20% for 80%

• lab experiments are highly controlled, establishes cause and effect relationship, lacks ecological validity
• field experiments are real world setting, higher ecological validity, less control
• natural/quasi uses naturally occurring variables, no control over IV
• indepedant groups have different participants in each condition, no order effects, participant variables
• repeated measures same participants in all conditions, controls participant variables, order effects
• matched pairs are pairs matched on key variables, reduces individual differences, time consuming
• random sampling have an equal chance of selection, unbiased, difficult to achieve
• opportunity sampling are first available participants, easy and biased
• volunteer sampling is self selected, ethical and unrepresentative
• stratified sampling is proportional, representative and time consuming
• quantitive data is numerical, easy to analyse but lacks depth
• qualitative data is descriptive and rich in detail but hard to analyse
• mean uses all data but is affected by extremes
• median is not affected by outliers but ignores some values
• mode is useful for categories but least precise
• range is a simple measure of spread but is affected by extremes
• standard deviation is more precise measure of spread but harder to calculate
• reliability is consistency of results
• internal reliability is consistency within the study eg test refers or split half
• external reliability is consistency over time
• validity is the accuracy of measurement
• internal validity is when is measures what it claims to
• external validity can be generalised to real life (ecological, population, historical validity)
• nominal data uses the chi squared test
• ordinal data uses the Mann-whitney (independent) and the Wilcoxon (repeated)
• interval data uses t-test (parametric)
• p is less than or equal to 0.05 (results unlikely due to chance)