class: center, middle, inverse, title-slide # How to publish a paper (as a lead author) ### Andrew Parnell<br> <a href="https://is.gd/how_to_publish">is.gd/how_to_publish</a> --- # Contents 1. Basic steps of publishing a paper 1. What to do with rejections and corrections 1. What to do after the paper has been published 1. Common things that can go wrong <hr> I'm not going to talk about the mechanics of writing a paper as this is a much longer and idiosyncratic process. These days you might be able to scrape through a PhD without at least one published lead author paper but it's becoming less and less likely. In Statistics, most people submit their thesis when they have 2 or 3 published papers. --- # The overall proccess - Step 0: Write the paper - Step 1: Choose a journal - Step 2: Decide on authorship order - Step 3: Go through the submission process - Step 4: Wait and wait and wait for peer review - Step 5: Paper published! - Sit back and wait for Nobel prize, Fields Medal, etc --- # What actually happens - Step 1: Pick a journal and write the paper around that structure - Step 2: Spend three days re-formatting the paper in the journal's proper structure because you forgot they had one - Step 3: Submit the paper. Wait 6 months to a year - Step 4: Paper is rejected. Make changes to paper (if possible) - Step 5: Find a different journal - Step 6: Repeat from step 2 until it's accepted --- # Step 1: Choose a journal - Getting your paper into the right journal is paramount if you want the right people to read it, or to get maximum exposure, or to get people to change their behaviour because of it - For example, if you're publishing new methods associated with climatology, you might choose a more technical journal (higher esteem) or a more applied journal (more readers and citations) - Quite often your supervisor will choose the journal for you as they have the expertise in that area - ...but more often you need to go out and find appropriate journals yourself --- # How journals are structured - Usually there are one or two editors who are senior academics and know the subject area very well - Below them are a load of Associate Editors who deal with the day-to-day running of the journal. The papers are passed to them by the editor and they find peer-reviewers - Working with the editorial board are a set of copy editors who deal with typesetting and similar problems. They are normally located in Asia - Above everyone are the publishers who own the journals, receive the subscription fees, and maintain the submission and review process --- # Finding the right journal - Usually there are multiple journals that are suitable for your paper - You might judge the differences between them by: - Quality metrics (such as `\(h\)` index or impact factor) - Esteem of editorial board - A special issue in your particular field - Turnaround time in getting the paper reviewed - Other papers in the journal (i.e. are they similar or not) - Broadness of readership - Open access - Most journals will have a style guide consisting of: - Word/latex templates - Word and figure limits (some still charge for colour pictures) - Structure and referencing styles --- # Using Google to find journals - Most journals have a web page which you can look at to find the editorial board and the recent papers in it. - However, if you want a quick summary of how good a journal is use Google Scholar: - Go to https://scholar.google.ie - Click on the menu bar and choose 'metrics' - Click on 'Categories' to find good journals - [Demonstration](https://scholar.google.com) --- # Step 2: Decide on authorship order - This can be surprisingly sensitive and cause major disagreements between groups - It also varies between subjects (e.g. maths is alphabetical) - In Ireland, the two most important places to be are either first author or last author (SFI counts these as 'senior' author) - My rule of thumb is: - The person who did the most work goes first - The person who did the second most work goes last - Everyone else goes in the middle and can fight over the rest - This can be very tricky to decide upon, e.g. what about the person who brought in the money? The person who brought in the data? The person who ran the samples through the lab? - This has become so contentious that some journals require you to specify the percentage contribution of each author, or the roles that each author played in writing it > Advice: decide on this as early as possible to avoid fallout --- # Step 3: Go through the submission process - Journal submission processes are usually a nightmare - Each website often has its own registration form (which it claims is linked to other websites but never really is) - Sometimes you will have to submit it in a crazy format, e.g.: - Separating out all the figures (e.g. in separate files) - Writing a separate letter of support to the editor (sometimes important if your paper is slightly outside of the journal's scope) - Putting all the files in the correct format (e.g. all figures must be `.eps`) - Listing all the authors and their affiliations on a separate form - Some journals allow you to bypass all this at the first stage of review