A crisis is when you can’t say:
‘let’s forget the whole thing’.
Having just finished my dissertation thesis it is time to look back on the challenges a dissertation provides. I think that there are three main crises: a beginning-dissertation crisis, and mid-dissertation crisis, and an end-dissertation crisis. Having already written one posting about the difficulties of doing a PhD thesis, I going into the crises in more detail here.
Beginning Dissertation Crisis
Being overwhelmed by the demands of the job
Description: You have a diploma or bachelor or master and now you’re working in your domain to do research. You think you should be able to work effectively as a PhD student because you did study the topic but it’s harder than expected.
What breaks you: Not taking the time to learn the basics of practical research. You will have to learn a lot, probably more than you did while studying, and you need time and energy for it.
Possible Solution: Realizing that studying a topic is mostly learning facts and theoretical knowledge about methods and techniques, but being a PhD student is a qualification (= training) to actually do research and apply these techniques (i.e., something completely different). Accepting that doing a PhD thesis is a learning process and that it is natural to make mistakes and not know everything about it, is an important realization. Note: This does not mean that it is okay to deliver bad work, but it means that having to learn how to do research efficiently and effectively takes effort and time.
Not knowing which research question to examine.
Description: Everyone else is following a specific topic and working hard on it, but you still don’t know which research question will be in the focus of your PhD thesis.
What breaks you: Switching from topic to topic, asking around for interesting research questions, never getting to know a topic deeply enough to develop your own research questions. Dropping a topic as soon as the first problems occur, in the hope of finding the golden topic where everything runs smoothly because you see this as a sign of a good topic where you are competent. Avoidance of the whole “what’s my thesis topic” issue is also deadly in the long run.
Possible Solution: Unless you continue with your diploma/bachelor/master topic, finding a research question takes time (in my case one year until I knew which topic to examine and how). You have to get familiar with the area you are working in and that takes time. This does not mean that you can slack off, but it does mean that working hard and not finding a research question until 6 to 12 months have passed is natural. Ask the ones who just finished their PhD thesis how long they have taken to find their topic and make sure that you choose comparable cases — PhD thesis is not PhD thesis. It is also normal that topics turn out to be more complicated than originally thought. After all, it is research and if it were simple someone would have done it already and it would be without merit for a dissertation thesis (no titles for replications ;-)). Take time to learn the domain, discuss your ideas with others. Make sure you pick a topic that you like, is manageable (usually, PhD students want to solve all the problems in one study which is a recipe for disaster) and that you could do yourself. If you are dependent on other people, their motivation will always be less than yours and they might leave you standing in the rain when things get tough (and they will get tough).
Trying to learn everything.
Description: Starting a PhD thesis in a department brings you into contact with the departments resources. Usually, when a lot of smart people are working together, there are a lot of interesting things one can learn or sample. New statistical methods, new technology, interesting research topics, and much more.
What breaks you: You stretch yourself too thin. You start to learn everything and anything and end up knowing nothing good enough to really do something with it.
Possible Solution: Realizing that you cannot learn everything. The time of universal experts is long gone. You have to set priorities what to learn and when. And you will have to delegate some things. For example, a colleague tried to learn R (an open-source statistical program), Objective-C (programming language for iPhones/iPod touch’s), do the studies described in an DFG project (German national science foundation project), find her own thesis topic and establish herself as a speech trainer. That’s way too much. Yes, the offers are there, the possibilities to learn very interesting things (like R) can be tempting. But in the end, you have to prioritize. Ask yourself: What is crucial for your thesis? The topic, yes. But besides that — you do not need to learn everything if you can delegate it. If you want, for example, learn how to do the programming of a mobile device yourself, delegate it first and then look at the program to understand how it was done. If you already know how to do statistical analysis in SPSS, don’t learn R when the time is short. You can first do the calculations with SPSS, then check them later with R if you have time to learn it. In most of the cases the results will be the same and this way you can make sure that you did use R correctly.
Having no time for the dissertation project.
Description: You are working 24/7 but despite all your work you do not make improvements in your PhD thesis. You are working for thousands of other projects but not your PhD thesis. You supervisor is confident that you will manage it, but you are not so sure.
What breaks you: Not setting limits and saying no. It is hard to reject work offers or assignments, but if your supervisor has many PhD students and does not know how much work is involved in the tasks, he might easily crush you under work orders, first leaving you no time for your PhD thesis, then leaving you no time for any life at all.
