Psychology Chapter 9 - Factorial Designs

Chapter 9 - Factorial Designs

• Factorial designs are experiments with more than one independent variable (IV).
• Each level of one IV is combined with all levels of the other IVs.
• This creates all possible combinations of conditions.
• Example: A 2 (cell phone use: yes/no) x 2 (time of day: day/night) design with driving ability as the dependent variable (DV) would have four conditions.
• Factorial designs can have any number of IVs and levels. A 2x2x2 design would test three IVs, each with two levels.
• Fewer IVs and levels are usually preferred due to the rapid increase in the number of conditions and required participants.

Upcoming Dates

• November 19th: Chapter 9 (factorial designs) lecture and tutorial
• November 21st: Chapter 10 (single-subject research) recorded lecture
• November 26th: Assignment 3 due (11:59 pm)
• November 26th: No in-person lectures or tutorials
• December 9th: Final exam (12-2 pm, Regent Theatre)

Assignment #3: Method (20%)

• Due November 26th (11:59 pm)
• Includes a "Current Study" section (revised from Assignment 2)
• Detailed "Method" section describing participants, materials, and procedure.
• "Study Limitations and Future Research Directions" section
• Resources (template and example paper) are available on Canvas.

Final Exam

• Monday, December 9th, 12-2pm
• Regent Theatre (50 King St E, DTR100)
• Multiple-choice and short-answer questions
• Covers chapters 1-10
• Closed-book
• 35% of final grade

Factorial Designs - Just Extra Variables

• Complex experiments have two or more manipulated or measured (individual difference) variables.
• Each variable can have a main effect.
• Variables can influence each other (interaction).

Assigning Participants To Conditions

• Between-Subjects: All factors are between subjects (e.g., each person experiences only one condition of a factor).
• Within-Subjects: Each subject receives all conditions.
• Mixed: One factor is between-subjects, the other is within-subjects; needs counterbalancing.

Non-Manipulated IVs

• Measured but not manipulated (e.g., gender, age).
• Often individual difference variables.
• Allow causal conclusions only for manipulated variables.

Non-Experimental Factorial Design

• Includes only non-manipulated IVs.
• Caution in inferring causation due to directionality and potential third variables.

Main Effects

• The effect of one independent variable (IV) on the dependent variable (DV), averaging across the levels of other IVs.

Interactions

• The effect of one IV depends on the level of a second IV.
• Often considered the primary research question.
• Can be spreading or cross-over interactions.

Simple Effects

• Show precisely what's happening in an interaction.
• Useful when there's a significant interaction.

Types of Interactions

• Spreading: Effect of one IV varies depending on the level of another IV.
• Cross-over: Effect of one IV is in opposite directions at different levels of another IV.