by simply submitting the pdf (though they never tell you) --- # Step 4: Wait and wait and wait for peer review - Once submitted you will usually receive a quick email confirmation (if you don't get one contact them and check the website) - Two things will now happen in the next week or two: 1. One of the editors for the journal will get back to you saying they don't think it's suitable (rejection without review) 1. It will be sent out for anonymous peer review - Sending a paper out for peer review means that it will be sent out to two anonymous academics (from PhD students to senior Profs) who will read your paper thoroughly and report back on it. - The journal usually sets them a deadline of anything from 2 weeks to 3 months. - More mathematical journals will usually take longer, and reviewers rarely review on time. --- # Waiting for peer review As soon as I submit a paper I put an entry in my calendar for 3 months' time so that I can remind the journal if they haven't replied. Whilst the paper is in peer review you can: - Run any experiments you think the reviewers might request - Keep checking the journal web page - Write your next paper! --- # A very long wait... <img src="laa.jpg" width="100%" style="display: block; margin: auto;" /> --- # Step 5a: Get a rejection - The two (or more reviewers) might write back saying that your paper is not suitable for publication. They *should* give good reasons for rejecting your paper - Having rejected papers is a sad fact of life for all of us, and a common side effect of being ambitious - A rejection usually hits you harder earlier in your career. Even the most famous academics regularly get papers rejected - If you disagree with a reviewer's opinion you can write back to the editor but you need very strong reasons for an editor to overturn a reviewer's opinion - Even if you disagree with the reviewer's opinion read it carefully and follow the rule: > If a referee did not understand something it is your fault for not writing it clearly enough, not the reviewer's fault for being stupid! --- # Step 5b: Major corrections and re-review - Most commonly, the referees will come back and say that they thought your paper was suitable for publication but there are major changes required - Sometimes the journal calls this a 'reject with resubmission' so that the journal can fiddle with their turnaround times - Your next job is to respond to the reviewers' points and make changes to the paper (more about this on later slides) - When finished, you re-submit the paper and it will be reviewed again --- # Step 5c: Minor corrections - Rarely (but happens more often as you get more senior) a paper will be accepted with only minor corrections - You usually still have to go through writing a response to the reviewers' opinions and the re-submission process but it will not go back out to peer review - If you have done a good job on responding to the reviewers the paper is normally accepted within a few days --- # What to do with a rejection - Get angry, upset, etc, but not for too long - Remember that some of the world's best and most cited papers were rejected from multiple journals - Read again the reviewers' comments and decide to either: - Appeal the decision (almost never works) - Submit to another journal - Go back to the drawing board and re-write the paper - If submitting to another journal I would always try to take on board some of the previous referees' comments. > Never submit the exact same paper twice to a different journal as if it gets sent back to the same reviewer (surprisingly common) it will be instantly rejected --- # Responding to reviewers and writing a rebuttal - If you get major or minor corrections you need to write a point by point summary of the issues raised by the reviewers and how you have changed the paper. - The key rule to responding to referees' comments is: > If the referee doesn't like something _always_ make a change to the paper, even if you believe this makes the paper worse - These rebuttals can often be longer than the original paper and can take months to create (journals usually set a 3-month deadline) - [Example](ResponseToRefs.pdf) --- # After the paper has been accepted - You might think that once the paper is accepted your work is finished, but this is not so - Once the paper is accepted it gets handed over to the copy editors who re-format your paper in the full journal style (usually a more detailed process than the style file you used) - They will send you back a copy of the typeset paper to check through. Read this thoroughly as they will often make mistakes with author details, placement of figures and tables (especially changing legends and captions), and equations - If you only submitted a pdf version originally you will now have to go back through and produce the proper format with appropriate figures (e.g. allowing for black and white printing) - You can also fix minor typos in the paper that you spot but major changes (such as changing paragraphs of text or figures) is usually not allowed - Once approved by you the paper is sent off for final production and you cannot change it any more --- # After the paper has been