Introduction to Biostatistics 01 - Summarizing Data
Introduction to Biostatistics 01 - Summarizing Data
This lecture provides an essential introduction to biostatistics, focusing on how to correctly describe and summarize biological data. It addresses the common misuse of statistics in medical literature and introduces the fundamental concepts of sampling, measurement, and the importance of accounting for natural biological variability. Using illustrative examples of "Martian" and "Venetian" populations, the session explains how to choose between parametric and non-parametric descriptors based on data distribution.
Learning objectives
By the end of this lecture, students will be able to:
- Explain the importance of biostatistics in medical research and the consequences of misusing statistical methods in clinical literature.
- Calculate and interpret core descriptive parameters, including the population mean, variance, and standard deviation.
- Differentiate between a normal (bell-shaped) distribution and skewed distributions to determine which statistical tests are appropriate.
- Apply the Central Limit Theorem to understand how sample means behave regardless of the original population's distribution.
- Distinguish between Standard Deviation (SD) and Standard Error of the Mean (SEM), identifying when to use each to report variability versus precision.
Topics covered in this lesson
- Biostatistics plays a crucial role in clinical research by helping identify therapies with real value and effectively managing medical resources.
- Mean and standard deviation are used as complete descriptors for populations that follow a normal, bell-shaped distribution.
- Median and percentiles (25th and 75th) serve as non-parametric descriptors to accurately characterize skewed or non-normal data distributions.
- The relationship between a whole population and a sample explains the mathematical rationale for using n-1 when calculating variance.
- The Central Limit Theorem demonstrates how the distribution of sample means creates a necessary bridge to advanced statistical testing.
- Distinguishing between standard deviation and the standard error of the mean clarifies the difference between reporting population variability and the uncertainty of a mean estimate.
External links
