Psychological Research Design Notes

Acknowledgment of Country

• UQ acknowledges the Traditional Owners of the lands.
• Respects are paid to Ancestors and descendants.
• Recognition of valuable contributions to society.

Stages in Psychological Research

• Theory: A proposed explanation.
• Hypothesis: A testable prediction derived from the theory.
• Study Design: Planning how to test the hypothesis.
• Data Collection: Gathering necessary data.
• Analysis: Using statistical tests to interpret data.
• Report Writing: Summarizing findings.
• Possible outcomes:
  • Not supported: hypotheses rejected.
  • Confirmed: repeated experiments support theories.

Research Design

• Key elements of research design include:
  • Control: Manage random variability, individual differences, and confounding variables.
  • Data Analysis: Statistical testing to analyze results.

Basic Research Designs

• Types of Designs:
  • Experimental
  • Quasi-experimental
  • Correlational

Independent Variables (IV) and Dependent Variables (DV)

• Independent Variable (IV):
  • Manipulated to observe its effect.
  • Identifies participant categories in quasi-experiments.
• Dependent Variable (DV):
  • Measures changes resulting from IV manipulation.
  • Depends on the IV.

Extraneous Variables

• Unwanted/Extraneous Variables include:
  • Random Variables
  • Situational Variables
  • Individual Differences
  • Measurement Error
  • Confounding Variables: can obscure cause-and-effect links between IV and DV.

Hypotheses

• Null Hypothesis (H0):
  • Claims no relationship exists between IV and DV.
  • Results are due to chance.
• Alternate Hypothesis (H1):
  • Suggests a relationship exists between IV and DV.
  • Tentatively accepted if results deviate significantly from chance outcomes.

Quasi Experiments

• Example:
  • Study comparing doctor visits between 50 smokers and 50 non-smokers.
• Groups equated in age and health as much as possible.

Correlational Research

• Conducted by measuring actions (e.g., smoking frequency) and outcomes (e.g., doctor visits).
• Scatterplot: Visual representation of relationships (Pearson’s r).
• Advantages:
  • Feasible when random allocation isn't possible.
  • Maintains ecological validity.
• Disadvantages:
  • Causality cannot be inferred directly.

Designing Experiments

• Importance of ensuring no systematic differences between treatment and control groups.
• Random assignment reduces bias and variability.
• True Experiments utilize randomized trials to affirm causal relationships.

Random Assignment Procedure

• Randomly assigning participants to conditions to ensure equal distribution among groups.
• Reduces variability and bias in results.

Aims of Research Design

1. Eliminate confounding variables to interpret IV-DV relationships clearly.
2. Minimize random variability to enhance detection of relationships.

Basic Experimental Research Designs

1. Independent-Groups Design: Participants are assigned to different groups.
2. Repeated-Measures Design: Participants engage in all conditions.

Repeated Measures Example

• All participants exposed to both conditions.
• Potential Confounds:
  • Order effects: fatigue or practice impact results.
  • Solutions: counterbalancing—rotate participation order to mitigate these effects.

Confounding Variables

• Occur when external variables affect outcome interpretations.
• Solutions include controlling for participant characteristics or balancing testing times.

Conclusion: Types of Studies and Control

• Different study types provide various insights:
  • Experimental studies provide clear causal connections.
  • Observational studies (quasi-experimental/correlational) show relationships but lack causal inferences.
• Essential to manage and control for variability to draw accurate conclusions in research.