published - You might think that after the paper is published your job is done, but this is also not so! - Always make sure you have a pdf copy of the final version that you have saved and shared with your co-authors - Once the paper is published you need people to see it and to read it - Advertise it e.g. on Twitter, send it to relevant people you think might be interested, and make sure to keep any associated software/code up to date. If it's really high profile or impactful, contact your communications office to do a press release - If people have email queries or questions, make it a priority to respond to these as it will increase your citations and possibly lead to further collaborations --- # Some tips from my own experience - Keep everything you used to create the paper in the same folder where possible, especially code used to create figures as reviewers often want to change these! - If you're feeling brave, choose a journal and then pick the journal one rank higher than it. If it gets rejected you can still go back to your original journal, and if it gets accepted it will get your paper into a better journal - Keep all code and data in a GitHub repository or similar where possible. Make it available to others so that if there is a mistake it gets uncovered quickly and you can fix it. Papers with associated code and software get many more citations --- # Publishing in open access journals and Plan S - Most highly-esteemed journals charge universities and other organisations large amounts of money to get access to the papers - Since the reviewers and associate editors are not paid, it's not clear why all these papers should cost so much to access - In recent years there has been a push toward open access publishing. Most commonly this means you pay some money when your paper is accepted to get the paper published and the pdf is available online for free - Quite a few journals these days are just completely free to publish in and view online - Even more recently there is a proposal (called Plan S) to stop all government-funded research being published in closed-access journals --- # Pre-prints - The big quandary is that at the start of your career you need to publish in highly-esteemed journals, but these are not open access and so are bad for scientific progress - A happy middle ground can be found by uploading the last submitted version of your paper to a pre-print server - These store pdfs online and make them available for free - When people search for your paper online they will usually find the free pre-print version as it will have the same title - Reading papers direct from pre-print servers is great because you get them much faster than waiting for peer review, but they also tend to contain more mistakes - I usually use Arxiv but there are lots out there. Most universities now also have their own pre-print servers - [Demonstration](https://arxiv.org) --- # Mistakes and (gulp) retractions - Almost all academics will admit to having published papers with mistakes in them - It is almost impossible to fix minor mistakes in journal papers, even though everything is now done online - If someone finds a serious mistake in your paper you should verify it and own up to that mistake as soon as possible. You can make yourself look much better by admitting mistakes and working with those who point them out to you - If a mistake is serious the journal might retract a paper or ask you to submit a response to the query. Traditionally it has been seen as very bad to have a paper retracted --- # Publishing in Nature, Science, and PNAS - (PNAS is the Proceedings of the National Academy of Sciences) - These journals are often called the 'tabloids' - In theory they are the best journals and the ones that are most desired to publish in - Most papers (>90%) get rejected within 2 days of submission without review - The papers tend to be shorter, have the methods section relegated to the appendices, and have higher impact (and therefore usually more dubious) conclusions --- # Crazy stuff that can happen - All kinds of weird stuff can happen when publishing papers. These are some of the things that have happened to me: - The journal can change the reviewers after requesting major corrections. The second reviewers might reject your paper - Some open access journals can charge per page and change the number of words per page - Some journals use double-anonymised peer reviewing - Journals can re-route rejections into their own family of journals - Editors can set up citation cartels between them that falsely boost the performance of their journals - Journals can take an age between accepting a paper and actually publishing it online - ... --- # Summary - Get the author order settled as soon as you start writing - Pick the journal carefully and write for that journal - Don't get too downhearted if the paper is rejected. It happens to everyone - Respond to the reviewers respectfully and follow their instructions - Ask for help/advice from other academics (e.g. me) if you need it - Read this presentation again before you submit your first paper: [https://is.gd/how_to_publish](https://is.gd/how_to_publish) # Questions?