Manipulate the IV, measure the DV (Artificial Task/enviroment)
+ High levels of control so easy to replicate making it more reliable.
+ High levels of controls means "cause and effect" can be established
-Artificial situation so low in ecological validity,less applicable to real life, so less useful.
- Participants may produce demand characteristics so results may be less valid as not measuring true behaviour.
Manipulates the IV,meaures the DV, Natural enviroment.
+ Natural environment means greater ecological validity so more
applicable in real life. + Lower demand characterisiics (if people are not aware of being studied) so results more valid.
- Harder to control so difficult to replicate therefore results may not be reliable.
- May be unethical (if people not aware that they are being studied)
IV naturally occuring,DV measured, natural environment.
+ High in ecological validity as natural environment,so more findings more applicable to real life so more useful.
+ Allows researchers to investigate areas thats would otherwise be unethical or impossible to manipulate.
- Lack of controls so difficult to establish cause and effects due to extraneous variables.
- Lack of controls mean results are less reliable.
Most experimental research starts with an experimental hypothesis, also known as alternative hypothesis.
The experimental hypothsis tests the cause of independent variables (IV) on the effect of the dependent variable (DV).
For example, "People who drink alcohol before driving will make significantly more driving mistakes than those who don't drink alcohol before driving."
In this example whether they have drunk alcohol or not before driving is the IV, the variable manipulated to see its effects on the DV, peoples driving performance.
One Tailed and Two Tailed Hypothesis
One tailed hypothesis, or directional hypothesis predicts the direction in which the results are expected to occur. Eg. alcohol will worsen driving performance.
Two tailed hypothesis, or non directional hypothesis, doesnt predict the outcome but predicts that there will be an effect of the IV on the DV, it just doesnt state in which direction. Eg. alcohol will affect driving performance.
Null Hypothesis and Operationalising Variables
A null hypothesis is a hypothesis that predicts that the IV will have no effect on the DV.
Eg. There will be no significant difference in the reading abilities of children whether they have blue eyes or brown eyes. Any difference is due to chance.
Operationalising the variables just means to define clearly HOW the IV is being manipulated and HOW the DV is being measure within a hypothesis.
Extraneous and Confounding Variables
Extraneous variables are variables which have not been manipulated by the experimenter but that could still effect the DV. Eg. emotional state, gender, age,light,noise,heat etc)
Confounding variables are extraneous variables which have not been kept constant and have confounded the experimenters attempts to find cause and effect because they have made it so the IV is not the only thing effecting the DV.
Independent measures design involves using different participants in each of the experimental conditions. The two groups of participants in this type of design will be exposed to different kinds of treatment. One group will be exposed to condition one, the other to condition two.
+ No order effects eg. practice, boredom, fatigue.
+ Participants cant guess the aim of the research because they only see one condition so low in demand characteristics.
- Increase in extraneous variables due to individual differences so low in validity because we may be measuring individual responses to the conditions and not the effect of the IV on the DV.
This involves using the same participants in both conditons.
+ No individual differences because the participant is compared against him/herself. This makes it more valid as we can be sure we are measuring the effect of the IV on the DV and not indiv diff.
- Order effects can effect the results because it is possible for the participants to perform better in the second condition due to practice, or possibly worse due to fatigue or boredom. This makes it less valid as we may just be measuring the effect of the order of conditions on the p's and not the IV on the DV.
Matched Pairs Design
This involves matching each participant to another as closely as possible on the variables the are considered relevant and important to the experiment. Eg age, gender, intelligence,memory. One of each pair goes into condition one, the other goes into the second condition.
+ Low in individual differences as particiapnts have been matched on important variables, therefore more valid.
+ No order effect as only one of each pair goes into each condition.
- Very difficult and time consuming to find a match.