 ## Energy efficiency

### Calculating Power in R

Power, or probability, is a statistical concept that describes how likely a study is to detect a difference. It is also a key part of design and analysis.

Power can be measured in different units. For example, electrical power is a measure of voltage and current. The standard metric unit of power is watts. A watt is equivalent to the amount of work that it takes for an object to travel a certain distance in a given time. In SI units, a joule is equal to the amount of work it takes to move a kg of mass m2 over a period of one second.

Choosing a sample size is a critical part of designing any study. If a study is to be reliable, it must be large enough to achieve the desired effect. Depending on the research question, it may be necessary to consider stimuli and covariates.

Using Monte Carlo simulations to calculate power is one way to do this. This method can be especially useful for complicated study designs. EGAP has written code that allows users to perform simulation methods in R.

Performing sensitivity analyses is another helpful technique. These will help researchers determine which inputs are most important. Refinements to the design can increase the power of the study. Some inputs are hard to assess ex ante, while others can only be estimated from publicly available data.

The best way to estimate the power of your study is to think about it as a function. To illustrate, consider this example: How many participants would it take to detect an effect with a r-value of.6? 