Types of Experiments

Key Features

·         Controlled Environment (extraneous variables controlled)

• --> don't affect results
• More certain that the IV causes change in the DV
• Extraneous variables --> therefore not confounding variable (not controlled: could affect DV)

·         P’s controlled by experimenter – know they are taking part

• Put into groups (IGD, RMD, MPD)

·         Experimenter controls IV too

ADVANTAGES

-       You can establish cause and effect relationships

-       If appropriate care taken, other researchers can find same findings = replication

-       Variables much easier to control in a lab

 -       Extraneous variables easier to control – results more reliable (likely to be replicated) – confounding variable do not affect to DV so certain that IV --> change in the DV

-       High ecological validity – realistic behaviour and results, they behave naturally

DISADVANTAGES

-       The more you control behaviour, less natural it becomes, lack of ecological validity

-       Not everything of interest can be investigated in a lab

-       Some bias results (sampling, demand characteristics) reduces the confidence in findings

-       Low ecological validity – not normal behaviour – behave differently in the lab as they know they are being studied SOCIAL DESIRABILITY BIAS

-       Demand characteristics – try work out aim of study, deliberately change to get it wrong ‘screw you effect’

-       Biased sample – not representative of population 

Field Experiments

·         Occurs in natural place/environment e.g homes, shopping centre, workplace
·         Experimenter manipulates IV to see if it affects DV e.g Piliavin et al (69) New York Subway
·         Experimenter manipulated characteristics of actor

 ·         Natural setting

ADVANTAGES

-       The p’s are in their natural habitat so there would be a greater ecological validity therefore there is more realistic behaviour with no demand characteristics

DISADVANTAGES

-       There is less control of extraneous variables and results can be confounded in some way

-       They’re more time consuming and expensive to design and run

-       Replicable?  

Natural Experiments

·         Experimenter does not manipulate the IV variable
·         Studies the effect of naturally occurring IV to see if it effects the DV
·         Comparing 2 existing naturally occurring groups of p’s

 ·         Not necessarily in natural environment, usually know they are being studied

ADVANTAGES

-       Can study impact ore effect of IVs- would be unethical to manipulate (Amputations)

-       Because IV is naturally occurring we often study behaviour (e.g school) more ecological validity and realistic behaviour

DISADVANTAGES

-       Lack of control over EVs likely to be CVs and affect DV – not sure if IV à Change in DV

-       Participant variables between 2 groups so might not be IV causing difference. Naturally different anyway, regardless of IV e.g more likely to be military or other high risk job with control group = less likely

-       Can’t set up experiment – have to wait for it to occur

-       Not easy, convenient, time pressure