Difficulty: Foundational | Prerequisites: None
Social psychology uses the scientific method to study how people think about, influence, and relate to one another. Researchers choose from several methods (surveys, correlations, experiments, observation) depending on whether they need to describe, predict, or explain behaviour. The key distinction is that only true experiments, with random assignment and controlled variables, can establish cause and effect.
Social psychology
The scientific study of how people's thoughts, feelings, and behaviours are influenced by the real or imagined presence of other people.
Operationalise
To translate an abstract idea or construct into a concrete, measurable behaviour or variable. Think of it as turning a fuzzy concept ("social anxiety") into something you can actually count or score.
Construct validity
The degree to which a measure or manipulation truly captures the theoretical concept it is supposed to represent. In simple terms, are you measuring what you think you are measuring?
Independent variable (IV)
The variable the researcher manipulates or controls in an experiment. Think of it as the ingredient you change to see what happens.
Dependent variable (DV)
The outcome variable that is measured to see whether the manipulation had an effect. In simple terms, this is what you are watching for a change in.
Confound
An extraneous variable that changes alongside the independent variable, making it impossible to tell which one caused the effect. Think of it as an uninvited variable crashing your experiment.
Mediator
A variable that explains why or how the independent variable produces its effect on the dependent variable. In simple terms, it is the mechanism in the middle of the causal chain.
Interaction effect
When the effect of one variable depends on the level of another variable (e.g., a treatment works differently for men and women). Think of it as "it depends" with data behind it.
Correlation
A statistical measure of the relationship between two variables. It tells you whether and how strongly two things move together, but not whether one causes the other.
Random assignment
Allocating participants to conditions by chance so that each person has an equal probability of being in any group. This is the feature that lets experiments claim cause and effect.
Demand characteristics
Cues in a study that suggest to participants what the experimenter expects to find, potentially altering their behaviour. Also called the Hawthorne Effect in some contexts.
Social psychologists follow a four-step cycle:
Develop a theory or idea, then operationalise the constructs and assess construct validity (does the IV really manipulate what you intend? does the DV capture what you mean to measure?).
Formulate a testable hypothesis.
Gather evidence to test the hypothesis.
Update the theory based on findings.
This cycle repeats: findings feed back into refined theories and new hypotheses.
Archival research
Used for examining cultural and historical trends by analysing existing records. Strength: accesses data that already exists. Weakness: limited to what was recorded, and you cannot control how it was collected.
Naturalistic observation
Watching people in their natural settings without intervening. Strengths: describes real behaviour, avoids disrupting what you are studying, useful when manipulation would be unethical. Weaknesses: does not explain why behaviour happens, raises ethical concerns (privacy, lack of informed consent), and different observers may record things differently (reliability problem).
Survey
Asking people to report their own attitudes, beliefs, or behaviours. Strengths: direct, cheap, quick, can reach large samples. Weaknesses: response bias, socially desirable responding (people say what looks good), and wording effects (small changes in phrasing shift answers).
Correlational research
Measuring two variables to find the relationship between them. Strengths: tests relationships that cannot be ethically manipulated, allows prediction. Weakness: correlation does not equal causation. A relationship between A and B does not tell you whether A causes B, B causes A, or a third variable C drives both.
Experiment
Manipulating an IV and measuring a DV, with participants randomly assigned to conditions. Strengths: control over variables, can eliminate extraneous variables, establishes cause and effect. Weaknesses: confound variables, experimenter bias, participant bias, and demand characteristics (Hawthorne Effect, where participants alter behaviour because they know they are being studied). Can be conducted in the lab or in the field.
Brain imaging techniques, psychophysiological measures (heart rate, blood pressure, skin conductance), and hormone measures (cortisol, testosterone) give researchers access to biological processes that self-report cannot capture. These complement traditional methods by providing objective physiological data.
Confound: An unwanted variable that varies with your IV, muddying the causal picture.
Mediator: The mechanism that explains why the IV affects the DV. Example: smoking leads to cancer, and the mediator might be the cellular damage caused by carcinogens.
