Experiments

Question 1

Why use experiments

Answer 1

They are good for testing causality

• Temporal order
• Associations
• No alternative explanations

Can observe causation

Question 2

How to randomly assign

Answer 2

Divide the collection of cases into 2 or more groups by random process

Question 3

Language of experiments

Answer 3

Treatments
- Independent varibale

Dependent variable
- Whats affected

Pretest
- Measurement of the dependent variable prior to the introduction of the treatment

Post-test
- Measurement of the dependent variable after the treatment has been introduced into the experimental situation

Experimental group
- Group that receives the treatment

Control group
- Group that does not receive treatment

Random assignment

Question 4

Type of design

Answer 4

Classical experimental design
- All designs are variations on classical experimental design

Pre-experimental design
- Experiments that lack random assignment or use shortcuts. These are much weaker than the classical experimental design

One shot case study
- Has only one group a treatment and a post-test

One group pretest-pose-test design
- One group, a pretest, a treatment and a post-test

Static group comparison
- Two groups, a post-test and a treatment
- lacks random assignment and pretest

Question 5

Types of design - Quasi-experimental

Answer 5

Two-group post-test only
- Very similar to static group comparison except the groups are randomly assigned

Interrupted time series
- Dependent variable is measured multiple times across a period of time

Equivalent time series
- Over time
- Repeated pretest, treatment, posttest

Latin square designs
- Researchers are interested in whether the sequences or order of treatment has an effect

Question 6

What is design notation?

Answer 6

The name of a symbol system used to discuss the parts of an experimental and to make diagrams of them

Question 7

What are threats to internal validity?

Answer 7

Selection bias
- Threat that research participants will not be from equivalent groups

History effects
- Threat that an event unrelated to the treatment will occur during the experiment and influence the dependent variable

Maturation
- Threat that some biological, psychological or emotional process within the subjects and separate from the treatment will change over

Testing effect
- The test itself affects the dependent variable

Instrumentation
- Threat occurs when the instrument for measurement changes during the experiment

Mortality
- Attrition when some participants participate throughout the experiment

Statistical regression
- Extreme values pull the average of other results up or down
- Random errors move group results toward the average

Diffusion or treatment of contamination
- Participants will communicate with one another

Experimental expectancy
- Through their actions experimenter conveys what they expect of participants

Double-blind experiment
- Designed to control for researcher expectancy

Placebo
- A false treatment or one that has no effect in an experiment

Question 8

What is external validity?

Answer 8

The ability to generalize results of an experiment beyond the study itself

Question 9

What is reactivity?

Answer 9

Participants may react differently in the study based on the fact that they are under study

Question 10

External validity

Answer 10

Natural experiments
- Quasi-experimental design when researchers can use a “natural” change in society and measure an outcome before and after the change