experimental designs 1 Flashcards
(27 cards)
what is validity, in simple terms
trueness
what is reliability, in simple terms
consistency and replicability
what do you look at for the reliability of research procedures
operational definitions
research protocols
what are the 3 categories of reliability of results
test reliability
experimental reliability
statistical reliability
what is an operational definition
specifies conditions resulting in prod/measurement of outcomes and variable values
defines concepts in terms of how measured
must be consistent and informative
what do research protocols include
operational definitions
instruction/ handling
what is included in intructions
what to tell participants for consistency
methodological information
what is test reliability and what does it mean
whether there are consistent conditions
large diff btwn measurements means poor rel and variation under same condition tested also means poor rel
what is experiment reliability
whether repeated experiment will have same outcome
what is statistical reliability
likelihood of obtaining results if null hypothesis is true
unlikely then result is statistically reliable; <0.05
how do you improve power of statistical reliability
more participants to increase power, as more likely to find difference if difference to find
can also increase trial number
what does an increase in power mean
decrease in type 2 errors
what 2 things MUST you do to experimental design and why
balance them out
minimise carryover effects so that you only change what ppts experience in one way; also want to avoid results being confounded
what are the 3 types of experimental design
Independent samples
Repeated measures
Matched pairs
what is independent sampling also known as
between samples or between subjects sampling
what is independent sample design
each ppt participates in one condition of experiment; performance independent of other conditions
give a strength of Independent samples designs
inexperienced ppts so cant guess expierment purpose
avoid order effects and no carryover effects in this way
give a weakness to independent samples designs
variance in ppts as may accidentally end up with similar ppts in one category.
how can this weakness to independent samples be overcome
random assignment and large sample to balance out
what are repeated measure designs
same ppts in all conditions so do everything
IV manipulated within ppts
what is repeated measures also known as
within ppts/subjects design
what are 2 strengths of repeated measures and why
no ppt variable so diff can’t be due to differing ppts
fewer ppts needed
both these things give more statistical power
what are weaknesses in repeated measures
order and carryover effects mean could learn about goal of study
differences in stimuli
how can these weaknesses of repeated measures be overcome
counterbalancing ppts so everyone equally practised/fatigued
randomisation/counterbalance stimuli
