In addition to the random sampling, this student decided to use the systematic sample strategy. Systematic was used so that there would be an objective and unbiased identification of which plots are used. The entire site was covered because there were representatives from all the parts of the cemeteries.
Site Chronology. Results of the two sampling strategies show that people were burying their dead at the Evergreen Cemetery for a period of 131 year. The earliest burial took place in 1859 and the latest was at the end of the 20th century, in the year 1990. Within the entire period the most number of graves were buried in the years 1859 to1880. The random sample shows the highest number of deaths were in 1859-1869 period while the systematic sample shows a similar result with most number of burials in 1870-1880 but this number was closely followed by the deaths in the previous 10-year period. From this data we can see that the most number of burials took place in the first 20 years of our samples. Since the cemetery was established in 1850 then it is logical to assume that the most number would be in the initial decades since it would still have a lot of space and can accommodate the most number possible. A hundred years later, although the cemetery still accommodate new graves, the number of persons being buried has significantly lowered. The reason may be little space and probably families of the disease would also prefer new cemeteries.
Demography. The average age of the people buried at Evergreen is 59.06 based on systematic sampling and 57.76 based on the random sampling. These numbers would only approximate the exact average age because almost half of the gravestones do not show enough information. However, the figures for the age in these samples are actually close to the overall average of 61.4 for the entire number of graves (994) in the Evergreen cemetery.
The most number of deaths occurred in the 77-87 age group based on the systematic sampling. Random sampling shows the age groups of 66-76 and 77-87 as having the most number of deaths. As regards sex of the people interred at Evergreen, there is slightly more women (45.83%) than men (41.66%). However, a substantial number of graves did not have distinguishable marks thus, 12% of the total population could not be classified whether male or female. Since the most number of deaths occurred to the 77-87 age group and the average age at death is 57-59 years old, then it can be concluded that in the past century and a half, the people are able to reach old age. The population at Santa Cruz at these times was fairly health as many of those who died were in their seventies and eighties. There was even an entry for 104 years old.
Preservation. Random sampling shows 4.1% of graves are damaged while twice of this figure is the result of the systematic sampling which shows that 8.3% are damaged. Considering that the cemetery is already more than 150 years old, the rate of disturbance at Evergreen is not very high. Probably, damages of the graves result from the yearly effects of natural causes such as wind and rain. However, as regards data interpretation, the accuracy of results is affected by the absence of important information such as the birth year and names. Without such data it was difficult to pinpoint the average age and the exact gender distribution of those buried in the cemetery.
Comparison of Data Samples
Gender distribution. There were more female than male in the both the random and systematic samples. This is contrary to the results for the entire population which showed that there were more males than females among the people buried in the Evergreen cemetery. These contrasting results can be attributed to the considerable number of graves which could not be identified if a male or female was buried there. In a small sample, the situation of having lacking data in several entries would have a huge effect on the results as shown in this case.
Age profiles. The graph covering the entire population showed that the age at death of most of the persons buried at the cemetery was between 70-79. There were 150 people in this age group and this was followed by the 80-89 age group. In comparison, the results of the systematic sampling and random sampling are not far beyond. This student’s random sampling showed the highest to be in the 66-87 age group while the systematic sampling generated the most persons at the 77-87 age group.
Frequency of internment. Random sampling showed that the period with the highest number of internment was from 1859-1869. Systematic sampling indicated that the years 1870-1880 had the highest number. Results of the entire population showed that it was actually the period of 1860-1879 that has the most number of internment. For this variable, results of the systematic sampling were closer to those of the overall figures.
More effective strategy. In this sampling exercise, the systematic sampling strategy proves to be more effective. The preceding sections support this claim because when compared to the results using the entire population, the figures generated by the systematic strategy are in line with the results for the entire population. Thus, systematic sampling is more effective in gauging general trends.
Variables that diverged the most. The gender variable diverged the most from the total population data. As mentioned in the section above on gender, the results of the both the random sampling and the systematic sampling are in contrast to that of the entire population. Entire population has more males buried while both sampling methods show that there were more females buried. The reason for this is related to the (a) kind of variable and (b) data availability. In terms of the kind of variable, gender is a nominal variable, meaning there are distinct qualifications. In this case, it would either be a male or a female. Unlike continuous variables, where there could be values in-between or results can be grouped in ranges or age groups, the male-female choices cannot be grouped. Thus, if not a male, then a female. However, such classification would be dependent on whether such data is available. The second reason for the discrepancy is the lack of available data. From the samples, the number of damaged graves was 4% to 8%. Thus, it was difficult to identify whether those buried in those damaged graves were male or female.
Sampling strategies. For the data on gender, internment year, and age of death the data are similar. In the damaged graves data, systematic has twice the number of damaged graves than that of from the random sampling method. The benefit of using these two strategies is to have a validation of the results. In this case all the results of both strategies are similar. The only contrast was for the gender variable and this was in the comparison with the results for the entire population.