Implementing different versions of Student’s t-test
The diabetes dataset contains data from subjects that are with diagnosed diabetes (CLASS: Y
), predicted diabetes (CLASS: P
), and controls (CLASS: N
). Using the available data, we will perform an analysis comparing average values of HbA1c between diagnosed diabetes and controls. The biological parameter we will be analyzing in this exercise is known as HbA1c, or glycosylated hemoglobin, which is higher in diabetes subjects due to the glycation of hemoglobin due to increased blood glucose. Let’s explore how grouping can be made based on the CLASS
variable, which contains information on the presence or absence of diagnosed Type 2 Diabetes Mellitus (T2DM).
Figure 8.1 – Comparing HbA1C in diabetes subjects and control group
For this exercise, the main setting is that we are comparing two groups, diabetes subjects (CLASS: Y
) and control subjects (CLASS: N
). There is a third CLASS, P, or...