<aside> 💡 Experimental design covers topics in the design and analysis of statistical experiments.

</aside>

Intro

Experimental design is the study of scientific claims. To do that, a hypothesis is formed, and either rejected or not rejected based on the results of a proper designed experiment.

When you have eliminated the impossible, whatever remains, however improbable, MUST be the truth - Sherlock Holmes

In the most basic experiment, we setup two groups of objects we want to conduct an experiment on. Ideally we want both groups to be identical in every possible way. We then apply a treatment on a group and measure something of interest, a response. If the groups differ in the response, it must be due to the treatment variable.

Definitions

• Experimental Unit: An experimental unit is an object (Eg: person, thing, transaction) on which measurements are taken.
• Treatment: Something that researchers administer to the experimental units. Each group of experimental units get a different treatment level.
• Response: The measurement of something interesting conducted on the experimental units.
• Population: A population is the set of all possible experimental units in the world (Eg: population of humans).
• Parameter: A parameter is the summary measure that describes a characteristic of the population (Eg: Mean, Standard Deviation).
• Sample: Typically, there are too many experimental units in a population to consider everyone, hence a sample is a subset of the population.
• Statistic: A statistic is a summary measure computed from a sample. Statistics are used to estimate population parameters.
• Variable: A variable is a characteristic or attribute of an individual experimental unit that will vary (Eg: Age, Gender, Education level).

Branches of Statistics