Experimental designs Flashcards
Repeated Measures design
Participants take part in both conditions of the IV (both lvls of IV)
A03 - Strengths of repeated measures
Pps are the same so you have no differences between people –> reduced participant variable. e.g. IQ
A03 - Weakness of Repeated measures
Order of the conditions may affect performance - “order effects” (confounding variable)
1. Practice effect (do better in cond 2 as learned from 1)
2. Fatigue effect (boredom, fatigue)
3. Increase in demand characteristics
They are the different effects not a list
How to deal with order effects
Counterbalancing - order in which the conditions are encountered are BALANCED out
* con A = real words con B = nonsense words
* AB or BA (then you can allocate which group does which)
* Half of the pps experience the conditions in one order while the other in the opposite
Independent groups
Pps are randomly allocated to one condition of the IV. One group undergoes one condition and the other group undergoes the other.
A03 - strengths of Independent groups
- No order effects - only done it once
- same materials can be used in cond A+B - meaning a test (e.g. word list) does not need to be standardised/assessed for lvl of difficulty.
- reduced demand characteristics
A03 - Weakness of independent groups
Researcher cannot control the effects of participant variables
* individual differences - e.g. IQ
* Ps in cond A may be different to those in cond B
Solution to independent groups
Random allocation - Ps can be allocated to diff groups to ensure an even spread of the participants characteristics
* e.g. done by hat/online randomiser
Matched pairs
Pps are matched based on similarities or key characteristics
e.g. similar age, gender, IQ
* believed to have an influence on the DV
* The pps pairings will be then allocated to diff conditions (in attempt) to control confounding variable of participant variables
A03 - Strengths of matched pairs
Reduce the num of participant variables (IQ, age, gender, memory test)
A03 - weakness of matched pairs
Difficult to reduce all the participant variables, have differences unbeknown to the researcher