In Computer Science and Mathematics, different faculty treat the senior paper in different ways. In what follows you will find some answers I (Walt) give to frequently asked questions. I'm still working on this, and will add more to it as I think of conversations I've had with students over the years.
Q. What is the point of the senior paper?
A. The point of a senior paper is to have the student do some more-or-less advanced research or investigation in a major and write a paper regarding the question and results in the manner of a professional in the discipline.
Q. How do I find an adviser? Is one selected for me, or what?
A. We do not assign advisers in CS or Math. As a rule, each of us has some areas of the discipline where we are more knowledgeable than are the others -- or more interested. Sometimes a student will talk with a prof because (s)he had a course from us in an interesting area. So for the most part, the choice is yours. Talk with several of us if you would like or are not certain. We do have different ways of doing senior papers, so that may also influence your choice.
Q. What are the areas in which you personally direct senior papers?
A. I have mostly moved from computer science to mathematics. My
mathematics specialties are numerical analysis and differential
equations (though I've not taught the latter for several years). There
are other interesting areas that arise in the calculus sequence and in
linear algebra. These provide ideas for mathematics research as well.
During the past year, I have directed a program between Luther
and IBM that involves software testing. That is an area in which
I'd be happy to direct student work.
Q. How do I find a topic?
A. Frequently topics come from the "beyond-the-scope-of-the-course" comments encountered in advanced courses. Sometimes profs will have mentioned possible areas of investigation, or there will be suggested advanced topics in textbooks. There may be topics that had to be skipped in the course due to lack of time, and these topics may appeal to you.
Q. I didn't see anything there that interested me.
A. Examine copies of three journals -- "Communications of the ACM", "ACM Computing Surveys", and "Computer". And remember, this doesn't have to be your life's work; just a major exercise in research and writing.
Q. How broad or specific should I be, and how can I know until I am into the research?
A. Good question. And as you do get into the problem, the breadth question will begin to answer itself. It is very common for the scope of the project to change as a student proceeds.
Q. I'd really like to write a program that I have in mind. Can that become a senior paper?
A. Probably not -- at least for me. This is a research/writing project. It may very well be the case that you write programs as part of an investigation -- say to time different algorithms for performance analysis. But keep thinking research and writing.
Q. I want to write in Computer Graphics (or Databases, or ...) , but haven't taken Luther's course in that area. Is that okay?
A. Not with me.
Q. Luther does not offer a course in Parallel Computing (or Theory of Computation, or ...). I have a real interest in the area and would like to write my senior paper there. Does that work?
A. Yes, if you have taken our appropriate preparatory courses, as Operating Systems in the example above.
Q. I've done an internship in databases, and would like to build on that in writing my senior paper. Does that work?
A. Not usually (for me). The key is that we are doing computer science, not computer technology or computer software. If the investigation fits the academic discipline of computer science, we can make it work.
Q. How long should the paper be?
A. 20 pages.
Q. What form of citations do you want?
A. Use as your model a paper published in one of these three journals -- "Communications of the ACM", "ACM Computing Surveys", or "Computer". If this is a Math senior paper, model the "Math Monthly". And above all, be consistent within your paper.
Q. Luther's library doesn't have most of the papers I need, and I'm concerned about the delay in loans. What do I do?
A. Only research libraries tend to have the journals we need to do research. Luther does not and should not carry them. If it is practical, I recommend a trip to a research (university) library some weekend. The closest are Universities of Iowa, Wisconsin, and Minnesota -- and Iowa State. Many times in the past, students have combined a weekend visit to a friend at one of those places with a trip to their library. It is often an education in itself.
Q. How many sources do I need, and what about Internet sources?
A. I wish you hadn't asked that. But since you did, let's say it would be uncommon for a good paper to have fewer than six or eight sources, and most have more. Web page sources can contain the best available information. They can also contain garbage. Most senior papers I've seen in the past couple of years do have several web citations. The key here is judgement as to quality. When in doubt, let's talk.
Q. I am interested in the Internet, and would like to learn all I can about IPv6 (or I love Apple and want to study the new Mac OS, or ...) so that I can be knowledgeable as I hit the job market. Can I research that and have it be my senior paper?
A. Sure, as long as you understand that such a paper will get a grade of 'C' or lower (for me). It fits a category of paper I call a "book report", the lowest level of senior paper that I accept.
Q. Ouch, that surprises me! Are there other pitfalls I should be aware of?
A. Let's see.
Q. Okay, I get the idea. Are there some positives to look for?
A. The best papers set out to compare two or more "things" -- algorithms, data structures, languages, architectures, operating systems, file systems, ... The comparison can be theoretical (as long as the theory is not just copied from a book or web page), or experimental. The key again is that we are "doing computer science (or math) and writing about it". So any paper that I judge to be excellent is going to involve some kind of intellectual investigation.