I’ve been thinking a lot lately about what makes a good and successful scientist, and I keep coming back to resilience. A lot of people don’t realise that doing scientific research can be incredibly frustrating, and it often feels like you are taking one step forwards and two or three steps backwards. A bit like playing a game of snakes and ladders.
In scientific academia, “success” is generally measured by getting your research peer reviewed and published, and, ideally this then being used by other scientists to help their own research. But successfully completing a research project and getting it published is a lot of hard work – loads of things can go right or wrong, and sometimes this is just down to luck rather than skill.
Let’s think about all the steps in an academic science research project, and some of the more common pitfalls.
- First, you might need to collect samples to carry out experiments on. Depending on your research question this could be as easy as opening a chemical store cupboard, or it could require intensive fieldwork in remote places (or, for planetary scientists and cosmologists, sending robots to even more remote places like Mars or comets). For a selection of examples of how things can go wrong in the field, check out #FieldworkFails on Twitter. Also, thanks to the human error element in shipping companies, getting your samples and / or equipment back to the lab isn’t always straightforward.
- Once you have your samples, you need to carry out your experiment. This might involve expensive, specialised equipment. There might only be one or two labs in your entire country (or even none) that have this equipment, and so those people-skills, that scientists are stereotypically famous for not having, come in really handy here for negotiating instrument time.
- You have your experiments scheduled – great stuff. Often, everything will go according to plan. But some instruments and techniques are so sensitive they could be considered temperamental; malfunctioning air conditioning, computer crashes, power cuts, vacuum pump failures, urgent and unanticipated instrument maintenance, unexpectedly incompatible samples – these are all issues that have delayed my data collection at some point in the past.
- Once you have your data, things get interesting. In an ideal world, this data will answer a deep and meaningful scientific question. But in reality, you look at the data and realise that you weren’t quite asking the right question, or that the experiment design wasn’t quite right and you either need to re-do the experiment in a different way, or collect some extra data. This is something that does improve with experience. But most science is about discovering new stuff, or confirming existing hypotheses in different ways, so this risk never goes away completely. In short, science is an iterative process where you are constantly testing, evaluating, and retesting your ideas (and coming to terms with that can be quite painful for scientists-in-training).
- Once your data is fit for purpose, it is time to start evaluating it and writing it up. In my experience, this usually involves a few more iterations of the hypothesis, and occasionally some more rounds of data collection to polish things up. This is also the part where the quality and enthusiasm of your collaborators can have a big impact on the project’s success – especially for junior researchers or for projects in a politically sensitive environment. Collaborator attitudes can either be a springboard to success, or a dragging weight that can even kill off a project all together.
- Congratulations – you and your co-authors have successfully created a manuscript describing your work. Time to submit it to a peer-reviewed journal. Now. If you haven’t submitted at least one paper, you probably have no idea just how hellish a paper submission can be (it is a close second behind university job applications for irritating faff!). Even if you carefully read all of the author instructions when formatting your paper, there is usually something that gets missed. References aren’t formatted correctly. You thought you could submit figures as jpegs but they only accept tiffs. You need to suggest 5 reviewers, not the usual 2-3. Your abstract is 7 words too long. You have to provide a “highlights” file. A graphical abstract is compulsory. Things are improving, and more and more journals are now accepting initial submissions in “your paper your way” form; that means you submit a single word file with figures in line with the text, rather than uploading a separate manuscript file with figure captions listed at the end, and uploading each image separately. But it is still a pain. I always think it should be a half-an-hour job to submit a paper, but it always takes me at least half a day.
- Now, sit back and wait for the peer review. Peer review is one of the foundations of science. Your scientific peers check your work, highlight any problems, and suggest ways to improve the work. This is supposed to ensure that only high quality work gets published. I am a big fan of the concept of peer review, but it has its problems, mostly relating to reviewers and editors being human, likely overworked, and things like egos and bias getting in the way. Now, I am a firm believer that, 90% of the time, if a reviewer hasn’t understood an aspect of the work, then that means I need to explain it better. But most researchers have, at some point in their careers, received a peer review where the reviewer has clearly not even tried to understand the material, and is intent on forcing their own biases onto the subject. Communities such as Facebook’s 11,000+ member group Reviewer 2 Must Be Stopped and the Twitter account Shit My Reviewers Say, with more than 37,000 followers, are testament to the ubiquity of this problem. Sometimes, the scientific editor handling your submission will realise what is happening and place less weight on this kind of review. Other times they might not.
- Once the peer review is complete, the scientific editor will make a decision: Accept as is (this is quite rare, at least in Earth Sciences); Revise (i.e. address all the reviewer comments – sometimes this will go back out for peer review; and nowadays this may come in the form of a “reject and resubmit” decision so that journals can appear to keep their article processing times down); Reject. Reasons for rejection might just be that the journal isn’t the most appropriate host for the work, or that the science is thought to be unsound.
- For revisions, the journal usually allows between 1 and 3 months to complete the revisions, depending on how major they are. They might include things like changing the language style, clarifying points, making diagrams clearer, considering new ideas or information, and even collecting a bit more data or re-evaluating some conclusions. Sometimes a major revision decision can leave you feeling like you are back at square 1.
- Once accepted, then comes all the publication admin. Transfer of copyright, creative commons licences, etc. (I have a life goal to one day properly understand these). Colour printing options / charges (some journals even charge for colour figures in online versions of papers!). And then there are the copy editors. The journal passes your accepted manuscript to a copy editor who formats the article ready for publication and ensures the text matches the journal’s preferred style. They then send you an article proof to check over and approve. A lot of the time, this is a smooth process, but there are also plenty of times where the proof seems to have been created by badly programmed robots. I have had to correct things like figures appearing in the text 2 pages before they are first discussed, tables being mis-formatted, and place names having their first letter changed to lower case. One time I spent nearly a week persuading a copy editor to reinstate some hyphens when talking about “pre- and post-combustion CO2 capture” because the journal policy was to not hyphenate words.
- A few years ago, you might still be waiting a few months for your paper to be published once it has been accepted, but this is less of an issue nowadays, thanks to journals being online. Having said that, some journals are still oversubscribed; one of my recent papers was accepted back in 2016, but not officially published for almost a year, In the meantime, the unformatted accepted manuscript was available online, but I had to wait months for the journal-formatted version.
- So, finally – your paper is out. It has passed peer review and been published. Congratulations! But the journey to success doesn’t stop here. You need to do this again. Lots. And the papers you have published need to be impactful – other scientists need to read them, use the information, and cite your papers. But that is a whole new chapter to think about 😉
There are so many things out of your direct control, that can go wrong or right during research. Skill takes you a long way. But you also need luck. Or failing that, a LOT of resilience.
My PhD fieldwork involved collecting rock samples from subglacially erupted lavas in Iceland, which involved climbing up 100 m high scree slopes; for every 3 steps up, I would slide 1-4 steps back down. So in many ways, my PhD fieldwork was a fitting metaphor for scientific research. Similarly, in many ways, it is like the game Snakes and Ladders.
A sense of humour and feeling of solidarity is one of the best things to help maintain resilience. So, to help my fellow researchers, I created Snakes and Ladders: The Science Edition.
The full file is designed to be printed in landscape orientation on A3 paper. There are even science-themed counters to cut out and play with. If the above image won’t download in a high enough resolution, you can download a full resolution version here (select “download” from the top right hand corner of the window).
So, if your research is getting you down, take a break. Play a game, and remind yourself that you need to keep rolling that die to move forwards.