Finding patterns in data: simulation tutorial

Data is a way of trying to understand the world, we can observe it and try to see if there are any meaningful patterns. 

Public health officials are especially interested in disease patterns in a populations. Patterns help us predict who might get sick, possibly help prevent some diseases or target treatment for certain diseases.

For example, if public health experts observe a pattern in which people who live in areas which get flooded often tend to get malaria, they might focus providing mosquito nets to those people since they are closer to mosquitos.

Why are some people at greater risk of getting sick than others? We can answer questions like this by exploring data. Data lets us see patterns. These patterns might not be obvious without looking at the data. Today, we will be using a computer model to investigate a group of people. 

In this model, each circle represents a person. So if there are 20 circles, there are 20 people in our dataset.


 

Anemia is a serious disease. Many people in our country suffer from anemia. It can be caused by deficiency of Iron in blood. 
In our model, each person is either anemic (a yellow circle) or not anemic (a magenta circle).



The people in our data set have other traits as well. For example, in addition to being anemic or not anemic, each person is ALSO either male or female. In order to “see” each person’s gender, we need to sort them. In this screenshot, males are grouped on the left and females are grouped on the right. ​