James Lewis discusses whether the traditional PhD thesis is the best way for assessing PhD students, and what a good alternative might be.
Contributor James Lewis
Writing up several years of research into a thesis is one of the most challenging hurdles in the long process of getting a PhD. I began writing up in April 2015 and I find myself frequently wondering whether such a mentally taxing task is the best way to examine students before sending them off into the wider world. At the end of July I published a two question poll on my personal blog designed to find out the opinions of others in academia. Over the course of 12 days, 111 people responded. The full results for the poll can be found directly here.
I encouraged people to comment and give their thoughts and suggestions. It is widely accepted that a thesis is the best form of assessment for the humanities so here I focus on whether it’s compatible with the sciences, where research is primarily shared through peer reviewed journal articles.
The traditional thesis
The first poll question asked whether the PhD thesis is an effective way to assess students and 63% stated that they believe it is.
Those in favour argued that it’s the most flexible format for developing writing skills, demonstrating knowledge and allowing discussions of ideas, speculation and negative results that wouldn’t make it into a published paper. Thesis writing also directly demonstrates a student’s ability to plan, carry out and write up a research project to examiners and future employers. Details of method development can be particularly useful resources for laboratories as those following in the student’s footsteps can quickly find out what worked and what didn’t.
Around a third think that the thesis is ineffective. One comment stated that unless a thesis is coupled with publications, there is little contribution to wider scientific knowledge. Others felt that the formality of printing and binding a thesis should be relaxed and that large scale alteration of published papers into the format of thesis chapters is outdated and unnecessary.
The viva (an oral examination of a student’s thesis) was singled out as valuable but requiring better quality control with student support and the ease of selecting examiners varying widely from supervisor to supervisor. One commenter accused academia of being too archaic to consider other options for examining students and several respondents stated a desire to present some of their research in more innovative ways such as a series of videos.
The stress of combining thesis writing with job hunting was also highlighted and it was felt that more time should be set aside for students to finish their thesis and find employment. Many students are left writing up with no further financial support remaining. From my personal experience and talking with other students I believe the UK standard of expecting research to be completed in three years is too short, particularly for laboratory-based projects. Method development is time consuming and involves a lot of trial and error and inevitably the most exciting results come right towards the end of a project. Allowing at least a few more months for finishing up work would allow time for key projects to be finished up and manuscripts to be prepared, which would make the subsequent thesis writing much easier.
The alternatives
The second question asked what the best alternative to thesis writing might be. Assembling a portfolio of papers and prepared manuscripts was by far the most popular of the five options and received 40% of the votes. The runner up was using a criteria based assessment, which received 18% of the votes. In this method students would typically have to demonstrate proficiency in relevant research skills, have prepared a set number of manuscripts and presented their work at a set number of conferences.
For long and complex experiments the publishing process can be slow and many expressed a concern that any assessment that requires or strongly encourages publication could place a great deal of additional stress onto students. Those students whose work is under confidentiality agreements would also need to be considered. One point I agreed with is that writing the thesis introduction chapter is a very valuable tool for getting students to step back and think about the broader scientific context of their work. It was also suggested that there should be a single worldwide thesis database accessible to all.
Global perspectives
I study in the UK and several of the comments I received were from students elsewhere in Europe who expressed surprise that a monograph remains the standard here. In countries such as Sweden, Spain and the Netherlands a thesis typically consists of a short introduction followed by published work stapled together.
An ideal solution?
Introducing the flexibility of a sliding scale from a classic monograph to an introduction with a series of published papers and/or prepared manuscripts might an ideal solution. Writing an in depth introduction is extremely valuable for students and I feel it should be encouraged regardless of the format of the rest of the thesis.
In a compilation of papers and manuscripts it would be easy to include small chapters on method development/negative results etc. If a student hasn’t yet published then discussion of their manuscripts during the viva would be very valuable before the more rigorous challenge of peer review.
Though the classic thesis is undoubtedly effective at assessing students I feel it doesn’t fit well with the demands of finding a job and communicating research and creates a great deal of unnecessary stress. I hope that the issue of thesis reform in the UK can be further debated.
What are your thoughts? Do you think the classical thesis and viva form of assessment is helpful? Or do you think there should be an alternative method of assessing PhD students? Please share your thoughts in the comments below so that we can continue this discussion!
James Lewis is a PhD student at Imperial College London and is currently 55,000 words into a thesis studying the interactions between minerals and organic matter during life detection experiments on Mars.
Report this comment
I feel your pain at having to write the ‘t’ word and look for a job at the same time. And agree that things may only start to get interesting after three years of research, when you have to down-tools and write it all up. But I think that simply stapling several manuscripts together (whatever stage they may be in the publishing process) and calling it a ‘thesis’ will lose thevital learning process of writing (and you do learn by writing a thesis). Furthermore, you can never be sure of the contribution of each author in such a compedium of multi-author research papers; not just who did the work?, but where did the ideas come from? and who did the analysis? and the writing? In a conventional thesis, such things are clearer. And form a permanent record (to gather dust on a shelf in a university library…).