Difficulty: Introductory | Prerequisites: None
This chapter lays the groundwork for every study you will encounter in the course. Social psychology is an empirical science, and the methods here (experiments, correlations, surveys) determine what claims researchers can and cannot make. If you understand why random assignment matters and what internal validity means, the rest of the course clicks into place. If you have missed early lectures, start here.
Social psychologists use experiments, correlations, surveys, and field studies to work out how people think, feel, and behave around others. The critical distinction is between correlational research (which finds relationships but cannot prove causation) and experimental research (which can, thanks to random assignment and controlled variables). Understanding internal validity, external validity, and statistical significance is essential for evaluating every study you will read this term.
Theory
An integrated set of principles that explain and predict observed events. Theories summarise observations and generate clear predictions.
In simple terms, a theory is a big-picture explanation for why something happens.
Hypothesis
A testable proposition that describes a relation that may exist between events. Hypotheses allow us to test a theory.
Think of it as a specific, checkable prediction you derive from a theory.
Random sample
A sample in which every person in the population has an equal chance of being selected for study. The sample should represent the population.
In simple terms, everyone has a fair shot at being picked, so the results reflect the wider group.
Random assignment
The process of assigning participants to the conditions of an experiment such that every person has the same chance of being in any given condition. It eliminates extraneous factors and defeats systematic bias.
Think of it as shuffling participants into groups by coin flip so that pre-existing differences wash out.
Independent variable (IV)
The experimental factor that a researcher manipulates.
In simple terms, the thing the experimenter changes on purpose.
Dependent variable (DV)
The variable being measured, so called because it depends on manipulation of the independent variable.
Think of it as the outcome you are watching for.
Internal validity
The ability of an experiment to establish the hypothesised cause-and-effect link. Confounds reduce internal validity by creating alternative explanations.
In simple terms, can you be confident that X caused Y, or could something else explain it?
External validity
The extent to which an experiment's results can be generalised to the larger population. Depends on the representativeness of the sample.
Think of it as: do these lab findings hold up in the real world?
Demand characteristics
Cues in a study that make participants think they need to give the response the researcher wants to see.
In simple terms, participants figure out what the experimenter is looking for and adjust their behaviour accordingly.
Mundane realism
The degree to which an experiment is superficially similar to everyday situations.
Experimental realism
The degree to which an experiment absorbs and genuinely involves its participants. Experimental realism is considered more important than mundane realism.
Think of it as: does the participant take the task seriously, regardless of how "lifelike" it looks?
Correlational research
The study of naturally occurring relationships among variables, asking whether two or more factors are naturally associated. Cannot draw cause-and-effect conclusions.
Experimental research
Studies that seek clues to cause-and-effect relationships by manipulating some factor to see its effect on another.
Deception (in research)
When participants are misinformed or misled about a study's methods and purposes, sometimes used to achieve experimental realism.
Random error
Uncontrolled, chance events (e.g. a pen dropping during one session but not another) that may affect results. Random assignment defeats systematic bias but does not eliminate random error.
Behaviour = f(person, situation)
Person = individual differences (personality, attitudes, genetics)
Situation = the social context surrounding the person
This formula (attributed to Kurt Lewin) is the foundational equation of social psychology: behaviour is always a product of both who someone is and where they find themselves.
Lab vs. field research
Laboratory research takes place in a controlled setting with maximum control over variables.
Field research takes place in natural, real-life settings with incomplete control over variables.
Lab research favours internal validity; field research favours external validity.
Correlational vs. experimental research
Correlational research examines naturally occurring relationships among variables.
Advantage: examines important variables in natural settings.
Disadvantage: cannot determine cause and effect; interpretation is ambiguous.
Experimental research manipulates one factor to observe its effect on another.
Advantage 1: random assignment ensures conditions are not confounded by participant differences.
Advantage 2: high degree of control allows cause-and-effect conclusions.
The best research uses both strategies to capitalise on their respective strengths.
Surveys must be constructed carefully to avoid biasing responses.
Order of questions matters.
Wording matters.
Framing: the way a question is phrased affects how people respond.
Random assignment is not the same as random sampling.
Random sampling = who gets into the study (representativeness).
Random assignment = how participants already in the study are sorted into conditions (causal inference).
Experiments use at least two groups: one receives the experimental treatment, the other (the control group) does not.
Internal validity: the ability of an experiment to establish the hypothesised cause-and-effect link.
Confounds threaten internal validity by introducing alternative explanations.
External validity: the extent to which results generalise to the broader population.
