Discussion: View Thread

I've had an issue with a researcher who wants to include me in the acknowledgment section but seems to feel that it is too much trouble to show me the publication before it is submitted. Actually, he did show it to me, but when I asked to be notified if there are any radical changes or if he plans to submit to a different journal, he balked. We agreed that it would be best not to mention me in the acknowledgment section, but I'm curious what other statisticians have done.

Have you been included as a co-author on a presentation or a publication without finding out until after the fact? What did you do about this?

Are acknowledgments different than co-authorship? Would you insist on seeing something beforehand? Would you need to hear about major changes or submission to a different journal?

-------------------------------------------
Stephen Simon
Independent Statistical Consultant
P. Mean Consulting
-------------------------------------------

Prety bad form on that researcher.
I don't consider an asknowledgement as very important, but I agree that if there are stats in it, I would pull my name if I didn't have a chance to review it.  What you are refering to is close to counter what most academic places require for articles.  On the other hand having to make a point that I desrve to be a co-author because of all the effort, has happened all too often.  But remember, getting funded does not mean you don't have to be included - after all NIH grants fund and expect co-authorship.  Legit (and ethical) authors don't balk at letting you see a manuscript that has major rewriting. After all you wouldn't want to be a co-author OR acknowledged by one of these papers that has to be withdrawn.  You have a good case for contacting the institution about this; however it depends if you wan't to completely burn your bridges. (I think I would with this investigator anyways, but that's your decision). 


-------------------------------------------
Raymond Hoffmann
Professor
Medical College of Wisconsin
-------------------------------------------

Stephen, I agree pretty much down the line with what Raymond has said.  Of the issues that you raised, only with regard to publication in a different journal can I perhaps make a case for failing to inform you -- but even then it just sounds absurd.

-------------------------------------------
David Mangen
Mangen Research Associates, Inc.
-------------------------------------------



Original Message:
Sent: 11-05-2014 16:31
From: Raymond Hoffmann
Subject: Unexpected co-authorship or acknowledgment

Prety bad form on that researcher.
I don't consider an asknowledgement as very important, but I agree that if there are stats in it, I would pull my name if I didn't have a chance to review it.  What you are refering to is close to counter what most academic places require for articles.  On the other hand having to make a point that I desrve to be a co-author because of all the effort, has happened all too often.  But remember, getting funded does not mean you don't have to be included - after all NIH grants fund and expect co-authorship.  Legit (and ethical) authors don't balk at letting you see a manuscript that has major rewriting. After all you wouldn't want to be a co-author OR acknowledged by one of these papers that has to be withdrawn.  You have a good case for contacting the institution about this; however it depends if you wan't to completely burn your bridges. (I think I would with this investigator anyways, but that's your decision). 


-------------------------------------------
Raymond Hoffmann
Professor
Medical College of Wisconsin
-------------------------------------------

Hi Stephen,
Most of the researchers I've worked with are happy to have me look at the manuscript and work submitted. They want to be sure they said everything correctly.  So definitely this situation you have raises a big red flag!
In my personal experience I haven't had someone add my name to a paper without my consent.  However, I once did some sample size calculations for a clinical trial protocol and found out a year or so later, when they wanted some more calculations, that my name was on the protocol as an investigator. I just asked that my name not be included as an investigator of the study and that was that. And yes, we are still talking to each other.
Best,
Elaine


-------------------------------------------
Elaine Eisenbeisz
Owner and Principal Statistician
Omega Statistics
-------------------------------------------



Original Message:
Sent: 11-05-2014 16:36
From: David Mangen
Subject: Unexpected co-authorship or acknowledgment

Stephen, I agree pretty much down the line with what Raymond has said.  Of the issues that you raised, only with regard to publication in a different journal can I perhaps make a case for failing to inform you -- but even then it just sounds absurd.

