1D - preventing error and bias

extraneous variables

any variable that is not the independent variable but may cause unwanted effect on dependent variable

confounding variable

variable that has directly and systematically affected the dependent variable apart from the IV

• extraneous variable that may not have been controlled for

• provide alternate explanation for the results

• can only be identified at the end of experiment

sampling size and procedures to prevent error and bias

large sample = large representativeness, findings are more likely to be unbiased

using more objective sampling procedures (random or stratified) —> more representative sample

experimental design choice to prevent error and bias

not choosing within subjects prevents order effects

counterbalancing preventing error and bias

• reduces order effects that involve ordering experimental conditions in a certain way

  • completing experimental conditions in different orders so results can still be compared