Depends on how representative the sample is.
Experimental research (especially lab research) tends to maximise internal validity at the expense of external validity.
Correlational research tends to maximise external validity at the expense of internal validity.
Mundane realism: superficial similarity to everyday life.
Experimental realism: the experiment absorbs and involves its participants.
Experimental realism is more important. An experiment can look nothing like daily life and still produce genuine psychological engagement.
Achieving experimental realism sometimes requires deception.
Even with random assignment, a difference between conditions could be due to chance.
Statistical tests estimate this probability. The conventional benchmark is p < 0.05, meaning you would see the difference fewer than 1 time in 20 by chance alone.
Replication, both within and across labs, is essential for confidence in findings.
Students often confuse random assignment with random sampling. Random sampling is about who enters the study; random assignment is about how those participants are sorted into conditions. You can have one without the other.
Students often assume that correlation implies causation. A correlation between two variables tells you they move together; it does not tell you that one causes the other. There may be a third variable driving both.
Students sometimes think mundane realism is the most important quality of a good experiment. Experimental realism (genuine psychological engagement) matters more. A study can look artificial and still produce valid results.
The p-value threshold is p < 0.05, not p < 0.5. The source notes contain a typo ("p < 0.5"). The standard in psychology is p < 0.05, meaning fewer than 5 chances in 100, not 5 in 10.
⚠️ Know the difference between random assignment and random sampling cold. Expect a multiple-choice question that tests whether you can tell them apart.
⚠️ Be able to identify the IV and DV in a described experiment.
⚠️ Understand the trade-off between internal and external validity, and which type of research maximises which.
⚠️ "Correlation does not equal causation" is a staple exam point. Be ready to explain why, using the concept of third variables.
⚠️ Know what demand characteristics are and how they threaten an experiment's validity.
⚠️ Experimental realism vs. mundane realism: know which one matters more and why.
True or False: Random assignment ensures the sample is representative of the general population. (False. Random assignment sorts participants into conditions; random sampling is what makes a sample representative.)
Fill in the blank: The variable a researcher manipulates is the ______ variable. (Independent)
True or False: Correlational research can establish cause and effect. (False.)
Fill in the blank: The conventional threshold for statistical significance in psychology is p < ______. (0.05)
True or False: Experimental realism is more important than mundane realism. (True.)
Q: A researcher finds that ice-cream sales and drowning rates are positively correlated. Can she conclude that eating ice cream causes drowning? Why or why not?
A: No. Correlation does not establish causation. A third variable (hot weather) likely drives both ice-cream sales and swimming frequency, which in turn increases drowning risk.
Q: In an experiment, a researcher randomly assigns participants to watch either a violent or a non-violent film clip and then measures aggression levels. Identify the IV and DV.
A: The IV is the type of film clip (violent vs. non-violent). The DV is the measured aggression level.
Q: Why is random assignment important in experiments?
A: Random assignment ensures that each participant has an equal chance of being placed in any condition, which eliminates systematic differences between groups. This means any observed effect on the DV can be attributed to the IV rather than to pre-existing participant differences.
Q: A lab study finds that people are more helpful after watching a kind act on screen. A critic argues the finding has low external validity. What does this mean, and how might researchers address it?
A: Low external validity means the results may not generalise beyond the controlled lab setting to real-world behaviour. Researchers could address this by conducting a complementary field study in a natural environment to see whether the same effect holds.
Q: What is the difference between mundane realism and experimental realism, and which matters more?
A: Mundane realism is how closely an experiment resembles everyday life on the surface. Experimental realism is how much the experiment psychologically engages participants. Experimental realism matters more, because a study can look artificial yet still produce genuine reactions if participants take it seriously.
This methodology chapter underpins everything that follows. When you read about the Milgram obedience experiments (Chapter 6), you will need to evaluate their internal validity, ethical use of deception, and experimental realism. The distinction between correlational and experimental evidence comes back in Chapter 4 (Attitudes and Behaviour) when the course asks whether attitudes cause behaviour or the other way round. Understanding the p < 0.05 benchmark prepares you for interpreting every study result in the course.
Research methods, social psychology methodology, theory vs. hypothesis, random assignment, random sampling, independent variable, dependent variable, IV, DV, correlational research, experimental research, lab research, field research, internal validity, external validity, demand characteristics, mundane realism, experimental realism, confounds, control group, statistical significance, p-value, replication, deception in research, representative sample, survey design, framing effect, random error, NYU social psychology, Chapter 1