Statistics can be pretty boring in the abstract. Examining a real
world issue shows the value of applied statistics. The project will
be evaluated on: usefulness— what are the benefits for business, policy,
or consumers?; ambition—what was the scope of this project?; methodology—
was the study carefully designed?; statistical analysis—were appropriate
techniques used?; communication— is the study clearly presented?
Proposal
The proposal should identify the topic of the study and major research
questions you are interested in answering. It should include a discussion
of who would be interested in the results of your study. You should
also discuss your initial plans for collecting data. If you choose to collaborate
with a classmate, you should submit a joint proposal.
Paper
The paper should contain the following sections. I’ve included
some questions you should consider when you write each section.
1. Introduction: What was the topic? What was the reason for this study? What research questions did you try to answer?
2. Description of population and sample: What was the population of interest? How was the sample selected? How random was the sample? Is the sample representative of the population? How was your data collected?
3. Statistical sections:
a. Basic descriptive statistics, include tables or graphs
Discuss your findings.
b. Confidence intervals for important variables
(at least one mean and one proportion)
Explain the meaning of each interval.
c. Hypothesis testing of means or proportions.
(meaningful tests for at least 3 variables)
Include references to sources.
Clearly present your hypotheses.
Explain the results of your tests.
d. Hypothesis testing of the difference of means or proportions
(at least 3 tests comparing groups)
Clearly present your hypotheses.
Explain the results of your tests.
e. Multiple regression
(Estimate at least 2 regression equations.)
For each equation, explain why the explanatory
variables were chosen and your expected sign (+, -) for each
coefficient.
Interpret your results, including statistical significance.
4. Conclusions: What does your statistical analysis show? What limitations are apparent in this study? How might further studies address the research questions?
5. Appendix: copy of survey, copy of data, computer printouts, references
Website
The Website presents a clear, non-technical discussion of your study
and your conclusions. You can include a link to the full-text of
your paper, for readers who desire more detail, or more technical information.
The website will be the main way of communicating your results with people
who can use them.
back to Quant II homepage
Department of Economics homepage
Questions, problems, or comments?
email: cstull@kzoo.edu