Lab 2-3 (Development and Quality of Life) entails the construction of a large spreadsheet from scratch. If you have plenty of time (as in my case when I worked at Saint Anselm’s) you can apply it as a cooperative assignment in which students are assigned different sections. This does not work if you are pressed for time (as I am in my current job), so I used a different strategy: I took the students to the computer room and had them navigate the CIA factbook website to create only the first two columns (country and population). Once I was satisfied that they had learned how to incorporate this data into a spreadsheet, I simply gave them all the sample data from the instructor’s guide (gathered in 2007) and instructed them to use this to make their six scatter plots. As a result, nearly all the students were able to complete the entire assignment well within the 3-hour lab period. Even though the scatterplots they made were not based on the data they gathered on their own, the main point of the assignment was not compromised.

Understandably, both students and instructors will feel overwhelmed by the scope of this exercise in its original form, so this serves as a reasonable compromise because students still get some experience collecting and compiling data. If I get the chance to do it again I will have the students collect data only on infant mortality and per capita GDP (instead of population). I will then require them to make a scatterplot of this data that they gathered first-hand. Only then will I give them the 2007 data from the instructor’s guide. I would also make them hand in this scatterplot they made from the data they collected because would provide tangible evidence that they learned to make at least one small spreadsheet from scratch.

This abbreviated protocol can also be applied to two other labs involving large spreadsheets, Labs 2-4 (Risk Factors and Health Outcomes in the U.S.) and 2.5 (Risk Factors and Health Outcomes in Africa).  It does not matter whether or not they finish during the lab period because once they learn to make and interpret at least two graphs in class, they can do the rest on their own time.

Most of the scatterplots will have plenty of "noise" (points that fall nowhere near the pattern) and some will have no pattern at all. A good example of an amorphous scatterplot is per capita GDP vs. per capita oil production. College students make this plot expecting to see a positive correlation. Those of us who are more seasoned know that per capita GDP is more often the result of what some refer to as "intangible capital" (education, government policy, financial institutions, social stability...). In effect, there is as much to learn from the scatterplots that have no pattern as from those that do.