MAT 442: Final Project
Due Monday, March 15
Find a suitable dataset to work with for the final project by Monday,
March 1. See the end of this PDF for possible places to find good data
sets.
You may work by yourself or in pairs. When working on the project, I
must know what each of the partners contributed to the paper and have a
copy of each person’s objectives and R and/or SAS code.
Requirements of the Dataset: The data must have a minimum of 12
variables with at least 7 quantitative variables and a minimum of 200
cases. If you find an interesting dataset with a good design that does
not meet these minimal requirements, e-mail me and I may let you use
it.)
As soon as you find your dataset, e-mail it to me so I can see if it
will be sufficient for the project. I will then be happy to help you
to outline you procedures and make suer your research objectives are
met. Do not hesitate to ask me for help.
Background Research for Project Topic: Every paper should begin with
appropriate background information about the topic. References used
in obtaining this information should be cited in the paper. You should
have at least five references dealing with previous studies about your
dataset or references about the topic of your paper and possible studies
completed even though different than yours.
Statement of Research Objectives: Once you find a suitable dataset to
work with, you must formulate a set of research hypotheses that your
project will address. The research hypotheses that you create will de-
termine what tests and methodology will be needed from the course to
address these needs. The statements of your research hypotheses can
be included as pat of your purpose or after your descriptive statistics
in the body of your report. As soon as you formulate these objectives,
e-mail them to me so I can make sure they are suitable.
Statistical Procedures to Use in the Report: The procedures to use
besides the numerical descriptive statistics should include at least three
or more of the following concepts with SAS or R:
1
• Multiple regression procedures
• Two-way ANOVA
• Repeated measures ANOVA
• Chi-square testing
• Logistic regression
• Log-linear regression
• Generalized linear models
Guidelines for Submission of Written Report
Be sure to include the title and source of the data.
• Abstract – An abstract is usually necessary because your report may
have multiple readers. Some of them will need to know the details of
your report, including the supporting data on which you are basing
your conclusions and recommendations. Other readers will not need
as many details but will want only the preliminary study information
and main conclusions. The abstract or executive summary should
cover the general subject of the research, the scope of the research,
identification of the type of methodology used, and conclusions. You
can describe the project’s objective, what you have done, and general
findings. It should be limited to no more than 200 words and is often
confined to less than 100 words.
• Background and Purpose of Study – This is an introduction to
the report. Review of literature should be included here, too. The
brief literature review is mandatory for the final project. You most
likely have many goals in mind and you can state these or your research
hypotheses here. It is these research hypotheses you will be addressing
in the paper.
• Sample Collection Techniques, Methodology, & Design (if ap-
plicable) – Describe the data collection process, if you know it. If not,
tell us how and when you acquired the data. Describe the size of the
sample and information about the variables of interest. Include infor-
mation about margin of error for percentages quoted in report based
on sample sizes used. Discuss the design, and if applicable, the power
of the test.
2
• List of Variables, Responses, and Codebook of Variables in
the Analysis (if applicable) – If this is very lengthy, this codebook
may appear in the Appendix.
• Routines Used in SAS or R – This can be in the introduction or
scattered throughout the report. This is up to you.
• The Body of the Report – The body is the bulk of your report. It
is the major part of your paper. In a logical order, present the results.
– Begin with the tabular and numeric descriptive statistical tech-
niques. These descriptive statistics findings should be supported
by tables or graphs. This is followed by a summary of the descrip-
tive findings. Be sure to include margin of error for all percentages
quoted in the reported.
– The next part summarizes the inferential procedures used. These
should correspond to the research hypotheses previously stated.
Comment on the rationale for using certain procedures. For all
statistical tests contained in the report, be sure to state the hy-
potheses, assumptions, test statistics, P -values, and conclusions
found. All results should be in narrative form and supported by
tables or graphs when needed to enhance the explanation. Do
not be afraid to be creative!
• Summary/Conclusions & Recommendations – This section con-
tains a brief summary of your findings. The conclusions are generally
stated first and are the outcomes and decisions based on your research
results. Recommendations, on the other hand, are suggestions for how
to proceed based on your conclusions. List any limitations, if they ex-
ist, and give suggestions for future research.
• Limitations (if any) and Other Variables You Wish You Wish
You Would Have Included in the Study – Be specific about your
findings. How did they compare with your research hypotheses? What
other variables would you have tried to include in the study, and were
there variables you would not use in further studies that were contained
in this study?
• References Section – Include any sources where you may have done
a search on background information on the topic. Include the source of
the data reference site (if not confidential) and other texts or software
you may have used in the report.
3
• Appendix – If you used survey data, a copy of the survey should be
included in the appendix. Some people also put the codebook in the
appendix.
Where to Find Datasets?
The following is a list of sites where you might be able to find datasets. If
you have a dataset from your place of employment that fits the criteria for
this project, feel free to use it.
• https://datadryad.org
• https://www.kaggle.com/datasets
• http://www.statsci.org/datasets.html
• OzDASL – http://www.statsci.org/data/index.html
• DASL – http://lib.stat.cmu.edu/DASL/
• JSE Data Archive – https://ww2.amstat.org/publications/jse/
jse_data_archive.htm
• Datasets from UCLA’s Department of Statistics – http://www.stat.
ucla.edu/cases/
• More datasets from UCLA’s Department of Statistics – http://www.
stat.ucla.edu/projects/datasets/
• U.S. Government Datasets – https://www.data.gov/
4






























































































































学霸联盟