Deskins Research Tips
These are a collection of tips for MQP/IQP/graduate students. Some may apply more than others.
1. Learn to do a Literature Search. This is one of the most important skills you can have. Also learn what makes a Good Scholarly Resource. You'll need to know what others have done and how your work fits into the "big picture" of previous knowledge and research. No research is completely new and original, but builds on previous knowledge and research that others have done. You are really adding another brick on top of bricks that others have already laid. Also, you need to be aware of what others have done so you don't unknowingly repeat others research. Nothing is more frustrating than finding out someone has already done what you've been working on for the last six months.
“Six hours in the library may save six months in the laboratory.”, An Introduction to Scientific Research, E. B. Wilson
2. Understand your problem. What question are you trying to answer? You can waste a lot of time just performing experiments or calculations without really making progress. Think about what you need to do to answer this question and do these things. Also if you understand your problem you should be able to explain your question/problem to someone else like a friend or relative.
3. Start simple. You will not solve your problem in one day, but must work towards the eventual solutions. Students are often tempted to jump right in and try to model the final product or run the last big experiment. They get frustrated when it doesn't work and don't know why. You often need to build up your skills and make sure you are doing the research correctly. Start with a simple system (like small molecules for modeling) to make sure you are doing it right. Get someone else's previous research results and make sure you can repeat it. If you can't even repeat other's work then you are in trouble! Once you are confident in modeling "simple systems" then you can move to more relevant important problems.
4. Be consistent. There are literaly thousands of variables that may influence an experiment (temperature, pressure, concentration, etc.) or simulation (number atoms, box size, basis set, other simulation parameters, etc.). You need to control the variables in order to identify how they influence your research. For instance if two catalysts are run at different temperatures their results may not be comparable since they are under different conditions. The temperature may be the reason for their different reactivities. Similarly, you should keep your simulation parameters consistent from one set of calculations to another.
5. Check and analyze your results often. You don't want to spend a few weeks running some calculations only to find out you've been doing them wrong the whole time and have to start over. If you check your results (look at output files, view xyz files) periodically you can find problems quickly and fix them before you've wasted too much time.
6. Work hard. Research can sometimes be tough. It requires patience and hard work. You will make many mistakes along the way - this is part of doing research. But it is rewarding and like many things you sometimes just have to stick with it for a while to see success. Your eventual solution may require months to years of work, so don't give up!
"Genius is one percent inspiration, ninety-nine percent perspiration.", Thomas Edison
7. Be independent. If you want to be a good researcher you need to learn to work on your own. A lot of your previous classroom work has been very structured, with only one solution to every problem, and the instructor closely monitoring your activity or giving you all you need to know. The "real" world isn't so straight-forward. You will eventually likely be working on a team of independent engineers or researchers, with a boss too busy to "babysit" anyone. You need to learn to make contributions on your own and demonstrate your competence. Doing good research requires intelligence, creativity, and hard work, as well the ability to think "outside the box". If you don't understand something you need to find out why, what knowledge you lack, and how to get this knowledge. Be willing to try new ideas on your own, but also don't be afraid to ask for help from your advisor or others if you get stuck.
8. Learn to share your results. Doing great research is fine, but others need to know what you're doing for it to be useful. You need to learn to share your results with your advisor, fellow team mates, fellow students, colleagues, other engineers, and the general public. Sharing takes place in a variety of ways, such as formal presentations, written work (your final report!), or weekly meetings. Learn to talk to others about your research. See these pages: Making Presentations, Deskins Tips on Tables and Charts, Deskins Writing Tips.
9. Learn computer skills. We live in the digital age and computers are an essential part of research. Not having good computer skills is like trying to cut down a with a dull saw. It's much better to sharpen the saw then cut it down. If you are doing simulations, learn about Getting Started or Linux/Unix. If not, make sure you can use a word processor (like Word or OpenOffice) and possibly spreadsheet (like Excel). Also learn about Citations. The WPI library has all sorts of classes to teach computer skills. Also, ask others for computer help/tips.
10. Make progress. You have limited time to complete your IQP/MQP/thesis/dissertation. Not everything you do will help you get to the finishing stage. Always ask yourself if an activity will help you get to your final goal. If not, don't do it. Run the simulation if you need to. Start writing if you need to. Visit the library if you need to. Always be making progress to that final document!