Possible Solution: Many supervisors underestimate the time it takes for beginners to do a study. Proposals for the national science foundation are written in a way that experts could do the study in the assigned time. But beginners (like PhD students who do the actual work) need more time. Supervisors often do not see this and they need explicit and direct information about the time it takes to do a study from a PhD’s perspective. Unless they know this, they cannot assist in doing the work. And a PhD student needs assistance. But what a PhD student need to do more is to demand the time needed for the PhD thesis. Make sure you (try to) defend the time budget which you need for your PhD thesis and separate it from the work you have to do for the institute (making an agreement to work always the same three days in the week (e.g., Mon-Wed) for the project and the other two days (e.g., Thu-Fri) for your PhD thesis might work). If you get more assignments than you can handle in that time, ask which assignments have priority (do a forced choice à la either-or, don’t accept a “both” or “all of them”). Keep a list of the tasks you did and the time it took (you will need it anyway when you want a recommendation for another job). Also, do present your PhD thesis work frequently, so that your supervisor sees that you didn’t slack of in the other half of the time you work but actually used it to further your PhD thesis work. In comparison to a fleshed out proposal the PhD thesis is like a newborn baby fighting against a cage fighter. It will loose unless the PhD student actively demands time for the thesis. Do not accept arguments like “Just do the next planned study and then we will have time for your PhD thesis.” Unless the project for the national science foundation and your PhD thesis run parallel you will not be able to finish your work and in the end, you will have worked three years for meager pay without receiving a PhD.
Avoiding conducting the first study.
Description: You have ready countless articles, you have designed studies (kinda), but you do not feel ready. You read another article, and another. You change the design of the study, talk to others about it, change the program, change the question. The only thing you do not do is the damned study.
What breaks you: Thinking that the first study must be perfect. When you read the articles about a topic you will inevitably notice that the topic is more complex than you already thought. You might even find papers about studies that did (almost) exactly what you wanted to do. You continuously find new possible moderating variables. It’s easy to think that you’re “wasting a study unless you’re sure about it”.
Possible Solution: Do the study, at least a pilot study, even when you are not sure that you have covered everything. Because you never will think that you have covered everything. The only way that would happen would be if you find a topic where the literature pieced together would solve the question without any doubt, but then, you would not need the study anyway. Research is working under uncertainty. There is always the chance that a study turned out to be a “waste of time”. But this rarely is the case. It will give you experience in conducting experiments and analyzing the data, point you to other important variables, lay the groundwork for further studies, and give you an idea whether you can find gold or only turn up dirt.
And no, there is no perfect study. You can criticize any study, no matter who the author is, because we are all human, we all make mistakes, and even if we would not, our equipment might malfunction (or the study participants). Doing the first study will give you a good idea what others will criticize and you can amend these concerns in the next study, and the next, and the next …
The only time when you must be very, very careful is when you are using up resources that you cannot replace. Money for equipment or participant compensation, a sample that can be used only once (e.g., a group of people who have never watched TV), or the like. In that case you should work together with an expert on that study (or do studies with other samples until you are the expert).
Having no significant results.
Description: If you have decided to examine a particular question, formulated hypotheses and conducted studies, the worst thing possible are results that are not significant. Treatments that show no effect because groups do not differ, correlations below significance, or insignificant effect sizes.
What breaks you: Thinking that because your hypotheses failed, you did too.
Possible Solution: First of all, welcome to the world of research. Unless you are working in a well-researched area, it is quite natural to find nothing. It’s research. Next, make sure that you really did find nothing. Normally, when you check the results in more detail (or segmented to different groups) it is possible to find significant and interesting results (that are not random but actually make sense). This does not mean that you should conduct post-hoc research (i.e., doing the study, than claiming that the result is what you expected), but that you should “make friends with the data” and find out what they say beyond the hypotheses. In most cases you will conduct more than one study and the results of the first study will give you valuable hints where to dig to find the gold.
Having no time to work on your thesis.
Description: Once you become established in a work group, you will notice more things you can do. Also, other people will give you more chances and opportunities you (feel you) have to use. From participation in project that are only marginally related to your PhD thesis to conference trips to teaching duties. During the PhD thesis, time can become surprisingly short even after you have learned the necessities to deal with your topic.
What breaks you: Loosing the focus on the reason you are there — doing your PhD. That always has to come first.
Possible Solution: As usual with creative projects: Set Priorities. Your main focus on the 3 to 6 years for your PhD thesis is getting the job (the PhD thesis) done. Sure, having an eye on possible future topics (after all, the PhD thesis will be done one day) and taking unique opportunities is crucial. But make sure that carefully evaluate non-critical opportunities in comparison to the time it will cut from your PhD thesis budget. For example, when I realized that conference trips took a lot of time, I stopped submitting papers for conferences until I was finished with writing my thesis. I might have missed a few interesting conferences, but at least, I have finished my thesis.
You got scooped.
Description: After conducting a study you notice that someone wrote an article about exactly the same topic.
What breaks you: Thinking that your work is worthless because someone else has published the same results.
