OCT25-PART_2_Experimental designs

Experimental Designs Overview

Instructor

  • Dr. Stelian Medianu

Key Concepts

Internal Validity and Confounding Variables

  • Understanding how confounding variables impact the internal validity of experiments is essential.

  • Internal validity is necessary for establishing a cause-and-effect relationship.

Basic Experiment Steps

  1. Manipulate the independent variable (IV).

  2. Measure the dependent variable (DV).

  3. Control for confounding variables.

Types of Experimental Designs

Independent Groups Design

  • Different participants are assigned to each level of the independent variable.

Repeated Measures Design

  • Same participants experience all levels of the independent variable.

  • Also known as a within-subjects design.

Matched Pairs Design

  • Participants are paired based on a key variable, then assigned randomly to one condition.

Repeated Measures Design

Structure

  • Participants experience every level of the independent variable:

    • Example: Level 1 and Level 2 for each participant in the study.

Study Example

  • Varao Sousa et al. (2013) explored memory and mind-wandering using a repeated measures design.

    • Treatments: Reading Aloud, Reading Silently, Listening.

Advantages

  • Fewer participants needed than independent groups designs.

  • Increased sensitivity in detecting differences.

Disadvantages

  • Potential for order effects:

    • Practice Effects: Improvement in performance due to repetition.

    • Fatigue Effects: Deterioration in performance due to boredom or exhaustion.

    • Contrast Effects: Participants might compare conditions and react differently based on their experience.

Solutions to Order Effects

Counterbalancing

  • Complete Counterbalancing: All possible orders of presentation included.

  • Partial Counterbalancing - Latin Square Technique: Each condition occurs once in each ordinal position and follows/preceded by each other condition.

  • Spacing Time Intervals: Helps to mitigate fatigue and contrast effects but may increase dropout rates.

Choosing the Right Design

  • No design is inherently superior; context determines suitability.

  • Consider:

    • Reversibility of effects.

    • Generalizability of results.

Matched Pairs Design

Process

  1. Match participants on key variables (e.g., genetic makeup, environment).

  2. Randomly assign one participant from each pair to a different condition.

Example

  • Diet and cognitive functioning study comparing Mediterranean and Standard American diets.

Usage

  • Used when participant availability is limited or running large numbers is costly.

Summary of Methods

Assignment to Experimental Conditions

  1. Independent Groups Design: Random assignment to conditions.

  2. Repeated Measures Design: All participants experience all levels of the IV.

  3. Matched Pairs Design: Match and then randomly assign one from each pair to a level of IV.

Key Terminology

  • Contrast Effect: Variations in responses based on the order of conditions.

  • Counterbalancing: Method to control for order effects.

  • Fatigue Effect: Decline in performance due to fatigue.

  • Mortality: Dropout of participants during a study.

  • Order Effect: Changes in responses due to the order in which conditions are presented.

Review Questions

  • Define confounding variable and internal validity.

  • What are strategies to ensure groups differ only in IV?

  • Discuss the strengths and weaknesses of each design type.