Glossary - Research Basics and Terminology

• Significance level
  
• Probability of rejecting H0 when H0 is true
  
• Usually set at α=0.05 or α=0.01
  

•  Experimental hypothesis
• The population parameter differs from H0
• When H0 can be rejected, it does not mean that H1 can be accepted, because of the remaining probability of α- and β-error

• Systematic error or deviation in results
  
• Example: selection bias
  

• Process of hiding which comparison group a particular participant belongs to
• Used to minimize risk of bias
• Common methodology in clinical trials
• Types:

• Measure of uncertainty around main finding of statistical analysis
• Usually set at CI=95%, also CI=99% or CI=90%

• Information that can take any value within a certain range
  
• Opposite of discrete data
  
• Example: Weight of a person: The weight can be any value within the range of people´s weight (e.g. 60.3 kilogram).
  

• Outcome that results from varying independent variable
• Example:

• Information that can only take certain values
  
• Opposite of continuous data
  
• Example: Number of people in a room: It is impossible to have 40.3 people in a room.
  

= Negative test when individual actually has the disease

= Positive test when individual does not actually have the disease