Experimental Designs

• independant groups design

  - 2 separate groups of Ps experience 2 different conditions of the experiment

  -experimental or control (not both)

  2 groups behaviour is compared
• + of independent groups design

  no order effects - Ps are less likely to guess the aims
• - of independent groups design

  participant variables - Ps are different in each group (personalities) DV may be due to PV rather than IV
• random allocation

  solves participant variables in independent groups by allocating Ps to experimental or control groups using random techniques and attempting to distribute PV evenly
• repeated measures design


  - all Ps experience both conditions (experimental/control condition)

  - 2 sets of data compared to see differnce
• + of repeated measures design

  participant variables are controlled
• - of repeated measures design

  order effects

  - each P has to do 2 tasks so the order completed may cause issue with findings, 1st condition may have an impact on 2nd

  - repeating task may be boring/tiering so performing the 2nd task may not be completed as accurately by Ps

  - performance may improve with practice, Ps may do better on the 2nd task

  - Ps may work out the aim of the study, experience demand characteristics
• counterbalancing


  solves order effects in repeated measures - as half Ps do condition A then B and half do condition B then A (ABBA)
• matched pairs design

  - Ps are paired together on a common variable/s relevant to the experiment e.g. IQ in memory test

  - 1 P from each pair does different conditions

  - often requires pre-test
• + matched pairs design

  Ps only do 1 condition - order effects/demand characteristics aren't a problem

  and it reduces PV
• - matched pairs design

  Ps can never be matched exactly - differences may effect DV and its time consuming/expensive if pre-test required