what is an experiment
an experiment is when the researcher deliberately changes one thing in order to find out if it has an effect on another thing
why are experiments different and important
they tell you wether one variable has a causal influence on another variable
what is an independent variable (IV)
variable that is manipulated as we think it links to the DV
what is a dependant variable (DV)
variable that is measured
what can we conclude from the IV and DV
-if IV is manipulated and there is no change to the DV, there can be no causal relationship between the 2
-if the DV changes when IV is manipulated, this suggests that changes in the DV was caused by changes in the IV
what is operationalisation
Explaining precisely how IV can be manipulated and how DV can be measured.
Why is operationalisation important?
-enables replicability
-removes subjectivity and bias increasing the reliability
what is a confounding variable
any uncontrolled variable that affects the conditions differently eg being certain that changing the IV was the only different between the conditions not another variable
what are 2 possible problems with the existence of a cofounding variable
1-any difference you find in the DV between the conditions might have been caused by the uncontrolled variable so you might conclude that the IV has affected the DV when it hasn’t
2-uncontrolled variable may cancel out the effect of the IV so you end up concluding that the IV had no effect on the DV when it actually did
what are the three different types of experiments
independent groups, matched pairs and repeated measures
explain how independent groups work
1-recruit a sample
2-divide them into 2 groups randomly
3-each pp is then exposed to only one condition
4-the DV is measured and the averages are compared
what are some strengths of the independent groups design?
1-it’s easier to keep the conditions the same meaning it’s also easier to control confounds
2-pp are less likely to guess the aim as they only take part in one condition, they don’t know about the other condition so demand characteristics are reduced
what are some weaknesses of the independent groups design?
1-groups will somewhat differ from each other as your using different participants for each condition
2-have to recruit twice as many pp
3-individual difference will mean that there will be variables between the pp so c=any comparisons made between the 2 groups are unreliable
describe the repeated measures design
1-recruit a sample
2-both groups are exposed to both conditions
what are some strengths of the repeated measures design
-fewer pp are needed as they are used twice so it is more economical than the independent group design
what are some weaknesses of the repeated measure design
1-chance of pp displaying demand characteristics as they have knowledge of all the conditions and so more likely to guess the aim
2-practice/order effect; performance from one condition is influenced by the previous condition eg practice and fatigue which introduces new confounds
how can we fix the problems from a repeated measures design
1-using diff materials for each condition to remove the effect of practice however they all need to be equivalent or it introduces a new confound
2-reduce demand characteristics by using a single blind technique
3-control order effects by using randomisation or counterbalancing to ensure pp experience the conditions in diff orders
what is randomisation
selecting randomly which condition the pp does first
what is counterbalancing
An attempt to control for the effects of order in a repeated measures design: half the participants experience the conditions in one order, and the other half in the opposite order. so AB then BA
describe how the matched pairs design works
1)recruit a sample
2)look at them individually and recruit a matched sample to each pp
3)randomly assign them so person 1 from the pair does condition one first and person 2 does condition 2 first
what are some strengths of the matched pairs design
ensures that conditions can be compared reliably and that any difference found between the results of each condition is likely due to the DV and not individual differences allowing causation to be established
what are some weaknesses of the matched pairs design
1)time consuming to recruit 2 well matched samples
2)difficult to match pp on all possible characteristics that could effect the DV
3)small sample as if one drops out you actually lose 2 pps
why is a matched pairs design better than both ID and RM
-reduces demand characteristics
-reduces any individual differences
-reduces the effect of practice and order
what are some strengths of correlational studies over experiments
-often occur in a natural setting without manipulating variables so more reflective of real life behaviour