Bookmark and Share
|
  • Text +
  • Text -

Instructions for Senior Thesis

Before beginning data collection, remember that you will need to turn in to your advisor the "Request of Review of Research Using Human Subjects Proposal" form. Once you have received approval from your advisor or the Human Subjects Committee, you may begin collecting data for your senior thesis.

Be sure to check the senior thesis deadline page (TBA).
Writing Your Thesis: The Final Draft
We are often asked about the length required for a senior thesis. Your advisor is the best person to ask about this. He or she knows how long yours should probably be. Most of us do not want a long paper. The economical journal format and its concise style is one frequent model for the thesis. Other faculty, and sometimes students, want to include their entire literature review in the thesis. Again, it is best to work this out with your advisor. Please consult the Publication Manual of the American Psychological Association (4th edition) for details about content and style.

The 100-150 word abstract should give the reader a clear picture of the main contribution of your thesis. If your advisor prefers the journal article format for the thesis, the Introduction is not a critical and historical review of the literature; rather it is a vehicle to motivate the research and involve the reader's interest in it. Six to eight pages is usually adequate for this purpose. The Method section does not give every detail of procedure, only the essentials (but all of the essentials) that would allow the reader to replicate your study. The length of the Results and Discussion sections depend greatly on the nature of your study and your data, but there, too, economy is an important goal. You will probably find that a total length of 20-25 pages is adequate space to report your study. You will also find that it is easier to write a loose, careless 40 page paper than a tight 20 page paper. It is certainly worth the extra effort to bring your paper down to an elegant length, in a form that says everything you want to say (and says it well) but without redundancy. Below are a few pointers relevant to the question of length:

Things that lengthen the paper and detract from it:

Irrelevant articles or theories cited and discussed in the Introduction (Ask yourself: Does this point contribute to the thesis I am developing?) Irrelevant details (e.g., color of experimental room, time of day) in Method section.

Material reported in wrong section. If you put results in the Discussion section, for example, you usually need more words to describe them than if you had put them in the Results section.

Redundant or verbose descriptions, explanations, discussion, etc. Ask yourself: Is this word really necessary? Is this sentence really necessary? Can I use a single sentence to say everything in this paragraph?

Things NOT to skimp on:

Report all of your research, or follow-up research you did after your original study. Seniors have sometimes assumed their thesis could report only one experiment. Actually, preliminary and follow-up research can add a lot to the report.

Introduction--giving the overall theory and exactly how it generates the hypotheses tested in this study. Don't make your experimental hypothesis sound like a hunch or an educated guess rather than derivations from a more general theory or hypothesis.

Discussion--explain how your findings relate to the theory.

Some other issues to keep in mind:

Don't report significance of an effect without reporting the data. e.g., "The results showed groups A and B to differ at the .01 level ..."). Report means, standard deviations and statistics (e.g., the values of t's or F's.). See APA Publication Manual for details on reporting statistics.

Don't present complex tables when a figure plotting the data would be easier to interpret.

Don't abandon ship when the data don't "come out." In fact, data are sometimes most interesting when they are unexpected. Don't bail out of the task of interpreting the results just because they are unexpected or marginally reliable. (On the other hand, if your data are obviously noise and nothing more, don't spend pages speculating over a null effect.)

Don't make your experimental predictions sound like educated guesses rather than derivations from a more general theory or hypothesis.

Oral Presentations

Talks will be scheduled for a total of 15 minutes each. Speakers should plan a talk that would take about 10 minutes without interruption and assume that about 5 minutes will be occupied by questions at the end. A convention talk of this length might have 1 or 2 minutes of general orientation at the beginning, about 3 minutes to describe the procedure, and about 5 minutes to present and discuss the results. (Of course, some talks will require a different apportionment of time.) As you can see, the time limitation forces you to be very concise and to get to the main point quickly. There is no time for a literature review. Spend at most a minute on a focused citation of studies that motivate your study very directly. At best, you can explain to the audience what the problem is, why it is interesting, and what you found out about it.

Make sure that the audience understands what you did. If they don't understand your talk, you might as well not have given it. Since you will have only 15 minutes, concentrate on the essentials of your study. Don't waste time on irrelevant details, however interesting they may be. Show good judgment and good taste in isolating the essentials of your study. If you are using a survey or questionnaire, it is a good idea to give your audience a few samples of items so that they will have a clearer idea about the materials used in your study.

Graphics may help your talk go quickly, or at least make everything clearer at a given speed. You should therefore try to use graphs, figures, and diagrams to illustrate your talk whenever possible. Do not feel you have to restrict their use to the presentation of results. Visual aids can be of great use in illustrating contrasting theoretical viewpoints or describing an experimental design. A time-saving trick is to make a figure or table that shows the results and makes clear the main variables in your study and, if possible, its design. You can present this visual aid while discussing your design. Your audience will thus be learning two things at once because they will see the results and the design at the same time. An alternative is to present an empty matrix that outlines the design and then add in the results when it is time to give them. A transparent overlay can make this type of presentation very elegant, but just having a second slide or poster of the matrix with the numbers added is fine.

Oral presentation is an art, a skill that everyone can master. You will do your best if you adapt these pieces of advice to your own personal style. For example, some people do better reading a prepared (and well practiced) talk while others prefer speaking more informally from notes and cue cards. Both approaches can be effective. What is better for you depends on your own style. Remember that in giving a talk you are on stage. Keep the attention and interest of the audience. The best way to do this is make sure that they understand what you did and what you found. give the essential details, skip irrelevancies, and be sure that the audience does not have to guess about anything of importance.

VISUALLY PRESENT YOUR RESULTS! If you have means, present them! Whatever your overall measure, present the actual numbers in a visual form. Never report only a general summary of the results, such as "there were no differences between the conditions" or "the control and experimental conditions differed significantly." It is very frustrating to be told that some effect was or was not significant without being shown what were the means. Do not be obsessed with details of statistical tests; it is usually sufficient simply to say that a relationship or difference was significant--the means and the main features of the results are generally of more interest to the audience than the exact levels of significance. You may want to present data on the effect size.

In order to be sure that your talk will fit smoothly in the time limit, you should practice it beforehand under timed conditions. It would be good to have an audience for your practice to make sure the talk is understandable as well as of the right length. Make sure the audience will understand what you did.

Please ask your friends to come to your talk (and to others'), if only to be your cheering section. We find that a moderate audience tends to improve the quality of both the presentations and the questions asked. Also, we want all speakers to stay through all the talks. It is poor form just to show up for your own talk and then leave when it's over. Let me know in advance if you have a schedule conflict that will preclude your attendance for the whole convention.