Class #1: How many tombstones are in Trinity Cemetary?

Welcome to Stimulus Science!

Hopefully this storm will clear out by morning, and we'll able to sun ourselves at Trinity Cemetery. New York City cemeteries are fascinating objects of study. The original solid state databases, you could spend your life mining their data-rich substrates. We will invest a simple question, and apply a basic scientific technique this Friday. Your mission, should you choose to accept it, is to figure out how many tombstones are in Trinity Cemetery.

If you take "Biology 101" in college, you'll likely experience some field work, where field really means walking a field, and your work will be trying to figure out the number of some species living in the field. You may smirk at the simple minded task, but take a moment to think about what sort of scientific procedures you may have to implement in order to establish a valid number. People spend their lives, writing books about this seemingly simple task.

The key issue in this sort of investigation is finding a scientifically credible way to be lazy, that is, not count all of the objects of inquiry, and yet claim to know to a reasonable degree of certainty, what the total number of objects is. In other words we need to sample a limited number of objects of interest, and project from that number, what the total number of objects in the field is.

The biggest problem to sampling is making sure your sample isn't biased. We could overlay a grid on the field, and count the number of objects in every third square of the grid. But systematic sampling isn't necessarily representative sampling. Can you think of a case where picking off every third object would lead to a biased result? Here's an example.

The key, therefore, is to develop a random sampling technique. What I'd like you to think about between now and our field trip, is what sort of scientifically credible way could we sample a number of tombstones from Trinity Cemetery, and deduce the total number of tombstones in the cemetery. You don't have to read everything in the links provided, but think about the question, and get ready to develop a scientific procedure on site tomorrow!