cognitive research methods

Seite 1

Types of Experiments

Lab experiments

-artifical

-iv manipulated and dv measured, fully operationalised

-standerdiesd procedure

-variables controlled

Evaluate

-easy to replicate; reliability

-easy to establish a valid cause and effect

-scientific credibility

-lack ecological validity

-particpants aware of their involvement which can lead to demand characteristics

Field experiments

-natural environment- tested where behaviour is normally displayed

-iv manipulated and dv measured

Evaluation

-greater in ecological validity

- ppts unaware of participation so no demand characteristics

-greater chance of extraneous variables impacting dv

-ethics: being unaware means no consent or right to withdraw

Seite 2

Experimental design

Independent Groups

participants are only involved in one of the experimental conditions

Evaluate

- reduces chance of demand characteristics

- need to recruite twice as many ppts

- may need to control ppt variables; random allocation allows for even distribution as they all have a equal chance of being selected

Repeated Measures

all participants take part in all conditions

Evaluate

fewer ppts needed so more economical

reduces issues with individual differences

increases demand characteristics

order effects- being influenced by the previous condition so are now practiced for fatigued in the next condition

Combating order effects

counterbalancing; ppts split in half and complete condition A then B and the other half do it in reverse order

Matched Pairs

ppts are matched on a certain characteristic(s) and assigned to different conditions

characteristics are established by pre-testing

Evaluate

conditions are compared more reliably

cause and effected; difference in results per condition is down to manipulation

time consuming

Seite 3

Types of Data

nominal data

grouped or categorical and values in each group are calculated

little would known about the individual differences- validity

ordinal data

data is ranked in order or position

doesn't show the difference between ranks

arbitrary scale so intervals are not equal

interval/ ratio data

the difference between each value is known and equal

gathered using a recognised scale/ psychological instrument

there is an absolute zero