1-Introduction Flashcards
What is a summary statistic?
a single number summarizing a large amount of date, e.g., a percentage
What are the two types of variables?
numerical/quantitative (continuous and discrete)
categorical/qualitative (nominal and ordinal)
What are associated variables? What are positive and negative associations?
variables that show a connection with each other (dependent variables)
association DOES NOT imply causation
positive association = upward trend in both variables
negative association = upward x, downward y
What are independent variables?
variables that do not show a connection with each other
What are explanatory and response variables?
the explanatory variable causes a change in the response variable
What is an observational study? Can it be used to establish casualty?
collect data in a way that does not interfere with how the data arises
NOT used to investigate the possibility of a causal connection
What is an experimental study?
used to investigate the possibility of a causal connection
What is the difference between a target population and a sample?
every member of a group vs a representative fraction of the group
What is a confounding variable?
a variable that influences both the explanatory and response variables
often unknown
What is a simple random sample? Is it generalizable?
o raffle
o each case in the population has an equal chance of being included
o likely representative / generalizable
Name 3 common sampling biases?
CNV
o convenience bias: easily accessible cases are more likely to be included in the sample
o non-response bias: only a non-random fraction of a random sample respond, e.g., starts with a random sample, but not everyone responds because not everyone sampled is reached
o voluntary response bias: only people who volunteer actually respond
What’s the difference between a prospective and a retrospective study?
ongoing (as events unfold) vs past
What is a stratified sample?
see stratifiedsample
divide the population into groups (strata); group similar cases in each strata
use a second sample method, e.g., SRS, select SOME cases from within EVERY stratum for the sample
What is a cluster sample?
see clustersample
divide the population into groups (clusters)
ALL cases from SOME clusters
pick a few clusters and include ALL cases from EACH selected cluster in the sample
What is a multi-stage sample?
o divide the population into groups (clusters)
o pick a few clusters and include SOME cases (SRS) from EACH selected cluster