Publishing with Students

Publishing original research with students: the case of ‘MPs on PMs’

Nicholas Allen, Royal Holloway

Inspiration for teaching can occur at the most unlikely of times. At risk of sounding like Alan Partridge, the inspiration for this particular exercise came to me late one Sunday evening as I was approaching the Stokenchurch Gap on the M40 (I was heading back towards London). By the time I reached the exit for the M25, I had a clear sense of what I wanted to do, and how.

The basic aim of the exercise was simple enough: to undertake original research with a group of final-year undergraduate students and publish the findings. I had done this once before with a previous cohort. That project had involved taking Patrick Dunleavy’s measures for analysing ‘positional power’ in the cabinet committee system and applying it to David Cameron’s coalition government. It had been a great way of linking research to teaching and of introducing students to the professional side of academic research.

Surveying MPs

The new project would be slightly more ambitious and involve the partial replication of Kevin Theakston and Mark Gill’s attempts to evaluate prime ministerial success. But whereas Theakston and Gill had surveyed survey academics, we would survey MPs.

At this point, it is worth noting that there appear to be two basic ways of incorporating group projects into taught modules. The first is to have a project as an intrinsic part of a methods course—a module concerned primarily with developing students’ knowledge and understanding of the research process—and to use the project as a frame for introducing different methods and requiring students to apply them. This model has been employed successfully by the LSE’s Ed Page as part of his GV314 module.

The second model, and the one I pursued, is to run a project in parallel to the content of a substantive course—a module that focuses on developing students’ knowledge and understanding of the content of research rather than the process per se—and to use the project as a way of motivating students and introducing them to activities they would not otherwise have experienced.

The course I convened at the time was a final-year module on the British prime ministership, which was capped at 20 students. In order to embed the project, I reorganised the order of the topics so that we addressed notions (and measures) of prime ministerial success in the second week. This meant that all students were immediately familiar with Theakston and Gill’s original study.

Over the following three weeks, I set aside 20 minutes ahead of each class to develop with the students a two-page questionnaire that we would post to all MPs. During these periods, we discussed all the survey items at length and the challenges of achieving a good response rate. We also discussed issues of anonymity and confidentiality, and the logistics of processing and coding the responses.

The final two-page mail-back questionnaire was posted to all MPs, together with a pre-paid return envelope and covering letters written and signed by pairs of students. As I explained to the class, I reckoned that an MP would be less likely to reject a short survey from two young voters than from one. All the students assisted with preparing and stuffing the envelopes on one evening outside of class. (The costs of printing were covered by a teaching prize I had received for the earlier cabinet-committee project.)

Much to our delight, we received responses from 158 MPs. Each student was responsible for coding just over a dozen responses and then checking the coding of a dozen more. Once all the data had been entered, it was simply a case of generating an SPSS dataset that could be analysed.

We discussed basic frequencies in the last week of the autumn term, and the students collectively presented the project to our departmental research seminar in the last week of the spring term. In the meantime, I set out to write a first draft of the paper that would report the research—coordinating nearly 20 separate authors was simply unpractical—which students then read and commented on.

I then submitted the paper to the journal Politics and forwarded the reviews to the students. Students again read the revised paper and response to the reviews. To our collective delight, the paper was accepted for publication prior to the students’ graduation ceremony.

What went well?

From my point of view, the project was a great success. All the students signed up to additional non-assessed work, and it generated excitement throughout the module. They were all exposed to the practicalities of research, including the peer-review process, and the division of labour meant that all took an equal part in preparing the survey and mail-out, coding the responses and discussing and presenting the results. Given that the project was not assessed, the division of labour was probably a necessary condition to its success.

From their feedback, the students really enjoyed the project. The results were picked up on by the national press, and most importantly of all, we contributed an original academic article to the literature on prime ministerial success!

Lessons learned

Running projects with students takes up an enormous amount of time. If the project is undertaken outside of class, then your commitment is unlikely to be recognised in any institutional workload model. I would strongly encourage tutors to embed projects within a module they happen to be teaching, even if the module is not primarily about research. I should have done this, and will do so in future.

Class projects obviously need to fit the module you happen to be teaching. There are limits to how many students can actually be involved if you want them all to be involved equally in data collection and analysis. You also need to accommodate diversity in student ability, as well as commitment. As a general rule, larger number of students participating in a project, the more researcher-led it needs to be.

It is obviously important to develop a project that is actually viable, both in terms of research design and what you expect students to contribute. If students are to collect original data, it needs to be accessible and easily located. Replicating an existing study may also help. At any rate, be creative and consider projects where student participation may be a crucial ingredient in accessing data.

Finally, have realistic goals for publishing the results of any class project. It is always best to aim for an article in a peer-reviewed journal, not least because it is a major part of being a professional political scientist. But if you are unable to place the research in a journal, writing a blog is also a great way of disseminating research and a useful exercise for students.