias in making any choice that could skew the results, you are supposed to declare (or it is a hidden assumption) that certain choices in the analysis were made a priori. Of course, in this case it is not relevant if you are going to forget about the first results and use whatever comes out of the second analysis. But it's still more of a problem compared to the first 4 options as far as the content of the article is concerned.

And option 4 isn't all that laughable, see the second footnote on page 12 of this article.

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