-------------------------------------------
David Mangen
Mangen Research Associates, Inc.
-------------------------------------------

I don't have much to add to the great points by Ray Hoffman and others, but I would like to share some pithy advice I received from my division chief about thirty years ago: "Never accept an acknowledgement - it is all of the responsibility with none of the credit."

-------------------------------------------
Deborah Dawson
Director of Biostatistics
University of Iowa
-------------------------------------------



Original Message:
Sent: 11-05-2014 16:31
From: Raymond Hoffmann
Subject: Unexpected co-authorship or acknowledgment

Prety bad form on that researcher.
I don't consider an asknowledgement as very important, but I agree that if there are stats in it, I would pull my name if I didn't have a chance to review it.  What you are refering to is close to counter what most academic places require for articles.  On the other hand having to make a point that I desrve to be a co-author because of all the effort, has happened all too often.  But remember, getting funded does not mean you don't have to be included - after all NIH grants fund and expect co-authorship.  Legit (and ethical) authors don't balk at letting you see a manuscript that has major rewriting. After all you wouldn't want to be a co-author OR acknowledged by one of these papers that has to be withdrawn.  You have a good case for contacting the institution about this; however it depends if you wan't to completely burn your bridges. (I think I would with this investigator anyways, but that's your decision). 


-------------------------------------------
Raymond Hoffmann
Professor
Medical College of Wisconsin
-------------------------------------------

------Original Message------

I don't have much to add to the great points by Ray Hoffman and others, but I would like to share some pithy advice I received from my division chief about thirty years ago: "Never accept an acknowledgement - it is all of the responsibility with none of the credit."

-------------------------------------------
Deborah Dawson
Director of Biostatistics
University of Iowa
-------------------------------------------



Original Message:
Sent: 11-05-2014 16:31
From: Raymond Hoffmann
Subject: Unexpected co-authorship or acknowledgment

Prety bad form on that researcher.
I don't consider an asknowledgement as very important, but I agree that if there are stats in it, I would pull my name if I didn't have a chance to review it.  What you are refering to is close to counter what most academic places require for articles.  On the other hand having to make a point that I desrve to be a co-author because of all the effort, has happened all too often.  But remember, getting funded does not mean you don't have to be included - after all NIH grants fund and expect co-authorship.  Legit (and ethical) authors don't balk at letting you see a manuscript that has major rewriting. After all you wouldn't want to be a co-author OR acknowledged by one of these papers that has to be withdrawn.  You have a good case for contacting the institution about this; however it depends if you wan't to completely burn your bridges. (I think I would with this investigator anyways, but that's your decision). 


-------------------------------------------
Raymond Hoffmann
Professor
Medical College of Wisconsin
-------------------------------------------

Dear Stephen,

Many years ago I read of a statistical consulting group that had as a policy to refuse acknowledgements. I wondered about it at the beginning, but through the years I've come to understand that it is the best policy.

I myself do the same nowadays. If I contributed  substantially, I deserve to be a co-author and have a say in the manuscript. If I didn't, then I don't want to jeopardize my reputation by being associated to something where I don't have a say. Often you don't get to see the manuscript, or you have no idea how your advice/contribution has been used. Thus, thanks, but no thanks.

I have a colleagues who was surprised to see her name on an article she had no idea was published and was not happy at all. Another who was acknowledged in an article which he thought was not good at all (at least the statistical part) and felt ashamed to be associated with it. Not worth it for something that will not show up in your CV or publication list.

Hope it helps.

Sharon
-------------------------------------------
Sharon Kühlmann-Berenzon, PhD
Sr Biostatistician
Public Health Agency of Sweden
Stockholm
-------------------------------------------

Dear Stephen,

I agree with Sharon. If you don't have a say, it's better not to be mentioned in the paper.

In the biomedical field there are pretty clear recommendations about authorship, contributorship and acknowledgment issues (you can see more on the http://www.icmje.org/recommendations/browse/roles-and-responsibilities/defining-the-role-of-authors-and-contributors.html). In the case of the acknowledgement, they advise editors to require that the corresponding author obtain written permission to be acknowledged from all acknowledged individuals.

