Experimental Research Design — Previous Year Questions
📊 Topic Overview
| Metric | Value |
|---|---|
| Papers Appeared | 13 of 28 (46.4%) |
| Priority | 🔴 HIGH |
| Section Distribution | Sec A: 8 |
🎯 Must-Know Concepts
- Between-group (independent groups) designs: randomised, matched, natural groups
- Within-subjects (repeated measures) designs
- Single-factor design — meaning and types
- Comparison between between-group and within-group designs
- Control group and randomisation
Key Question: Types of Experimental Research Design / Between-Group vs Within-Group
Answer (Section A — ~450 words):
Experimental research design is a research plan in which the researcher manipulates one or more independent variables, controls extraneous variables through random assignment, and measures the effect on a dependent variable to establish causal relationships.
Single-Factor Designs: Single-factor designs involve one independent variable (factor) with two or more levels. They are divided into between-subjects (independent groups) and within-subjects (repeated measures) designs.
Between-Group (Independent Groups) Designs:
1. Randomised Groups Design: Participants are randomly assigned to experimental and control conditions. Random assignment is the most powerful control for pre-existing individual differences. The simplest and most internally valid between-group design.
2. Matched Groups Design: Participants are matched on a relevant variable (e.g., IQ, age) before being randomly assigned to conditions. Matching reduces error variance and increases statistical power. Used when sample sizes are small.
3. Natural Groups (Ex-Post Facto) Design: Groups are formed based on a pre-existing characteristic (gender, diagnosis, age) rather than random assignment. The IV is not manipulated — participants "bring" it with them. Therefore causal inference is limited.
Within-Subjects (Repeated Measures) Design: The same participants serve in all conditions of the experiment. Each participant acts as their own control, eliminating individual difference variance entirely.
Advantages: More statistically powerful (less error variance); requires fewer participants; controls for individual differences completely. Disadvantages: Carry-over effects — exposure to one condition affects responses in subsequent conditions. Controlled by counterbalancing (presenting conditions in different orders to different participants).
Comparison: Between-Group vs Within-Subjects
| Feature | Between-Group | Within-Subjects |
|---|---|---|
| Participants per condition | Different | Same |
| Individual differences | Controlled by randomisation | Eliminated entirely |
| Sample size needed | Larger | Smaller |
| Carry-over effects | Not present | Possible; requires counterbalancing |
| Statistical power | Lower | Higher |
| Best for | Conditions that produce lasting changes | Conditions with minimal carryover |
Randomisation: Random assignment ensures that any pre-existing participant characteristics are equally distributed across conditions. It is the single most important design feature for establishing internal validity. It should not be confused with random sampling (which concerns external validity and generalisability).
Control Group: The control group provides a baseline against which the experimental group is compared. Without a control group, it is impossible to determine whether change in the DV is due to the IV or to extraneous factors (history, maturation).
🧠 Memory Aids
Between-Group Types (RMN): Randomised, Matched, Natural groups
Within vs Between key difference: Same participants (within) vs different participants (between)
📚 Related Content
✅ Self-Assessment
- What is the key advantage of a within-subjects design over a between-subjects design?
- What is carry-over effect and how is it controlled?
- How does a matched groups design differ from a randomised groups design?
- Why does random assignment control for individual differences?
- What is the difference between random assignment and random sampling?
💡 Exam Tips
- Section A: Define experimental design → Between-group types (3) → Within-subjects → Comparison table → Control group and randomisation.
- The comparison table (between vs within) is the most effective structure for this answer.
- Always distinguish random assignment (controls individual differences — internal validity) from random sampling (controls representativeness — external validity).