Chapter 4.2: Observational Study vs Experiment
Ethical Experiment
if not an ethical experiment then there cant be an experiment- observational study
Statistically Significant
if the observed difference in responses between the groups in an experiment is too large to be explained by chance variation in the random assignment
Blocking
if you have a known characteristic within your experimental units that you would expect to affect the response to the treatments- you must block -you must say why you are blocking -you can only compare results within each block
Inference (about population)
individuals taking part in the study are randomly selected form the larger population -about cause and effect- treatments are randomly assigned to experimental units
Matched Pairs Design
matched subjects (with known similarities) are used to compare a treatment or both treatments are done to the same subject
Explanatory Variable
("x" independent) helps explain or predict changes
Response Variable
("y" dependent) measures an outcome of a study
Four Principals of Experimental Design
-comparison: at the end of experiment -Random Assignment: experimental units to treatments -control: keep other variables constant, or have a control group -replication: enough experimental units -concluding inference: can only make inference about he population you selected from
Treatment
a specific condition applied to the individuals in an experiment
Experiment
deliberately imposes some treatment on individuals to measure their response
Single Blind Experiment
neither the subject or whoever is interacting with subjects knows who is receiving what treatment
Double Blind Experiment
neither the subject or whoever is interacting with the subject know who is receiving what treatment
Observational Study
observes individuals and measures variables of interest but does not attempt to infer the response
Confounding Variable
occurs if a variable other than the explanatory variable or response variable is associated with the explanatory and response variable
Factors
often used for explanatory variables, different categories of your treatment
Lurking Variable
something else that may be affecting your experiment
Experimental Units
the individuals of which the treatment is applied