Most of the journals in the biomedical field have authorship statements which have to be filled and signed by all co-authors prior to publishing the paper, so it's not so easy to be named as an author without knowing.

-------------------------------------------
Vesna Ilakovac
School of Medicine, J.J.Strossmayer University of Osijek
Osijek, HR-31000, Croatia
-------------------------------------------

In our academic consultancy, such rules are signed up front. I proactively point out that the rule, that naming me or my institution in an acknowledgement or as an author requires our written consent to the final draft and the published version, might require a re-writing by me. Why should I? Was I payed for it? Why wait for me? Your case may well have had the same result here. Discussing a project is similar to teaching, which goes without acknowledgment. If clients wish to name me or my institution nevertheless, I need to react, mostly by writing letters. I might rarely ask to be acknowledged for pointing to some technique rather than analysis and reporting, as research needs to be transparent about the means used. Once upon a time, acknowledgements ended with "Any errors are mine." I think this distinguishes acknowledged help from co-authors' contributions. Consequently, acknowledgements were banned by many jounals.
Having contributed enough to be an author, like analysis and reporting (did you?), changes things. I do not accept the frequently offered choice of either money or authorship. It's either enough contribution for both or neither of them, as a rule. Co-authors have to allot each other review time. Finding out authorship after the fact, I would review the report as usual. Sometimes I wrote to the editor, mostly to correct points irrelevant to the conclusion. Telling clients about corrigenda usually suppresses acknowledgements or brings the final draft on my desk, as appropriate. Those that hope to gain respectability by using my name will fear beeing publicly and permanently criticised by me. I am grateful to journal editors that require signed statements on the published version from all authors. Some even started to hassle the acknowledged like that.
Sometimes, I have to inform review boards that no co-operation on some project was agreed upon. The projects would be of the kind, R.A. Fisher called performing a post-mortem.

-------------------------------------------
Reinhard Vonthein
Universitaet zu Luebeck
-------------------------------------------

That is a frustrating situation, but unfortunately not at all uncommon.

I have learned to make it a rule not to be acknowledged: I'm either a co-author or nothing, and for consulting projects I am happy with nothing, because my payment is monetary not academic. As a statistician, there is little good that can come out of an acknowledgement. If there is nothing wrong with the paper, no one notices the acknowledgements. However, if there are analytic problems, you will get noticed as the consulting statistician and the responsibility will fall on you - particularly if there is no statistician as a co-author.

Some journals are moving to a style of listing what everyone's role was. If your role was just high level advice and it is listed as such, I would be more comfortable not seeing every detail. However without that designation, I think you need to presume that any questions about data integrity will fall on you. If you are "the statistician", you should have code available to quickly reproduce every result/figure/table in the paper in case there is an audit. In that sense I would insist on seeing final drafts of the paper.

Best of Luck,

Ben

-------------------------------------------
Benjamin Goldstein
Assistant Professor Biostatistics
Duke University
-------------------------------------------

As with most of life, a one-size-fits-all rule rarely is appropriate. While I agree that acknowledgement is generally not ideal for a statistician, there are cases where, in my opinion, it is perfectly appropriate.

One example is on a methods paper where the authors are primarily statisticians. Acknowledging a colleague's review and critique is perfectly acceptable. An additional example might be a clinical or applied paper where there is another statistician as an author. An acknowledgement referencing assistance with some part of the study design and/or analysis would seem perfectly acceptable if it was felt the criteria for authorship wasn't met.

The key in both of these cases is being explicit in why the person is being acknowledged. With this approach, it will be clear what the acknowledgement is for, and perhaps more importantly, what it is not for. A handful of journals are now requiring written approval of the person that is to be acknowledged. If this became more of the norm, perhaps the concerns regarding acknowledgements will be lessened since the acknowledgement text would be open to edits and approval by you. 

