Anest. intenziv. Med. 2017;28(3):183-187

Statistics in biomedical research IISpecial article

K. Langová^*, J. Zapletalová, L. Ličman

Ústav lékařské biofyziky, Lékařská fakulta UP v Olomouci

The article focuses on data processing and data analysis. It briefly describes basic descriptive statistics - measures of central tendency and variability for continuous data. Particular attention is paid to the correct choice of descriptive statistics of central tendency (arithmetic mean, median, modus) and variability (standard deviation and quantiles) regarding data distribution. It explains basic types of data distribution - symmetric, right skewed and left skewed. Attention is paid to the normal Gaussian distribution and its statistical properties. There are various types of distribution for continuous biomedical variables.
It describes the basic steps of hypothesis testing as a method of inductive statistics. It explains the difference between parametric and non-parametric statistical tests. Parametric methods are based on normal distribution and use mean and standard deviation. Non-parametric methods are based on rank. There is an overview of various kinds of parametric and non-parametric tests for hypothesis testing of continuous variables. The importance of choosing the correct parametric or a non-parametric method is emphasized.

Keywords: descriptive statistics; mean; standard deviation; quantile; normal Gaussian distribution; parametric and non-parametric tests

Received: October 11, 2016; Accepted: January 5, 2017; Published: June 1, 2017  Show citation