First page Back Continue Last page Overview Image
Randomization: Assigning experimental units to each of the treatments randomly is key in an experimental design.
Randomization: Assigning experimental units to each of the treatments randomly is key in an experimental design.
- Chance is used to ensure that the responses that we observe are due to the effect of the treatments and not the underlying characteristics of the experimental units or other variables which we haven’t controlled.
- Reduces bias, the systematic favoring of certain outcomes, (both from the experimental units and the experimenter).
- In some experiments subjects are blindly assigned to treatments – they don’t know what treatment that they are receiving to help eliminate bias (e.g. placebo effect).
- In double-blind experiments neither the subject nor the experimenter knows to which treatment the subject is assigned.
Notes:
Relate each of the above points to the Stats 202 problem