Coronavirus Case Study USA Assignment | Essay Help Services
Mini Research Project
(50 points extra credit possible)
Due in Canvas on Tuesday, May 12 before 10 p.m., no exceptions)
Department of Economics Professor Siegler
Sacramento State Spring 2020
To give you more practice with the topics from this class, and to provide some extra
credit (up to 50 points or 10 percent of your course grade), I am giving you the
opportunity to complete a mini research project.
The default topic is the determinants of COVID‐19 death rates, using a cross‐section of
the 50 U.S. states. If you prefer to do something else, please e‐mail me with your topic
to get approval first. Approval requests for alternative topics must be to me by May 5
before 10 p.m.
The dependent variable should be the COVID‐18 death rate per 100,000 population for
each of the 50 U.S. states, and you can find this data at the Centers for Disease Control
(CDC), the New York Times, Johns Hopkins University, and other sources.
The explanatory variables are up to you to decide, but you should discuss and cite
articles that lead you to include these variables. You should also discuss what the
expected signs are of the regression coefficients before you estimate models (e.g., “All
else equal, states with higher population densities should have more cases and more
death rates per 100,000 people because it is more difficult to social distance when people
are living in closer proximity to each other.”). There are some easy ones that almost
certainly should be included in your model, including state population density and/or
percent urban, percent of state population over the age of 65 (or the average age of the
population in each state), the percent African‐American, average incomes, poverty
rates, etc.). This list is not complete – I want you to decide what might explain the
dependent variable. You might want to include things that measure the health care
systems in each state, such as physicians per capita or other measures.
There are also some other interesting things you could examine, such as the impact that
social distancing may have had on death rates (Apple, Google, and Unacast have
measures of how much social distancing is occurring in each state). You could also
examine whether states with Republican governors (and citizens who vote for
Republican governors) have higher or lower death rates (after controlling for other
factors) compared to those states with Democratic governors. Or you could come up
with something else I haven’t thought of yet.
In general, official data sources are preferred, so data from the U.S. Bureau of Labor
Statistics, for example, are better than from some random website.
The goal is to report a multiple regression model (or models) that explains at least some
part of the variation in COVID‐19 death rates across states. You may or may not find
statistically significant and large results, but you should interpret your results, based on
what you’ve learned this semester.
You should also subject your model or models to a series of misspecification tests to see
if one or more of the CNLRM assumptions is likely violated, with a discussion of how
this might affect your results. What is your best model or models and why?
Your final paper should consist of the following components:
Abstract (while this is the first part of the paper, it is usually written last. The
abstract should summarize the research question and main results in a couple of
hundred words, at most. See the “Multiple Regression Examples PowerPoints
for some examples of abstracts)
Introduction of research question and testable hypotheses
Economic modeling/Empirical methodology
References/bibliography using a standard format such as APA or University of
Appendices? (Additional results, detailed data descriptions and transformations,
etc., if appropriate)
There are no minimum word or minimum page counts. Be concise, yet complete.
Finally, you need to attach your entire spreadsheet of data and the R script you used so
that I can replicate everything reported in your paper quickly. Reproducibility is an
important part of the research process. Since much of this data is changing daily, please
be sure to specify the day when your data were downloaded. For example, your
dependent variable may be current as of May 4, 2020. You should report this in your
data description section. Finally I am here to help, so please feel free to ask questions.