Welcome to Experimental Methods and Design!

Ever wondered if listening to music actually helps you study, or if it just distracts you? To find out the truth, psychologists don't just guess—they use experiments. In this chapter, we will explore how researchers design studies to discover cause-and-effect relationships. This is a vital part of your Research Methodology strand and is the specific method you will use for your class practical in the Learning and Cognition context.

1. What is an Experiment?

The goal of an experiment is to establish causality. This means we want to prove that changing one thing (the cause) leads to a change in another thing (the effect). Unlike just observing people, in an experiment, the researcher is "in the driver's seat," actively changing variables to see what happens.

The Three Golden Ingredients:

  • Independent Variable (IV): This is the factor the researcher manipulates or changes. Think of it as the "cause."
  • Dependent Variable (DV): This is the factor the researcher measures. It "depends" on the IV. Think of it as the "effect."
  • Control: Keeping everything else the same so that only the IV can influence the DV.

Analogy: Imagine you are testing a new fertilizer on plants. The IV is the type of fertilizer. The DV is how tall the plant grows. To make it a fair test, you must control other things, like giving every plant the same amount of water and sunlight.

2. Types of Experiments

The IB syllabus focuses on two main types of experiments that you need to know for your "Learning and Cognition" practical and your exams.

A. True Experiments

In a true experiment, researchers randomly assign participants to different groups. Because of random assignment, each person has an equal chance of being in the "experimental group" or the "control group." This helps ensure the groups are similar before the experiment even starts.

Key Feature: High control over variables, making it the best way to show causality.

B. Quasi-Experiments

Sometimes, we can't randomly assign people. For example, if you want to compare "people who have smoked for 10 years" vs. "non-smokers," you can't randomly pick a group and force them to smoke! In a quasi-experiment, the researcher uses pre-existing groups (like age, gender, or people with a specific health condition).

Key Feature: The researcher does not manipulate the IV because the participants already belong to a specific category.

Quick Review: If the researcher flips a coin to put you in a group, it’s a True Experiment. If you are in a group because of who you already are (e.g., your age), it’s a Quasi-Experiment.

3. Experimental Designs

Once you’ve decided on the type of experiment, you need a plan for how to use your participants. This is called the experimental design.

Independent Measures Design

In this design, you have different groups of people for each condition. One group tries "Condition A" and a completely different group tries "Condition B."

  • Pro: No "order effects" (participants don't get tired or bored because they only do one task).
  • Con: "Participant variables" (maybe one group is naturally smarter or faster than the other).

Repeated Measures Design

The same group of people does all the conditions. First, they do "Condition A," and later, they do "Condition B."

  • Pro: You don't have to worry about "participant variables" because you are comparing a person against themselves!
  • Con: Order effects. Participants might do better in the second task because they've had practice, or worse because they are tired.

Matched Pairs Design

Researchers pair up participants who are very similar (e.g., two people with the same IQ or the same age). One person from the pair goes to Group A, and the other goes to Group B.

  • Pro: Reduces participant variables while avoiding order effects.
  • Con: It is very time-consuming and difficult to find "perfect matches."

4. Connecting to the IB Core Concepts

In your Paper 2, you may need to evaluate research using specific concepts. Here is how they apply to experiments:

Causality: This is the main goal. We want to know if \(IV \rightarrow DV\). If an experiment is well-designed with high control, we can be more confident about causality.

Measurement: This is about how we turn behavior into data. For example, if we are testing "memory," do we measure it by how many words someone remembers, or how long it takes them to recognize a face? Measurement must be consistent and accurate.

Bias: Experiments try to reduce bias. Researcher bias happens if the scientist accidentally influences the results. Participant bias happens if the participant tries to guess the aim of the study and changes their behavior.

5. Common Mistakes to Avoid

  • Confusing the IV and DV: Remember, the Independent variable is the one I change.
  • Assuming all experiments are in labs: While many are, the defining feature of an experiment is the manipulation of the IV, not the room it happens in!
  • Ignoring Participant Variables: When using an independent measures design, always remember that differences in the results might be because the people in the groups were different to begin with.
Did you know?

To prevent bias, researchers often use a "Double-Blind" setup. This is where neither the participant nor the person collecting the data knows which group is getting the real treatment and which is getting a "placebo." It’s like a blind taste test where even the waiter doesn’t know which soda is which!

Summary Key Takeaways

1. Experiments are the only method that can truly establish causality.

2. A True Experiment uses random assignment, while a Quasi-Experiment uses pre-existing groups.

3. Independent Measures use different people in each group; Repeated Measures use the same people for everything.

4. The IV is the cause, and the DV is the effect we measure: \(IV \text{ changes } \rightarrow DV \text{ is measured}\).

Note: For your Internal Assessment (IA) and Paper 2 Section A, remember that you will be designing a research proposal. Understanding these experimental designs is the foundation for a high-scoring proposal! For more on how to pick your participants, see the chapter on "Sampling and populations of interest."