Possible Solution: Have a very close look at the article, at the study, at the theoretical background, at the discussion and conclusion. Unless you are working in a highly deterministic domain (like mathematics, physics, perhaps also chemistry), it is almost impossible that the work is exactly the same. Usually there are differences to the work you did. If not, it is either a very strange incident or a case of plagiarism (in which case a social network site that shows you your connection to the researchers via people you know might come in very handy). Others might not see these differences at first glance, and in your article you will have to point them out. And although it is very unsatisfying, it might also be possible to get a PhD thesis even if someone else did exactly the same work — if you are in the end phase of your work and it is a clear case of parallel creativity, it would be very unfair not to give you the title. After all, you did everything you needed to do.
(BTW, PhD Comics has a very good series about being scooped: Strip 1, Strip 2, Strip 3, Strip 4, Strip 5, Strip 6, and Strip 7.)
You project partner quits.
Description: You work in an applied setting, with a museum or a commercial organization, or you need the specialized equipment of a certain institute. Suddenly and without warning, your project partner, who you need to conduct your studies, ceases to work with you. They have good reasons, e.g., lack of time or resources due to budget cuts or reorganization. But that doesn’t help you.
What breaks you: Thinking that just because a project partner quits you cannot finish your PhD thesis and have to start all over again.
Possible Solution: In many cases there is more than one way to examine a question. And often there is an alternative to the way you have planned. It pays to keep this in mind at the beginning, have possible alternatives if something goes wrong, but even if this happens later, it is usually a setback and does not mean that you must flush the work of 2+ years down the drain.
Questioning the value of the research done.
Description: You have worked for three to six years on a topic. Now you write it together in a PhD thesis or prepare your thesis defense and you realize: Damn, is that all? I’ll never get my thesis for that. I’ve made too many mistakes. Or, even worse, the world isn’t better because of it.
What breaks you: Loosing perspective. There are thousands, millions of researchers. Your work is only a small part, but that does not mean that it is not important.
Possible Solution: There’s an old saying that you do not to see the wood for the trees. Your thesis (probably) won’t save the world. It’s only a small part of in the wider range of things. That’s okay. It’s also okay that you made mistakes. It’s a qualification work, you are supposed to learn in the process and you can’t learn without making mistakes. This is no justification for sloppy work, but even if you (later) find out that another solution would have been better, normally you had good reasons for doing something in a certain way. Hold tight to your original reasons. When you defend your thesis tell them why you did decide in a certain way and what you have learned in the process. Other researchers probably wouldn’t have acted differently.
Fear of getting slaughtered in the thesis defense.
Description: You have made a lot of mistakes during your PhD thesis. Unless you are micromanaged by an expert and work in the area of that expert, that’s perfectly normal. Like said before, it’s a qualification work. Nevertheless, when the thesis defense comes near, the fear of getting slaughtered can become overwhelming.
What breaks you: The enemy in your mind. You know your work better than anyone, you know exactly (afterwards) where you made your mistakes. And like Sun-Tzu said, you need to know the enemy to take him down. So you can easily attack all your arguments yourself and destroy everything you did.
Possible Solution: First of all, unless you work in the domain of another research who attends your defense, you are the expert in the domain. You have worked in that area for three to six years, it is very unlikely that someone knows more about the specific area of the domain than you do. But is is possible that someone will ask you about the method you applied (which they probably use themselves) or about the relation of your findings to the topics they know. Some might even ask questions that are totally unrelated to your work. You do not need to give a perfect answer to every question. They usually want to see how you handle yourself with difficult questions, and saying that “it is a good question, I have to check my data for that” or “that’s an interesting suggestion, but would require another study to answer” is also acceptable.
There are many crises during a PhD thesis, it is demanding work, it is a qualification work on a very high level. But most of theses crises are manageable. BTW, if you are troubled about writing your dissertation thesis, I highly recommend the posting: How to Write a Dissertation Thesis in a Month: Outlines, Outlines, Outlines.
Did you experience different incidents or found different solutions? Write a comment.
Not a science major, but I’ve been dealing with some dissertation crisis myself. This helped me a lot. Now if only my chair will agree with you….
Thanks anyways! Will be tweeting your article for my future references!
thank you for the nice feedback … if you are interested in more information, you find information about doing a PhD thesis on pages 328-335 (Acrobat pages 165-168).
All the best
To the point.I used to try to learn R and write everything in R for analytics for my thesis although I knew Matlab better. Now I’ve delayed learning R.
Good example, I love R, esp. as an alternative to SPSS (which has … some problems). But I also still use SPSS. One way to learn about R is to do analyses both in R and [the program you are familiar with], but still, it takes time (more than I have right now). While it is certainly a good program and well-worth learning, when you are in the middle of your PhD and you are swamped with other stuff, this might not be the best time. Redoing your analyses when you have time (if that ever happens) might be a way to learn it though.