The learning objectives of “Organizing, Summarizing, and Interpreting Data” go beyond learning technical aspects of statistics. They also include getting you to think critically about how to use data to analyze business problems. Textbook problems in statistics generally give you data and then ask you to perform a specific task such as computing a standard deviation or estimating a regression equation. Important as being able to solve them is, those problems capture only part of data analysis. Doing actual data analysis requires deciding what data to collect (or use) and what statistical techniques to apply. Moreover, even when someone includes data analysis in a memo or paper, the material included is typically a small fraction of all the data analysis that the researcher performed. Using data analysis effectively requires careful selection of what analysis (based on what data) to report in order to advance an argument or support a conclusion.
The National Bureau of Economic Research (NBER) is, as the name suggests, an organization that engages in economic research. NBER research associates are generally economics professors who publish articles in academic journals. NBER circulates early versions of these publications in its “working paper” series. Most NBER working papers are from 20-100 pages long and require extensive training in economics to understand. The home page for NBER is www.nber.org.
The NBER Digest is a monthly publication with brief descriptions of NBER working papers. The descriptions are supposed to be sufficiently non-technical to be accessible to people who are not professional economists. (In deciding which working papers to include in the NBER Digest, NBER seeks to include articles that are of interest to non-economists.) Current and past issues of the NBER Digest are at www.nber.org/digest/.
The articles in the NBER Digest generally contain a graph or chart. The purpose of these assignments is to get you to examine carefully the choice of the graph or chart, to describe how the authors constructed it, and then to explain how the graph or chart advances the main point of the article. In business, you are going to have to present your data analysis to people who are not data analysts themselves and, even if they are, do not have the time to examine all of the analyses you ran. The summaries in the NBER Digest are the type of exposition that you will have to do. Just as careful reading in general helps you learn to write, reading summaries like those in the NBER Digest and thinking about how they use statistical analysis to support a conclusion will help you learn how to present empirical effectively.
You are to turn in reports based on two NBER Digest articles. The NBER web site has the archived issues going back to 1997. (Go to the bottom of the page to see the earlier years.) Scroll through the titles and find topics that appear interesting to you. For most of the articles, the graph or chart is a scatter plot with a simple regression line. You should use one of those articles for your second report. Your first report should be an article in which the supporting graph or chart is not a scatter plot and/or regression equation. Your report should answer the following questions.
- What is the main point of the article? You may copy verbatim from the article as long as you put what you copy in quotation marks. Whether you copy from the article or use your own words, your description should be concise.
- Describe the graph or chart that accompanies the summary. How does it support the conclusions of the article?
- Describe in as much detail as you can the steps one would have to go through to produce the chart or graph in Excel. If you can tell, what are the underlying data? If you cannot tell what data the authors used, what data might you collect to reproduce the results? (In many cases, the graphs rest on calculations that are more complicated than you can figure out from the brief summary. In such cases, describe how you could collect and process data that would result in a chart or graph like the one in the NBER Digest.)
- Were there any alternatives to the graph or chart that the researchers might have considered? What were the advantages or disadvantages of the choice they made?
- Do you find the use of the chart or table convincing? Do you have questions about the results?