AP Stats: Unit 3

population

the entire group of people, animals, or things that you wish to know something about

sample

subset of the population that you are actually taking measurements of

census

survey that collects Data from every member of the population

observational study

observes and measures individuals without attempting to influence them

retrospective observational study

examines existing data

prospective observational study

tracks individuals into the future

experiment

imposes treatments on experimental units to measure responses

random sample

when all the members of a population are equally likely to be chosen

simple random sample (SRS)

when every group of size n is equally likely to be chosen; no stratifying or clustering

without replacement

individuals can only be selected once

with replacement

individuals can be selected more than once

stratified sample

dividing the population into homogenous groups and selecting some from every group; best to do when you notice that the groups differ regarding the relevant issue, and will result in less variability in responses if done properly

cluster sampling

dividing the population into heterogeneous groups and selecting one or a few groups to represent the whole

systematic random sampling

randomly selecting one of the first k individuals and choosing every kth individual thereafter 

bias

when certain outcomes are systematically favored

convenience sampling

selecting those members of the population that are easiest to sample 

voluntary response sampling

allows the members of the sample to select themselves to be in the sample; typically results in emotional and negative answers

undercoverage

when some members of the population cannot be in the sample 

nonresponse

when a member that is chosen for the sample does not respond 

response bias 

when there is something about the situation that is causing members of the sample to respond incorrectly

poor question wording

intentional or not, this can lead to biased results

confounded variables

two variables whose affects on a response cannot be distinguished one from the other 

experiment

study that imposes a treatment

treatment

specific condition applied to individuals in an experiment

experimental units/subjects

smallest collection of individuals to which a treatment is applied

factor

an explanatory variable in an experiment

levels

amount of the factors

control

having a group that receives no treatment for a basis of comparison

double blind study

neither the subjects nor the researchers know who is receiving which treatment

single blind study

either just the researchers or just the subjects know who is receiving which treatment

replication

using enough subjects to draw a meaningful conclusion

blocks

homogeneous groups created to account for another variable we think may influence responses *experiment only

completely randomized 

no blocking

statistically significant

result is beyond what we would expect due to random chance