BIOL 2000H Lecture Notes - Lecture 10: Random Number Generation, Statistical Hypothesis Testing
Document Summary
Be able to determine the causal question, hypothesis, testing method, etc. from a short scenario. Determine all necessary variables and how they would be manipulated for your experiment. Control: refers to a zero level of the treatment or treatment combination. Sometimes control refers to internal method checks. In other situations, the meaning of control is less clear. Some experiments lack zero levels or ambient conditions. Identifying the control units in these experiments is arbitrary and somewhat meaningless. You can design valid experiments that test hypotheses without controls if you are certain that you manipulated an independent variable and you can measure the appropriate response variable: 1. How many levels of each factor: 3. More replication means more confidence in the mean and variance for each treatment combination. More replicates mean more work and more cost: 4. Short experiment = little time for response to develop. Long experiment = responses could be modified by other confounding variables.
