What a laboratory experiment is
Definition
Laboratory experiment
An experiment carried out in a controlled setting where the researcher manipulates the independent variable and keeps other variables as constant as possible.
- A laboratory experiment is run in a setting you control, so that everything except the independent variable can be held steady.
- You decide what the independent variable will be and you set it yourself, then measure the dependent variable.
- What makes it a laboratory experiment is the level of control, not the room it happens in.
- A school hall with the blinds down, the phones collected and a stopwatch running is controlled enough to count.
Control bought at the cost of realism
Definition
Ecological validity
How well the findings of a study apply to real, everyday settings rather than only to the artificial conditions of a laboratory.
- The more tightly you control a situation, the less it resembles the messy places where behaviour normally happens.
- A study with high control therefore tends to have lower ecological validity.
- That is a trade you make deliberately, not a mistake in the design.
Common Mistake
- Do not write that a laboratory experiment is unrealistic and stop there, because that is description rather than evaluation.
- Do not label every indoor study a laboratory experiment, because control decides the method and the building does not.
Strengths
- Cause and effect can be established, because holding the extraneous variables steady means a change in the dependent variable is very likely to have come from the independent variable.
- The study is easy to replicate and to measure precisely, because the procedure is standardised and written down and behaviour can be timed or counted objectively.
Example
- Suppose you want to know whether background music harms revision.
- You book a quiet classroom, give every pupil the same passage and the same ten minutes, and play music to one group and nothing to the other.
- The music is the only thing you allow to differ, so any gap in test scores points back to it.
Weaknesses
Definition
Demand characteristics
Cues in a study that let participants guess its aim, so they change their behaviour instead of acting naturally.
- The task can be artificial, since learning a list of unrelated words in a booth is nothing like revising for a real exam, so the findings may not transfer to everyday life.
- Participants respond to demand characteristics, because they know they are being studied and may behave the way they think you want rather than naturally.
When to choose a laboratory experiment
- Choose it when your question is about cause, because no other method lets you rule out rival explanations so firmly.
- Choose it when the behaviour can be measured precisely in a short session, such as reaction time, recall or attention.
- Choose it when you want other researchers to repeat the study and check your result.
- Choose a different method when the whole point is how people behave in ordinary life, because the control that helps you here would destroy that.
Turning this into marks
- Name the strength or weakness and then say what it buys or costs you.
- Use the words control, cause and ecological validity, since those are what a marker is looking for.
- The mark most often lost is writing that the method has high control without adding that control is what makes a claim about cause believable.
Self review
- What is the defining feature of a laboratory experiment?
- Why does tight control make a claim about cause more convincing?
- Explain the trade-off between control and ecological validity.
- Give a research question you would answer with a laboratory experiment, and say why it suits one.
