If anyone had a plan at the beginning of grad school, it was me. When I began my program, I expected to finish in four years and start teaching as quickly as possible. To make this happen, I did everything I was supposed to do: chose what I thought would be a “straightforward” project, made a timeline, and did my best to keep up with the scheduled benchmarks in those early years. Seven years later, I look back and see that none of my original plans worked out, but I did learn lessons and gain experiences that I never could have anticipated in the beginning.
This story is common among PhD students. Many of us begin our programs intending to finish in 4-5 years, only to find ourselves in grad school several years longer than expected. Well-intentioned friends and family ask us when we are graduating, with confusion over why an academic program would take so much longer than planned.
So, how can we explain this? Why do some students take four years to finish a PhD, while others – in the same program – may take up to eight years or more? Of course, multiple variables are involved here: some students complete their program while working full time, and others deal with personal or family situations that temporarily take them away from their studies (just to name a couple examples). In the sciences, students may earn a master’s degree before beginning a PhD program, or they may enter the program with just a bachelor’s. But while everyone’s exact circumstances are different, I think that one factor is responsible for most of this variation: each student is doing an original research project. This means that no two projects are the same, and that any given project has never been done before.
Because PhD projects involve new research, it is *impossible* to know – with 100% certainty – how things will turn out. Sure, students can work carefully with their advisors to plan projects that have a good chance of success, perhaps by utilizing established methods or writing goals that are based on promising preliminary observations. But any good PhD dissertation, at least in the sciences, still contributes something new – and you cannot be completely sure of what that will look like.
My project changed a lot between my first graduate-level research proposal and my dissertation. Professors told me early on that this is normal, since the project evolves when you begin trying different methodologies and collecting data. Still – depending on the scope of the “changes” – these can significantly alter the direction of your research.
It can be frustrating to deal with these changes while they are happening. However, some of my most important discoveries came about through apparent “problems” and adjustments to my plans. I’ll describe two examples here.
Restricted Outcrop Access Leading to Better Exposures
During my first summer of grad school (2013), I spent almost two weeks doing field work at Coconino Sandstone outcrops near the town of Seligman, Arizona. The outcrops themselves were on public land, but the road to access them went through a private ranch. At the time, this ranch had a system in place by which visitors (mostly hunters) could sign out permits to enter the gated dirt roads.
The following summer (2014), I returned with a new field assistant for three more weeks of work at these sites. We drove to Arizona and set up camp at a campground in Seligman, but when we tried to check out a ranch permit – as I had done so many times – we found that public access had been restricted since my last visit. After multiple phone calls and conversations with locals, it remained uncertain whether we could gain permission to enter the ranch.
Sitting in dismay at a coffee shop, I spoke to my advisor on the phone, and he suggested that we drive to the nearby town of Ash Fork to see if we could find any helpful exposures in that area (quarrying Coconino “flagstone” slabs is Ash Fork’s primary industry). So that afternoon we set out with a smartphone map, not knowing if we’d even find anything.
As we drove north of Ash Fork, I spotted a clearing on the satellite map which I thought might represent a sandstone exposure. “Primitive” is a polite word to describe the road that led there, but somehow we managed to make it in our rental SUV. Still not knowing what we’d find, we got out of the car and explored the area, and ended up discovering an amazing outcrop. Finally – I thought – things might be turning around.
Then, as we were leaving, we got a flat tire.
So we had lost access to my original sites, found a great new outcrop, but were now unable to return to that outcrop (the road was really rough, and we didn’t want to take a chance with another flat). Since we couldn’t do anything else, we went back to Ash Fork to see if we could find any more good exposures. If we found one, there must be more, right?
Needless to say, we did not get access to my original sites or return to the “flat tire outcrop” that summer, but the new sites we discovered north of Ash Fork ended up being some of the best of my whole project. We wouldn’t have even gone looking for them (at least not that summer) if we had followed our original plan, so – while the permit issues were distressing at the time – this is a clear example of how a seemingly bad situation can lead to a better result.
Tedious Data Collection and Lab Challenges Resulting in New Methods
One part of my project involved collecting textural data (e.g. grain size and sorting) from the Coconino Sandstone. The easiest way to do this is to break down (“disaggregate”) the rock into individual grains, and then characterize textural parameters from the loose sand using sieve or laser diffraction particle analysis. The Coconino Sandstone, however, is well cemented in many outcrops and difficult to disaggregate. In fact, for years into my PhD program, I believed that disaggregation would be impossible. Unfortunately, the only alternative was to manually measure individual grains under the microscope (in thin section images).
After measuring thousands of grains, I couldn’t take it anymore, and I thought, “There has to be a better way.” They say that, “desperation is the mother of invention,” and accordingly, this began a quest to develop more efficient methods for characterizing textural trends in my sandstone.
Even though I had always believed that the sandstone could not be disaggregated, I noticed that some samples from one study area were more poorly cemented and readily crumbled apart by hand. So I thought: “If I could cut thin slices from these samples, maybe they could be disaggregated for textural analysis."
To my excitement, this worked fairly well, and I was able to disaggregate and analyze loose sand from these samples – but the well-cemented outcrops were still off limits in my mind. However, I began to notice that cutting the “soft” samples thinner increased the likelihood that they would disaggregate successfully, and one day, I had a crazy idea: *maybe* I could disaggregate even the well-cemented samples if I cut them thin enough.
While I still doubted this idea, I started to prepare a test sample. Appropriately, our old sonicator probe – which I needed for the methodology – broke on that exact day. After waiting another several months for a new one, I tried the method again and, long story short, it worked. Against all odds, we now had a methodology for disaggregating and collecting textural data from my sandstone, which impacted the rest of my PhD program and led to the first published journal article from my dissertation.
Throughout the process, I knew that I was taking some risks. Equipment delays and general experiment challenges took time, and I sometimes wondered whether the efforts would ultimately be fruitful. However, every time I almost abandoned this quest, I would notice something that motivated me to stick with it. At the end of the day, the trial and error of developing these methods really taught me what it means to solve problems as a scientist.
So the overall lesson here is this: when things go “wrong,” treat it as an opportunity to learn. Sure, you may not be following your original plans, but maybe those plans are not the best solution for the question you are trying to answer. In short, don’t be afraid of or surprised by the research “problems” that you’ll face in grad school. After all, apparent problems are the reason we have research projects: if everything was already figured out, we would have nothing left to study.
Comments