Benjamin Bell: Blog

For anyone thinking about embarking on a PhD, or has recently started the journey, a common question that arises is what format will you choose to publish your thesis? Do you go down the "traditional" route, writing a monolithic piece of text, or opt for the "thesis by publication" route (also known as alternative format, or article thesis), where you publish journal articles "as you go".

What option you choose may ultimately result from what question you trying to answer, the subject you are doing, or by the options offered by your university. Your future career choice may also dictate the type of thesis you write. Thinking of an academic career? then "publish or perish" may sound familiar to you!

Having recently completed my PhD thesis by publication, I wanted to share my thoughts and experiences on this format. Admittedly, before I started my PhD, I hadn't even thought about the different options for the thesis, but it seemed like a no brainer to do it via publications. After all, no one will ever read it, right?

What I also wasn't aware of at the time, was the "debate" surrounding the format - A quick google search of the pros and cons of different thesis formats or asking the question "should I do a thesis by publication?" should give you a good idea. Although, thesis by publication is now well-established, and some of the views presented may be out dated.

Another common question that crops up online is whether a thesis by publication is "easier" to write and easier to defend. This seems to be a common misconception based on the idea that writing an 80,000 word thesis is "hard", while writing a few 4000 - 8000 word articles is easy. So, is thesis by publication the easy route? Absolutely not.

Another common misconception is that a few published journal articles is all you need to complete your thesis by publication. The guidelines and requirements for thesis by publication will vary by university, but generally, you are still required to write an introduction, literature review, methodology and a conclusions chapter. Your published articles effectively replace the results and discussion chapters.

Therefore, you need to balance writing a "traditional" style thesis, and writing publishable journal articles. Your final thesis by publication may end up being the same length had you chosen to write it in the traditional format.

You also do not have to actually publish the articles to complete your thesis (although this may vary by university). They simply need to be written in "journal format". My final thesis included two published papers, a third paper which I had submitted for publication, and a fourth paper that I had not submitted.

So, why choose the publication route?

Ultimately, the biggest benefit is you get experience publishing straight away, and hopefully a track-record of publications by the time you finish - necessary for an academic career, but also useful for non-academic careers.

Is it easier?

Some key things you need to think about when publishing a journal article (which you may not need to consider for a thesis chapter) include: What is the "message" or purpose of the paper? What are key questions you are trying to answer? Who is the target audience? Is it novel?

This is where writing journal articles becomes more difficult than writing a traditional thesis chapter. If you have ever tried to write up a piece of work (such as an undergraduate or masters dissertation) as a journal article, you may find yourself running into a few stumbling blocks. For example, I have recently written up my MSc dissertation as a journal article which I hope to publish soon. But, this process was much more difficult than I had expected. While the dissertation was well written and had some good results (which I hope will be of interest to people!) - the text was quite general with no clear "message". I found myself having to start from scratch, re-writing the introduction, methods, discussion and conclusions. I also had to do more research to ensure the paper was still relevant and up to date. After all, my MSc dissertation was completed four years ago, and in science, that might as well be a lifetime.

So, does it make sense to write a traditional thesis chapter, which you'll try and convert to a journal article after submission?

The answer is that it makes no sense to do it this way, which leads to the other major benefit of doing a thesis by publication. You will become a much better writer. Writing journal articles teaches you to become more targeted with your writing, i.e. you will write with purpose. You'll write more clearly and concisely, and these skills will help you with the rest of your thesis.

Another benefit to thesis by publication, is that it allows you to more easily manage your entire project, by splitting it into smaller sub-projects. A PhD is a massive undertaking and can be quite overwhelming when you sit down and think about what you need to so. By splitting up your project, you can focus on the different aspects, write up the results, publish, and move onto the next part.

Writing up papers and submitting "as you go" provides a further benefit - you'll get peer-reviewed feedback on your work from experts in the field, which can be invaluable for improving your work.

But, be prepared to have your work rejected, and this can be a very disheartening experience.

The first paper I wrote for my thesis was rejected by the journal. I remember when I originally submitted the paper, thinking that I had written something everyone would want to read, and that I had produced something that was novel and exciting. How naive! I had spent months working on this manuscript, and it had gone through more revisions than I care to remember. But, the long hours you poured into your work have no bearing on the outcome as to whether or not it will be published.

I had managed to get past the "desk rejection" stage, and my work was sent out for peer review. Several months went by, and then I received the dreaded rejection email. Included were the reviewer comments, and although there were many positives, the feedback also pointed out what they considered to be some fundamental flaws which would require revisions before the work could be published, although in this case, not in the same journal.

At first, the feedback received was pretty hard to read. But, in hindsight, this feedback was invaluable, and looking back now I would consider the original paper I wrote to be pretty terrible.

One of the challenges you'll face when doing a thesis by publication is deciding how many publications to write. It can be tempting to try and write as many papers as possible, aiming for quantity over quality - but this approach makes its more likely your work will be rejected.

Part of the reason why my original paper was rejected, was that it did not contain enough data or evidence to support the message I was trying to get across. I had tried to split my "first paper" into two pieces of work, when really, it should have been presented as a single paper. I also wrote in a style that was more akin to a dissertation, by including loads of supporting literature, resulting in long introduction and discussion sections. As one of the reviewer's put it: "it's my opinion that the manuscript is too long" - and they were right, it was unfocused and lacked clarity.

I also needed to do some more lab work, to provide additional supporting evidence. So, I went back to the drawing board and started again. When I re-drafted the paper, I pretty much re-wrote the entire thing, with probably less than 1,000 words of the original surviving. This time, I had written up all the data, including the additional lab work, yet I still managed to write a paper that was now 2,000 words less than the original. Thankfully, on the second submission, the paper was accepted.

In other words, I became a much better writer and researcher as a result of the rejection. Had I not tried to publish my work, I may have faced some very difficult questions come the viva, or even failed (thankfully I passed with minors!).

All in all, the entire process of getting my first thesis paper published, from the original manuscript to the final published paper took about nine months to complete. This included completing additional lab work, countless re-writes and revisions, and two rounds of peer review.

The lessons I had learnt from this process made the second paper easier to write, which was accepted (with minor revisions) on the first submission. But, by no means could or should you consider a thesis by publication to be the "easy option". A PhD is hard work no what which format you choose to write your thesis, and anybody who has completed one will agree that it is probably one of the most difficult things you will do.

Thanks for reading, I would be interested to hear your own thoughts and experiences!

Further reading

Thesis by publications: you're joking, right? - Another viewpoint, and discussion on publishing.

A PhD by publication or how I got my doctorate and kept my sanity - And another look at this topic.

PhD Comics - A lighter look at academia.