Correlation vs Causation (Unit 2 Mod 6-1)

Correlation

When 1 trait or behavior accompanies another

Correlation is NOT Causation

Correlational Coefficient

• Statistical measurement of relationship between two variables

• Formula: r = + 0.37

  • + = Indicates strength of relationship (0.00 - 1.00)

  • 0.37 = Indicates direction of relationships (positive or negative)

• Absolute value indicates strength of of correlation

Scatter Plot

• Graph made of points using 2 variables

  • Slope of points = Direction

  • Amount of scatter = Strength of relationship

Illusory Correlation

Perception of a relationship where there isn’t one

Experimenter/Hawthorne Effect

When people change their behavior because they know they’re being tested

Independent Variable

• Factor manipulated by experimenter

• The effect of the IV is the focus of the study

• If = I.V. Then = D.V.

Dependent Variable

Factor that may change in response to the independent variable

If = I.V. Then = D.V.

Single Blind Procedure

Participants aren’t informed

Double Blind Procedure

Neither participant nor the person administering the test knows who’s receiving the treatment

Random Assignment

• Procedure to assign participants to control and experimental group randomly

  • Ex: Random number generator

  • Each participant has an equal chance of being in either group

Experimental Group (Experimental Condition)

Participants exposed to independent variable and compared to the control group

Control Group (Control Condition)

Participants not exposed to independent variable and compared to the experimental group

Confounding Variables

• Variables other than the independent variable that effect the dependent variable

  • Need to be controlled to show cause → effect

  • Usually eliminated with random assignment

Experimental Differences

• Any differences in the experiment’s condition between the control and experimental group

  • Want minimal environmental differences

  • Ex: Lighting, temperature, noise levels

Placebo

• Inactive substance with no effect given to control group

  • If there’s a placebo effect on the show group it shows in the changes in the dependent variable