Random error

Table of Contents:

Random error is residual error, and is, particularly, an example of how randomness can be applied to help you identify and quantify real data effects. It is the amount of uncertainty in the outcome of your study and it is the measure of spread in your data.

Traditionally its magnitude is tested against the magnitude of the main study result, expressed in a mean value, mean difference, proportion, odds ratio, etc. The ratio of the two magnitudes are the socalled standardized study result that should be larger than two standard errors in order for a trial to be statistically significant.

Random access

When you have random access to a data file, this is to say that you will have a direct access to each data value, and you do not need to take multiple prior steps for the purpose.

Random selection

Random selection is the procedure whereby participants in a study are randomly selected from the entire target population, which is the population we want to make predictions about.

< Prev   CONTENTS   Source   Next >