Experimental method: laboratory, field, natural and quasi-experiments. Aims, hypotheses, independent and dependent variables, operationalisation, extraneous and confounding variables.

What this dot point is asking

AQA wants you to describe the four types of experiment, aims and hypotheses, variables, operationalisation, and extraneous and confounding variables. The exam skill is to write fully operationalised directional and non-directional hypotheses and to distinguish the four experiment types and the two kinds of nuisance variable.

Types of experiment

The four types of experiment differ along two dimensions: where they take place and whether the researcher manipulates the independent variable. A laboratory experiment manipulates the IV in a tightly controlled environment, which gives high control over extraneous variables and high replicability (its strengths) but lower ecological validity and a greater risk of demand characteristics (its weaknesses). A field experiment also manipulates the IV but does so in a natural, everyday setting, gaining ecological validity and reducing demand characteristics, at the cost of less control over extraneous variables and harder replication. A natural experiment does not manipulate the IV at all; instead it uses an IV that varies naturally (for example, the introduction of television to a region), which lets researchers study variables that could not ethically be manipulated, but it can rarely be replicated and the lack of manipulation weakens causal claims. A quasi-experiment uses an IV that is a fixed, pre-existing characteristic of the participants (such as gender or having a diagnosis), which again cannot be manipulated, so cause and effect cannot be firmly established and participants cannot be randomly allocated. The common error is to confuse natural and quasi: the difference is that a natural experiment's IV is an event in the world, while a quasi-experiment's IV is a characteristic of the people.

Hypotheses and variables

An aim is a broad statement of what the researcher intends to investigate, while a hypothesis is a precise, testable prediction of the outcome. A directional (one-tailed) hypothesis predicts the direction of the difference (for example, caffeine will make reaction times faster) and is used when previous research points one way; a non-directional (two-tailed) hypothesis predicts only that there will be a difference, without specifying which way, and is used when there is no clear prior evidence. Both the independent variable (the one manipulated) and the dependent variable (the one measured) must be operationalised, meaning defined clearly enough to be measured, so "aggression" might be operationalised as the number of physical hits in ten minutes. Variables that are not the IV but could affect the DV must be considered. An extraneous variable is any nuisance variable that might affect the DV and so add random error; if controlled, it does not bias the results. A confounding variable is more serious: it varies systematically with the IV, so it changes alongside the conditions and provides an alternative explanation for the results, undermining the validity of the experiment.