Introduction to Research Design

Welcome to the "blueprint" of Psychology! Before a psychologist can conduct an experiment, they need a solid plan. Think of this chapter as the architecture of research. If the design is shaky, the whole study might fall down. We are going to look at how researchers decide what they want to find out, who they are going to study, and how they organize their participants to make the results as fair as possible.

1. Aims and Hypotheses

Every study starts with a "Why?" and a "What do I think will happen?"

Aims

An aim is a general statement of what the researcher intends to investigate. It’s the purpose of the study.
Example: "To investigate whether drinking coffee affects memory."

Hypotheses

A hypothesis is a clear, precise, testable statement that predicts the outcome of the study. It’s written at the start of the investigation. There are two main types you need to know:

1. Directional Hypothesis: The researcher predicts the specific direction of the results (e.g., higher, lower, more, less). We use this when previous research suggests which way the results will go.
Example: "Students who drink coffee will recall significantly more words than students who do not drink coffee."

2. Non-Directional Hypothesis: The researcher predicts that there will be a difference, but they don't say which way it will go. We use this when there is no previous research or when previous findings are contradictory.
Example: "There will be a significant difference in the number of words recalled by students who drink coffee and students who do not drink coffee."

Quick Tip: If the question asks you to write a hypothesis, always make sure it is operationalised (measurable). Instead of saying "coffee makes you better," say "200ml of caffeine results in more words recalled on a list of 20."

2. Variables: The Building Blocks

To test a hypothesis, we need to change one thing and measure another.

Independent Variable (IV): The variable that the researcher changes or manipulates. It is the "cause."
Dependent Variable (DV): The variable that the researcher measures. It is the "effect."

Operationalisation: This is a fancy word for making variables measurable. You can't just measure "intelligence"; you measure "an IQ score." You can't just measure "aggression"; you measure "the number of times a child hits a doll."

Extraneous Variables: These are "nuisance" variables that might affect the DV if they aren't controlled (e.g., the temperature of the room). If they actually do ruin the results by providing an alternative explanation, they become Confounding Variables.

3. Sampling: Choosing Your Participants

Researchers usually want to know about a large group of people (the target population). Since they can't study everyone, they select a smaller sample. The goal is for the sample to be representative so the results can be generalised.

1. Random Sampling: Every member of the population has an equal chance of being picked (e.g., names out of a hat).
Strength: No researcher bias. Limitation: You might still end up with an unrepresentative sample by chance.

2. Systematic Sampling: Selecting every \( n^{th} \) person from a list (e.g., every 5th person on a school register).
Strength: Objective and usually avoids bias.

3. Stratified Sampling: The sample reflects the proportions of people in certain sub-groups (strata) within the target population.
Strength: Highly representative, meaning results can be generalised. Limitation: Very time-consuming to calculate and execute.

4. Opportunity Sampling: Simply asking whoever is available at the time (e.g., people walking past in the street).
Strength: Quick and easy. Limitation: Highly biased as the sample is limited to one place and time.

5. Volunteer Sampling: Participants select themselves (e.g., replying to an advert).
Strength: Participants are willing and engaged. Limitation: Volunteer bias — people who volunteer might be more helpful or curious than the average person.

4. Experimental Design

This is how we arrange our participants into groups for the IV.

Independent Groups: Different participants are used in each condition (e.g., Group A drinks coffee, Group B drinks water).
Strength: No order effects (boredom or practice).
Limitation: Participant variables (Group A might just have naturally better memories than Group B).

Repeated Measures: The same participants take part in all conditions.
Strength: No participant variables; fewer participants needed.
Limitation: Order effects (they might get better at the task the second time because of practice).

Matched Pairs: Participants are paired together based on a variable (e.g., IQ or age). One person from the pair does Condition A, the other does Condition B.
Strength: Reduces participant variables and avoids order effects.
Limitation: Very difficult and expensive to match people perfectly.

Key Takeaway: To fix "order effects" in repeated measures, researchers use counterbalancing (half do condition A then B; the other half do B then A). This "balances out" any practice effects.

5. Controlling the Study

To make a study "scientific," we need to control for bias.

Demand Characteristics: When participants figure out the aim of the study and change their behavior to "help" the researcher (the "please-U effect") or "break" the study (the "screw-U effect").

Investigator Effects: When the researcher’s own behavior (conscious or unconscious) influences the participants' responses (e.g., smiling more when a participant gives the "right" answer).

Ways to improve control:
- Randomisation: Using chance to decide the order of tasks or the allocation of participants to groups.
- Standardisation: Ensuring every participant has the exact same experience (e.g., identical instructions read from a script).

6. Pilot Studies

A pilot study is a small-scale "practice run" of an investigation.
Why do it? To check if the procedure works, if participants understand the instructions, and if the timings are right. This saves time and money in the long run because it allows the researcher to fix any "bugs" before the real thing!

Don't worry if this seems like a lot of definitions! The best way to learn these is to apply them. Ask yourself: "If I were testing if music helps revision, who would I pick (sampling), how would I group them (design), and what exactly would I measure (operationalisation)?"

Quick Review Box

Directional Hypothesis: Predicts the specific way results will go.
Random Allocation: Randomly putting people into Group A or B to avoid bias.
Counterbalancing: An ABBA technique to stop order effects in repeated measures.
Operationalisation: Making variables measurable and specific.