Interaction effect: The effect of one variable changes depending on the level of another variable (e.g., a drug works differently across age groups).
Students often think correlation can show causation if the correlation is strong enough. It cannot. Strength of correlation has nothing to do with whether one variable causes the other.
Students confuse confounds with mediators. A confound is a problem (an unwanted variable contaminating your results). A mediator is a feature (the explanatory mechanism you are looking for).
Students assume experiments are always better than other methods. They are better for establishing causation, but they can lack ecological validity, may not be ethical for certain questions, and are not always practical.
Students forget that demand characteristics affect experiments too. Participants are not passive; they try to figure out what the study is about and may change their behaviour accordingly.
⚠️ Be able to identify which method is being described in a scenario and name its strengths and weaknesses.
⚠️ Know the difference between a confound, a mediator, and an interaction effect. These are frequently tested as short-answer or multiple-choice.
⚠️ "Correlation does not equal causation" appears in some form on nearly every social psychology exam. Be ready to explain why with an example.
⚠️ Understand construct validity for both the IV and the DV. Can you spot whether a study is measuring what it claims to measure?
⚠️ The Hawthorne Effect (demand characteristics) is a classic exam topic. Know the definition and be able to give an example.
True or False: A strong positive correlation between ice cream sales and drowning rates proves that eating ice cream causes drowning.
False. Correlation does not establish causation. A third variable (hot weather) likely drives both.
Fill in the blank: Translating an abstract concept into a measurable variable is called ______.
Operationalising.
True or False: Naturalistic observation allows researchers to determine why people behave a certain way.
False. It describes behaviour but does not explain it.
Fill in the blank: The ______ Effect refers to participants changing their behaviour because they are aware of being studied.
Hawthorne (demand characteristics).
True or False: A mediator is an unwanted variable that contaminates the results of an experiment.
False. That describes a confound. A mediator explains the mechanism behind the effect.
Q: A researcher finds that students who sit in the front row get higher grades. Can she conclude that sitting in the front causes better grades? Why or why not?
A: No. This is a correlational finding. Students who choose the front row may already be more motivated, which is a confounding variable. Only an experiment with random assignment to seating could establish causation.
Q: What is construct validity, and why does it matter?
A: Construct validity is whether a measure or manipulation truly captures the theoretical concept it claims to represent. It matters because if your IV does not manipulate the construct you intend, or your DV does not measure it, your entire study's conclusions are undermined.
Q: Name one strength and one weakness of the survey method.
A: Strength: it is a direct, cheap, and quick way to collect data from large samples. Weakness: it is vulnerable to response bias, including socially desirable responding and wording effects.
Q: Explain the difference between a mediator and a confound, using an example.
A: A confound is an unwanted variable that co-varies with the IV, making it unclear what caused the effect. A mediator is the mechanism through which the IV affects the DV. Example: if exercise reduces depression, the mediator might be increased endorphin release. A confound would be if exercisers also happened to get more sunlight, and sunlight, not exercise, drove the mood improvement.
Q: What is an interaction effect? Give an example.
A: An interaction effect occurs when the impact of one variable depends on the level of another. Example: a new teaching method might improve performance for younger students but have no effect on older students. The effect of the teaching method interacts with age.
Research methods underpin everything else in the course. When you study attribution errors (Perception), ask yourself: how were those effects demonstrated experimentally? When you encounter the IAT in the Prejudice unit, notice it is a clever operationalisation of implicit attitudes. Understanding methods helps you critically evaluate every study you encounter.
Scientific method, operationalise, operationalize, construct validity, internal validity, external validity, ecological validity, independent variable, dependent variable, confound, confounding variable, third variable problem, mediator, moderator, interaction effect, correlation, causation, random assignment, archival research, naturalistic observation, survey method, response bias, social desirability, wording effects, experiment, field experiment, lab experiment, Hawthorne Effect, demand characteristics, experimenter bias, participant bias, brain imaging, psychophysiology, cortisol, testosterone