Report Proposal

This proposal should also include all code that you use to generate descriptive statistics like the mean of a value or plots of the data. Though this is not a writing course, you need to proofread your papers before submitting them. If you need help with writing, there is a writing center on campus that provides this service for free or just ask a roommate to read the paper quickly. If they can't understand what you're talking about, then I probably won't either.

The proposal is not a final decision on your paper topic, but it probably should be. While changing your topic is something you can do up until you submit the paper trying to do this at the last minute will be difficult.

A good proposal should include what I mentioned above, as well as the following:

1. Your Question

What are you planning to address? Is this question clear (can you ask it in a ten-word sentence)? Is this question interesting? This should probably take a paragraph or two. This is a prototype for your introduction in the final paper.

2. The Concepts and How You Plan on Measuring Them

Your question will contain concepts. These concepts should be something that we can measure (it shouldn't be something entirely subjective). Explain these concepts, explain the measurement, and say whether the measurement makes any sense. For instance, if your concept is something like "wealth" and the measure is number of apples consummed in a year, I probably won't believe you or give you credit for the measurement component of this proposal.

3. Outcome, Explanatory Variables, Links

You should identify one of your measures as an outcome variable (say in a regression model) and one or two variables as explanatory variables. As we move into hypothesis tests, try forming a prediction about how your explanatory variable(s) will affect the outcome variable. In general, this will coming in the form of "Cases that differ on X will differ on Y." or "X will increase/decrease/have no effect on Y." A really good proposal will also include a brief explanation of why this occurs.

4. Dataset and Citation

You need data. You need it for this report. It you can't find, upload, or understand a particular dataset then do not wait to ask.

You should have a citation for your dataset, as well as an explanation of how they collected the data. Was it a survey? Are the numbers from a particular organization? Can we believe these numbers or did someone make it up?

5. Description of Data

What type of variable is your outcome variable? What kind of variable is your explanatory variable/control? Recall the early class periods and readings where we discussed ordinal, interval, and categorical variables.

Describe the data appropriately. What is the typical value for your variable? What about its dispursion? There are not set questions for this description (this will depend on the data that you choose), but you should use the techniques you learned in class.

A really good proposal will include a visual description, like a scatterplot, histogram, or box plot. Graphing the relationship between your variables will be required in your final paper, so if plots are difficult for you, this is the time to address the issue.

It is not necessary that you fit a linear model for the relationship between your outcome and explanatory/control variables, but you can. You'll be doing this for the final paper, so you may want to push yourself to do this in the proposal. This way, there's less work to do in late April when you work on the report itself.