Another concern that is linked with authorship and acknowledgements but is not often discussed is ghostwriting. That is perhaps a topic for another conversation.

In terms of the original question regarding being listed as a co-author without your approval, yes, that has happened. I typically start by giving the person the benefit of the doubt that they were looking to be inclusive and be a good colleague. From my experience, it seems best to take the high road and discuss why you would like to be more involved before the submission on future papers. Most papers come back for revision, so you can "fix" the paper if needed and positively influence the corresponding author at the same time. 

-------------------------------------------
Rickey Carter
Associate Professor of Biostatistics
Mayo Clinic
-------------------------------------------



Original Message:
Sent: 11-06-2014 09:19
From: Benjamin Goldstein
Subject: Unexpected co-authorship or acknowledgment

That is a frustrating situation, but unfortunately not at all uncommon.

I have learned to make it a rule not to be acknowledged: I'm either a co-author or nothing, and for consulting projects I am happy with nothing, because my payment is monetary not academic. As a statistician, there is little good that can come out of an acknowledgement. If there is nothing wrong with the paper, no one notices the acknowledgements. However, if there are analytic problems, you will get noticed as the consulting statistician and the responsibility will fall on you - particularly if there is no statistician as a co-author.

Some journals are moving to a style of listing what everyone's role was. If your role was just high level advice and it is listed as such, I would be more comfortable not seeing every detail. However without that designation, I think you need to presume that any questions about data integrity will fall on you. If you are "the statistician", you should have code available to quickly reproduce every result/figure/table in the paper in case there is an audit. In that sense I would insist on seeing final drafts of the paper.

Best of Luck,

Ben

-------------------------------------------
Benjamin Goldstein
Assistant Professor Biostatistics
Duke University
-------------------------------------------

I had a situation in which I asked to neither be a coauthor nor acknowledged.  I had built a simple demographic forecasting model for a customer, who then published results from it.  I was fully aware of the intent of the model.  The problem was that the customer was a controversial advocacy group, whose numbers are respected nonetheless.  It would not have been good for my career to be publicly associated with it or to appear to be competing with my governmental employer.  At least, this job was essentially programming to the customer's specifications.  I did not have to exercise any professional judgment about the results.

-------------------------------------------
Charles Coleman
-------------------------------------------



Original Message:
Sent: 11-06-2014 09:46
From: Rickey Carter
Subject: Unexpected co-authorship or acknowledgment

As with most of life, a one-size-fits-all rule rarely is appropriate. While I agree that acknowledgement is generally not ideal for a statistician, there are cases where, in my opinion, it is perfectly appropriate.

One example is on a methods paper where the authors are primarily statisticians. Acknowledging a colleague's review and critique is perfectly acceptable. An additional example might be a clinical or applied paper where there is another statistician as an author. An acknowledgement referencing assistance with some part of the study design and/or analysis would seem perfectly acceptable if it was felt the criteria for authorship wasn't met.

The key in both of these cases is being explicit in why the person is being acknowledged. With this approach, it will be clear what the acknowledgement is for, and perhaps more importantly, what it is not for. A handful of journals are now requiring written approval of the person that is to be acknowledged. If this became more of the norm, perhaps the concerns regarding acknowledgements will be lessened since the acknowledgement text would be open to edits and approval by you. 

Another concern that is linked with authorship and acknowledgements but is not often discussed is ghostwriting. That is perhaps a topic for another conversation.

In terms of the original question regarding being listed as a co-author without your approval, yes, that has happened. I typically start by giving the person the benefit of the doubt that they were looking to be inclusive and be a good colleague. From my experience, it seems best to take the high road and discuss why you would like to be more involved before the submission on future papers. Most papers come back for revision, so you can "fix" the paper if needed and positively influence the corresponding author at the same time. 

-------------------------------------------
Rickey Carter
Associate Professor of Biostatistics
Mayo Clinic
-------------------------------------------