Experiments

• A test carried out in controlled conditions in an artificial setting, aiming to establish a cause and effect relationship between 2 variables
• These are rarely used in sociology because of practical and ethical problems
• Positivists are more likely to favour them because they produce statistical data, however they acknowledge that it could sometimes be hard to carry them out

Evaluation:

• Reliability = laboratory experiments are highly reliable because researchers can replicate them
• Practicality = it is impossible to control all the possible variables that can have an effect; laboratory experiments cannot be used to study the past; laboratory experiments are usually small- scale, so it is difficult to investigate large-scale phenomena
• Ethics = it can be difficult to obtain informed consent without revealing what the experiment is, which then leads to deception; some experiments can cause psychological harm to participants
• The Hawthorne Effect = participants are likely to alter their behaviour because laboratory experiments are not carried out in natural settings
• Free will = human behaviour is complex and it cannot be explained with cause and effect, people respond differently in different situations

• They still aim to find cause and effect, however they do it in a more natural setting
• The participants of such experiments are often unaware that they are being observed
• These experiments can produce statistical data (for positivists), however interpretivists might also go for it because it is carried out in a natural setting so it might be able to produce textual data

Evaluation:

• They are more "natural", valid and realistic, avoiding the artificial conditions of laboratory experiments
• The more natural the setting is, the less control the researcher has on different variables
• Some critics argue that they are unethical since they involve carrying out experiments without the participants being aware they are taking part

• A thought experiment that does not involve the researcher actually experimenting on real people but still designed to find cause and effect
• This involves looking at statistics
• Positivists favour the comparative method because it produces statistical data

Evaluation:

• It avoids artificiality
• It can be used to study the past
• It has no ethical issues because it doesn't involve real participants
• Even less control of variables, so it is harder to establish a cause and effect relatuionship