In theory, there is no difference between theory and practice. But, in practice, there is.
– Walter J. Savitch
I love going out and experience nature. My studies were very practical with lots of field trips in soil science, botany, geology and faunal life and I spend heaps of my free time watching birds and other critters. Despite these practical experiences were the ones that last longest from what I can remember from my studies, and despite that my own interest in watching all kind of organisms in situ led to a profitable career as a freelancer, I decided to skip fieldwork for most of my research before and during my PhD.
The reason was easy to explain. To me it was less risky to work with existing datasets to answer my research questions. Fieldwork can fail greatly. There are myriads of reasons where unexpected circumstances preventing you from getting the data you need for your thesis. A bad season, exceptional weather conditions, technical issues, poor study design and no time to adjust it, no Plan B. Many (but not all) reasons vanish when working with existing questions – and it is easy to go for alternatives when something doesn’t work out as expected. Ok, maybe your question cannot be answered with existing data and you are prone to sample primary data yourself during your PhD. But maybe you should instead opt your initial questions in a way that existing data can provide the answers first place.
I still think that focusing on research capitalizing on available information during the PhD is an advantageous strategy to optimize outcomes for a good start into your academic career. For at least three main reasons:
- Publications: While others spend months for collecting data you can already start writing papers – and it’s more likely you will end up having published more than your peers spending ages in the field. In a publish-or-perish system that is an important issue to consider (but more on that in a later post).
- Money: You don’t rely on extensive funding to realize your work. With existing data at hand you don’t need to invest in getting your own data. Ideally, all you need is money for your own salary – and some workspace.
- Time: You not only have more time for your writing but also to finish you PhD in time. In Germany the time funded to acquire a PhD is often restricted to three years but on average German PhD students in natural sciences need 4.3 years. This sets up a lot of pressure to PhD studies if they don’t want to run out of money before their defense. In other countries the funding period is one or two years longer – which is a good thing. But in Germany you need to think twice if you can take the risk of getting your primary data when you only have two field seasons available (you need to write up in the final year to defend in time). Personally, this was my main motivation to strive for a topic that can be worked on using available datasets.
But is being efficient in your PhD by operating on existing data really makes you a better scientist too? I doubt so. The reality is that to cherish available datasets and to get a feeling of possible error sources you need the experience of sampling your own data. And you need to refresh this experience from time to time.
There is a decent chance that you will become routine-blinded if you get stuck behind the screen!
I spend years in collecting data for a population genetics project that later turned into my Master thesis; and I frequently send my bird sightings to ebird which makes me aware of the error sources when one wants to use those data later in spatial models. After my abstinence of obtaining practical data for almost six years I now returned in situ to launch a 3yr project on road edge effects on biodiversity where I’m Co-PI. Next to refreshing known or almost forgotten challenges that comes with fieldwork I notices a couple of things:
- Planning is not doing. You can brainstorm as much as you want with your project partners. Once you are in the field you experience all the (unforseeable) challenges that hide from you when sitting behind your office screen. I don’t want to go into the details of my specific project but only after you spend time out there for sampling you really see how much time it really needs, what equipment or technical setup actually works or how to make your sampling more efficient and less error prone.
- Once in situ there comes the ideas. Of course you start a project with a clear idea in mind that translate into research questions and testable hypotheses. But once out there you see lots more questions popping up – perhaps more relevant ones or ones that are detached from your project but rely on the same data. Either way, if you need to supervise more students your number of potential topics can be easily increased just by going out and ascertain new data.
- You detach from your desk-based academic routines. Writing papers and grant proposals, review journal articles, teaching, exhume yourself from ever increasing administrative work – all that more and more becomes a routine in your academic career. If you are lucky to work in a discipline that allows you to go out for data sampling you have a great opportunity to detach from those routines and look at them from a distance. I found that fieldwork helps me to restore focus in order to reframe some routines if they become too overwhelming in my work (or even private) life. This is something I experienced already during my freelancing business where I spend days in the field for surveys. Despite being physically exhausting, it was often enough mentally refreshing and I felt more productive and focused afterwards. Seeing fieldwork as a mini-sabbatical could be an important yet overlooked aspect to improve your academic life as a whole, no matter what stage you are at.
- You become a better supervisor. Not only you will have more ideas for possible theses work. You will also better empathize the potential struggles your students have in the field. If you – as a supervisor – are fully detached from fieldwork, even if you draw on lots of experiences from fieldwork taken place decades ago, you will have a hard time to understand the problems that appear to your students. Therefore, always set up the project in situ. That will make you the best possible supervisor in that project (I’m sure the same counts for lab work too).
If you are an ecologist spending regular time in situ you might think ‘Of course! That’s what we do for ages now! Whats the point?!’. Shame on me. But for me that was a big Eureka moment, simply because I thought I knew all this already from my past fieldwork experiences. But there is certainly a lot more to learn and perhaps (for the one more for the other less) there is dementia going on over time. At least that might explain some of my observations.
More and more ecologists focus on in silico work in their main line of research. At the same time I frequently review papers working in silico or with existing datasets that seem certainly detached from the real world. I talked about bubbles before. I see a real danger that future in silico scientists work in their network of in silico scientists celebrating their groundbreaking findings that have nothing to do with the real world anymore. The field of species distribution modeling / environmental niche modeling stands a good chance that its members become more and more routine-blind over time. Even with good intentions (e.g. informing conservation or environmental management) modellers or other in silico researchers will loose connection to the real world, both their systems and their peers from more applied disciplines. To anticipate this development, it’s good to go out for groundtruthing yourself and your research to avoid being routine blinded. No matter what academic stage you are.
To close, even if I think that there are certain episodes in an academic career where you should focus on optimizing your output (the earlier the better, and perhaps the PhD time is best for that), I think that a healthy mix of in situ and desk-based work will be key to improve yourself as a scientist (talking here about ecologists and the like of course).
In practice, there is a huge difference between theory and practice, in theory there isn’t.