stats module 0 Flashcards
what is a census
special sample that includes everyone and “samples” the population
issues with census
expensive, undercoverage (might not include everyone), time consuming
what are sample statistics
data that is collected from a sample and the summaries
two types of inferences
population inference, causal inference (cause and effect)
population inference
results from the sample that can be generalized to an entire population
causal inference
difference in responses that is caused by the difference in treatments when comparing the results from two treatment groups
when should you only make population inferences
when there is random sampling
what is randomizing
helps to eliminate the effect of unknown extraneous factors, on average, sample looks like the rest of the population, individuals are selected randomly from population
what is non-random sampling
leads to biased results, no way to fix biased sample or salvage useful information, best way to avoid is to sample at random
what are the different ways to randomly sample
simple random samples (SRS), stratified random sampling, systematic random sampling, cluster random sampling
what is simple random sampling (SRS)
each sample size of n in population has same chance of being selected, samples are drawn at random, generally differ from one another (sampling variability)
what is stratified random sampling
population is divided into different homogenous groups (strata), then taken an SRS within each stratum before results are combined, can reduce random sampling and variability)
what is systematic random samplign
start from a random selected individual, then sample every kth person, can give representative sample, less expensive than random sampling
what is cluster random sampling
splitting the population into similar groups (or clusters), selecting one or a few at random and perform census on them, gives unbiased sample
what is bias?
then tendency for a sample to differ from the corresponding population in some systematic way