Drawing Conclusions from Data: an Introduction

The practical part of the first session (after the break) will be dedicated to getting to know the software package R. Participants are encouraged to participate in both parts.

We start with concise graphical and numerical descriptions of data obtained from observational or experimental studies. The most common and frequently used probability distributions of discrete and continuous variables will be presented. Statistical inference draws conclusions about a population based on sampled data. Chance variations are taken into account such that a level of confidence is attached to these conclusions.

We present the reasoning behind significance tests for the comparison of observed data with a hypothesis, the validity of which we want to assess. We apply this procedure to data obtained either from one or from two populations.

The correct use of the t-test will be discussed. Nonparametric methods are considered as a possible alternative in case the requirements of the t-test are not met.

We cover the basic concepts of hypothesis testing for categorical data, including the chi-square test.

Quite often the relationship between two variables, where the outcome of one variable is seen as depending on the value of the other, is the focus of scientific interest.

We will give an introduction to linear regression analysis, where a regression line based on observations obtained in a sample describes this relation.

Hands-on exercises are worked out behind the PC using the R software.