Experimentation and Ethics Flashcards
Experiment.
A technique for establishing the causal relationship between variables.
An experiment is comprised of:
Manipulation and Randomization.
Manipulating the independent variable creates:
The control and experimental groups (exposure to violent TV).
After creating a pattern of variation through manipulation, we measure the:
Dependent variable (aggression).
After we measure the dependent variable, we:
Check whether variations in the independent and dependent variable are synchronized (correlation coefficient). Is there a causal relation?
Two types of randomization:
Self-selection and random assignment.
Self-Selection Problem:
Problem that arises when a participant’s inclusion in the control or experimental group is determined by the participant.
Random Assignment:
A procedure that uses random events to ensure that a participant’s assignment to the experimental or control group is not determined by any third variable (flip a coin or pull from a hat).
Flaw in Random Assignment?
Third variable problem can still exist. We cannot conclude that there is a causal relationship between the independent and dependent variable.
Inferential Statistics.
Calculations that are used to tell whether or not random assignment worked or failed. Worked only if p<0.05.
Srtatistical Significance Can Be Claimed If:
There is a 95% or greater chance that random assignment worked.
Internal Validity
The characteristic of an experiment that allows one to draw accurate inferences about the causal relationship between an independent and dependent variable.
Internal Validity Can Be Claimed If:
- Independent variable was effectively manipulated.
- There was random assignment.
- The dependent variable was measured in a valid, unbiased manner, with reliability and powerful measure.
- A correlation was observed between the pattern of variation in the independent variable nad the pattern of variation dependso n the dependent variable.
External Validity.
A characteristic of an experiment in which the independent and dependent variables are operationally defined in a normal, typical, or realistic way. Does not happen often, as experiments are rarely replicated in the real world.
Theory.
A hypothetical account of how and why a phenomenon occurs, usually in the form of a statement about the causal relationship between two or more properties.
