outcome 4 vocab Flashcards
population
the entire group of individuals about which we want info
sample
population from which we are actually collecting info
bad sampling methods
convenience sampling and voluntary response sampling
simple random sample
n individuals chosen in such a way that every individual has an equal chance to be selected
what to include when describing hat method
equal sized pieces of paper, shake the hat, without replacement
stratified random sampling
classify population in groups of similar individuals (strata). Then, choose separate SRS in each stratum and combine to form full sample.
EX: divide school pop into grades
pro of stratified random sampling
to get equal representation
how to choose variable for stratifying
choose a variable that reduces the variability the most, or is most strongly associated with the response variable
cluster sampling
Divide population into smaller groups (clusters should mimic characteristics of population), then choose SRS of clusters (all individuals included in sample)
cons of sampling
- under coverage bias: some groups in pop are left out of process of choosing sample
- nonresponse bias: an individual chosen can’t be contacted/ chooses to not respond
- response bias: systematic pattern of incorrect responses in a sample survey
- bad wording of questions
census
data from all individuals in population
systematic sampling
sampling every nth member of the population
pro of cluster sampling
easier
observational study
observes individuals + measures variables, doesn’t attempt to influence responses
experiment
deliberately imposes some treatment on individuals to measure their responses
confounding variable
when 2 variables are associated in such a way that their effects on a response variable cannot be distinguished from one another
treatment
a specific condition applied to the individuals in an experiment
experimental units
who’s receiving the treatment? smallest collection of individuals to which treatments are applied
keys to a good experimental design (4)
- comparison
- random assignment
- control
- replication
Completely randomized design
the experimental units are assigned to treatments completely by chance (hat method)
Randomized block design
the random assignment of experimental units to treatments is carried out separately within each block
block
group of experimental units that are known before to be similar in some way that will affect the response of the treatments
Matched pairs design
a randomized blocked experiment where each block consists of a matching pair of similar experimental units
double blind experiment
neither the subjects or those interacting with them and the person measuring the response variable know which treatments a subject recieved
when can we generalize an experiment to the whole population
if there is random assignment
when can we assume a cause and effect relationship from an experiment
if there is random selection
