Procedure:
We will be using data collected from the historic Oakland Cemetery in southeast Atlanta (
www.oaklandcemetery.com
). This 48 acre cemetery opened in 1850 and reached maximum capacity by 1884. It continues to serve as an active cemetery as plot ownership is passed along family lines or sold back to the cemetery. The cemetery initially served as a public burial ground for the City of Atlanta, and it is home to about 7,000 Confederate and Union soldiers from the Civil War. There are several unique sections of the cemetery, including the original 6 acres, German Jewish grounds (purchases made by the Hebrew Benevolent Congregation in 1878 and 1892), and African American grounds (reflecting a policy of racial segregation in 1866). Outside of these areas, burial plots are owned at random with respect to cause of death or ethnicity.
1. In the prelab, you are asked to review the ‘LifeTableAnalysis.csv’ file and look at what raw data is available. Think about what type of question you would like to ask that involves comparing the survivorship between two groups of individuals using the data from Oakland Cemetery. You practice comparing males and females in the prelab, so you should find a new comparison. For example, you could compare those who were born between 1821-1860 vs those born 1861-1900 (i.e. before and after the Civil War). 2. Once you are in class, form groups no larger than three students (preferably groups of 2) with others that are interested in similar types of questions. Pick your two groups and distill your question down into a research hypothesis, record it, and have a TA check it before proceeding.
Remember, your hypothesis should involve comparing the survivorship of no more than two well-defined populations. 1.
Two groups being compared:
2.
Research Hypothesis:
3.
What variable or variables (column or columns in the datasheet) will you need to use?
3. Use the cemetery data to construct a life table and survivorship curves based upon your hypothesis. To do this, you will need to randomly sampling at least 75-100 unique people per group you are comparing (150-200 total). You can randomly select people from the data based on their line number. You may choose to use an online random number generator and manually generate your dataset, or use R. Your prelab activity notes should be sufficient to guide you through this step. If you failed to take good notes, repeat the prelab to record the needed code.
Stop and Think: What in the code would need to be modified for your data set? For your
specific question? What should those changes be, and how would the code look? Would it be
line edits, or might it impact the order that you do certain steps?
Once you have your line numbers, if the sampled individual fits into one of the groups that you are comparing, then collect their data. If not, generate a new number. Continue until you have
