These lecture notes are based on Stata and wherever possible, corresponding codes in R for analysis.
Biostatistics differs from statistics in a number of ways where here our emphasis is on biological or health care related examples; in biostats we talk about survival analysis, and study designs and analytical techniques that are different from how these things are dealt in the regular statistics. Biostatistics is closely linked with epidemiology and epidemiological study designs. Here we shall deal with some essential principles.
Biostatistics connects Epidemiology and thinking in epidemiology with the real world data and provides you with the tools that you use in order to explore issues around biology and health care to explore distribution of exposure and health effects and study the association between exposure and health effects. Let's start with a quick recap of scientific thinking and setting up of studies.
Scientific thinking is based on the following rough principles and methods:
- Note observations and get the facts. When you observe the facts, they should not be colluded by your own feelings or emotions. Therefore a plan of measuring what you observe is important
- For explaining the world, you should have a theory that should explain the world of observations
- Based on the theory set up a set of predictions and test the predictions with new sets of data
- You should set up more than one theory that can explain the pattern and continue to refute theories till the best one stands
From this listing, we note that a few things are essential for the scientific process of induction, deduction, and abduction to proceed:
- We need a method and tools to describe the world
- We need a way to measure entities whose values vary
- We need tools and ways to then link the measured entities with each other
- In case of biostatistics, we deal with biological and health care related variables
Biostatistics provide us with the tools of thinking and using the thinking tools. We use concepts, techniques and tools. In this lecture note, let's cover a few points.
Four different types of variables
Tables and figures for nominal and ordinal variables
How do we estimate mean and standard deviation for continuous variables?
Links
The function for binomial coefficient in R is choose()