Basic concepts of statistical inference using R
>eR-BioStat
Topics
The course in slides format
We provide two (similar) courses that were developed by the >eR-BioStat team and by Julie Vu and Dave Harrington(https://www.openintro.org/book/biostat/), respectively.
Both courses cover the same topics and we provide the course files that were used to produce the slides.
The slides in this page are organised in three chapters that cover basic topics in statistical inference using R. The course is focused on the practical aspects of inference and not only on the theory behind and covers the following topics:
-
Inference for continuous data: one sample.
-
Inference for continuous data: two sample.
-
Inference for categorical data: one & two samples.
All examples are illustrated using the R software. Useful R functions include:
-
t.test()
-
prop.test()
-
chisq.test()
External datasets for illustration are included in the data repositories.
basic concepts of inference (one population)
This chapter introduces the important concepts in drawing estimates from samples from one population. We discuss methods of inference and estimation for a population mean and cover three topics:
-
Point estimates for a population mean.
-
Interval estimates methods.
-
Testing hypotheses about the population mean.
External datasets are available in the data repository.
Inference for continuous data (one population & two populations)
This chapter discusses the settings where
The methods discussed in the previous class are applied to multiple populations. This include:
-
The analysis of paired observations
-
The comparison of two or more independent groups.
Both normal and t distribution are used for both interval estimation and inference.
External datasets are available in the data repository.
Inference for binary and categorical data
In this chapter, methods for estimation and inference are extended to categorical data, such as binomial proportions or data in two-way tables. We cover the following topics:
-
Tests and estimation methods for proportion in one population.
-
Tests and estimation methods for proportions in two populations.
-
Chi-square tests.
External datasets are available in the data repository.
Recommended reading for the classes about inference for one and two population and for inference of categorical data are Chapter 4 , Chapter 5 and Chapter 8 in the book "Introductory statistics for the life and biomedical sciences", respectively, which is available online here:
This part of the course is based on unit 4 in Vu & Harrington course and it
covers the following topics:
-
Inference for one population.
-
Point estimate and interval estimates.
This part of the course is based on unit 5 in Vu & Harrington course and it covers the following topics:
-
Inference for paired data.
-
Inference for independent samples.
This part of the course is based on unit 8 in Vu & Harrington course and it covers the following topics:
-
Categorical outcome and chi-square tests.
-
Binary outcome and test for proportions.