Cemetery Demography
Adapted from Flood and Horn, Cemetery Demography, in the Ecological Society of America's Experiments to Teach Ecology.
We learned in the module on population principles that there are three types of survivorship curves. The type I curve is for species where most of the young survive and most mortality is in the oldest age classes. The type II curve is for species where mortality is relatively constant for all age classes. And the type III curve is for species that have high mortality in the youngest age groups and then relatively low mortality after that. Human survivorship generally fits a type I curve, though the shape of the curve may vary between communities and for people born in different decades. In this activity we'll plot survivorship curves for two human populations, born in different decades, in the same area. We'll compare the curves, hypothesize about why they might be similar or different, and make some predictions about how the shape of survivorship curves may be linked to environmental variables.
Instructions
1. In order to do this activity, you will need to find a large cemetery that has a significant number (at least two hundred) of headstones for people born before 1900.
2. You will choose two decades to compare (people born in the 1860's and people born in the 1890's, for example).
3. For each of the two decades you will collect 100 samples (200 in total). For each sample (headstone) you will record on the data sheet the birth year, death year, and age at death (see Table 1 below).
Table 1: Data Sheet for one decade (e.g. 1840-49)
| Individual | Birth Year | Death Year | Age at Death |
| 1 | |||
| 2 | |||
| 3 | |||
| 4 | |||
| 5 | |||
| 6 | |||
| .... | |||
| .... | |||
| 100 |
4. The next step is to summarize the data. You will create two tables (see example below), one for each decade. For each decade table you will start by grouping the data into age classes of ten year intervals.
In the second column of both decade tables you will enter the the total number of individuals who died in each age class.
So in the first row of the second column (C2,R1) you'll enter all the people who died at age 9 or younger. In the second cell (C2,R2), all the people who died between the ages of 10 and 19, and so on down the column.
| C 1 | C 2 | C 3 | C 4 | C 5 | |
| Age Class Years | # of deaths in class (dx) | # surviving from birth (lx) | survivorship per 1000 (S1000) | Log10 | |
| R1 | 0-9 | ||||
| R2 | 10-19 | ||||
| R3 | 20-29 | ||||
| R4 | 30-39 | ||||
| R5 | 40-49 | ||||
| R6 | 50-59 | ||||
| R7 | 60-69 | ||||
| R8 | 70-79 | ||||
| R9 | 80-89 | ||||
| R10 | 90-99 | ||||
| R11 | 100-109 | ||||
| R12 | 110+ |
5. The third column contains the number surviving from birth. You start filling this column by putting a zero in the bottom cell (C3,R12). In the next cell up (C3,R11) you add to the zero the number from C2,R11. To calculate the value in C3,R10 you add C2,R10 plus C3,R11. To calculate the value in C3,R9 you add C2,R9 plus C3,R10. And so on. The value in C3,R1 is C2,R1 plus C3,R2 - that number should be 100. Here's an example of a completed table.
6. In the fourth column standardize the data to per thousand. Because you have a sample of 100 headstones, you simply multiply the numbers in column 3 by 10.
7. in the fifth column calculate log10 for each number in column 4. The top number (C5,R1) should be 3.00 and the bottom number (C5,R12) is log10 zero which is undefined and you can just enter a zero there.
8. Now create a graph with the years (0, 10, 20, 30, etc.) on the x-axis years and the corresponding log10 values on the y-axis. Plot a curve for each of your two decades. Excel would be the easiest way to do that.
9. Using your graph think about the following questions:
- Do your two curves differ at all? Can you think of any explanations for the differences?
- In the future, how might a human survivorship curve differ in shape if the cases of AIDS increase and no cure is found?
- How might a survivorship curve differ in shape if medical advances reduce the infant mortality rates?
- If pollution related diseases increase, how might that alter survivorship curves?
- We are assuming that the survivorship curves you have plotted are representative of the the survivorship patterns of the local community. What are some of the assumptions you are making in drawing that conclusion?
10. Write up the activity as a brief report that includes the summary data tables for the two decades, the graph with the curves for the two decades, and a discussion of the above questions.
11. If you are interested in comparing your results to those for other communities and decades, there are several websites with student-collected data sets that you can use to plot survivorship curves. Sites at Hamilton College, and Bloomsburg University, both link to several datasets (though unfortunately some of the links are broken).
