Types of experiments
Laboratory experiments
These are highly controlled experiments carried out in an artificial setting. Any location can be a laboratory, but it must be one in which extraneous variables such as noise, temperature, light, seating arrangements, etc can be kept constant for all participants. Any tasks participants take part in tend to be artificial and unrepresentative of everyday life (e.g. watching film clips as opposed to real events, memorising lists of random words, etc). The important factor is that the researcher has complete control over the independent variable.
Field experiments
A field experiment is carried out in a more naturalistic setting than a laboratory experiment. This means that the tasks carried out by participants are more representative of everyday life, but there is still some degree of artificiality because the researcher still has complete control of the independent variable.
Natural experiments
Natural experiments are the most realistic way of researching human behaviour in everyday life as participants are in their natural environment carrying out their normal everyday tasks. The researcher has absolutely no control of the independent variable and must instead select one that happens to naturally occur in the environment. As the researcher cannot control what the participants are doing then there is likely to be a large number of extraneous variables.
Comparison of the strengths and weaknesses of laboratory, field and natural experiments
Experimental designs
Independent groups
In an independent groups design, each participant is only assigned to one condition of the independent variable. There can be several groups of participants, but each group only takes part in one condition of the IV and does not repeat anything.
Strengths of independent groups designs are that each participant only takes part in one condition of the experiment, meaning the dependent variable will not be influenced by order effects (practice, boredom or fatigue). Participants are also less likely to get fed up and drop out as they only take part in one of the tasks.
The main weakness of independent groups designs is that the participants in one group may be very different on some random variable from the other group, e.g. one group may just happen to have a higher average IQ than the other group, and so the results may lack validity. The design also requires twice as many participants as a repeated measure design.
Repeated measures
In a repeated measures design, each participant is assigned to more than one condition of the independent variable. The experimental groups consist of exactly the same participants repeating the same task but under a different condition.
Strengths of repeated measures designs are that both groups consist of exactly the same participants, and so there can be no difference between the groups in variables such as average IQ. The study can also be carried out with half as many participants as in an independent groups design, making it useful when participants are difficult to obtain.
The main weakness of repeated measures designs is that there are order effects (e.g. practice, boredom and fatigue). As participants are performing more than one task then because they have had some practice, become bored or tired it means that one condition could be performed differently for a reason other than the independent variable. It is possible to reduce the negative impact of order effects by counterbalancing, which means varying the order in which participants take part (e.g. group 1 does task A followed by task B and group 2 does task B followed by task A, which means that any effects of practicing the task, become bored or tired in equal for both conditions of the independent variable).
Matched pairs
In a matched pairs design, there are two equal groups of participants and each participant is matched with a similar participant in the other group (e.g. age, IQ, occupation). Both groups then take part in different conditions of the independent variable, as with independent groups.
Strengths of matched pairs designs are that it allows researchers to control for participant variables (a type of extraneous variable) that might otherwise affect the result, without introducing order effects as in repeated measures.
Weaknesses to matched pairs designs are that it is a time consuming method to employ and it may not be possible to truly match all participants in one group with a suitable partner in the other group.
Summary of the strengths and weaknesses of experimental designs
A Level exam tips
Answering exam questions (PSYA1 AQA A specification)
Questions about types of experiments tend to be 1 or 2 mark questions that require you to:
- Identify the type of experiment in an example
- Give 1 strength and 1 weakness of that type of experiment
- As only one strength or weakness will be required in the answer, it is sensible to learn 2 for each type of experiment (6 in total) and be able to explain them with a brief example.
- Identify the experimental design in an example
- Give 1 strength and 1 weakness of that experimental design
- As only one strength or weakness will be required in the answer, it is sensible to learn 2 for each design (6 in total) and be able to explain them with a brief example.