Chapter 11 - Experiments & Observational Studies

Elementary Statistics & Inference

What are the types of OBSERVATIONAL studies?

Retrospective & Prospective

What are the elements of a DESIGNED RANDOMIZED EXPERIMENT?

1. Experimental Units are assigned at RANDOM to TREATMENT.

2. A Quantitative Response Variable is measured/observed for each experimental unit.

3. We can attribute differences in the RESPONSE to the differences among the TREATMENTS.

What are the FOUR PRINCIPLES OF EXPERIMENTAL DESIGN?

1. CONTROL sources of variation other than the factors being tested.

2. RANDOMIZE the assignment of participants to treatments.

3. REPLICATE by applying each treatment to more than one participant.

4. BLOCK the experiment by grouping participants who are similar in important ways that you CANNOT control.

Single-Blind Study

A study in which either all those who can affect the results or all those who evaluate the results are kept ignorant of which subjects receive which treatments.

Double-Blind Study

A study in which both all who can affect the results AND all who can evaluate the results are ignorant of which subjects recieve which treatments.

Observational Study

A study based on data in which no manipulation of factors has been employed.

Retrospective Study

An observational study in which subjects are selected and then their previous conditions or behaviors are determined.

Prospective Study

An observational study in which subjects are followed to observe future outcomes.

Random Assignment

To be valid, an experiment must assign experimental units to treatment groups using some form of randomization.

Factor

A variable whose levels are manipulated by the experimenter.

Response Variable

A variable whose values are compared across different treatments.

Experiment

An ________ manipulates factor level to create treatments, randomly assigns subjects to these treatment levels, and then compares the responses of the subject groups across treatment levels.

Control 

We do this to aspects of the experiment that we know may affect the response, but are not the factors being studied.

Randomize

This is randomly assigning subjects to treatments to even out effects that are out of the experimenters control.

Replicate

We do this to emphasize that our data is representative of a population and not just a single anedotal occurence.

Block

We do this to reduce the effects of identifiable attributes of the subjects that cannot be controlled.

Placebo

A treatment known to have no effect, administered so that all groups experience the same conditions.

Placebo Effect

The tendency of many humans to show a response even when administered a placebo.