Chapter 4 Flashcards
Sample survey steps
1) What population we want to describe
2) What population we want to measure
Bias
design that systematically favors certain outcomes
Voluntary response samples
People call in for their opinion
Random sampling
the use of chance to select a sample
Simple random sampling
Set of n individuals chosen in such a way that every n has an equal chance
Stratified random sample
Group by similarity, SRS strata, combine SRSs
Cluster
Chose SRS of the cluster. All clusters are included in sample.
Inference
we predict or assumer things about a populatio
Undercoverage
when some group are left out of the process of choosing the sample
Wording bias
When a question is skewed for a certain answer
Observational study
Observes individuals, and infers
Experiments
seek to establish cause and effect
Lurking variable
Unaccounted for variable that effects response
Confounding
When two variables become indistinguishable
In experiments, explanatory variables are called
factors
Random assignment
Experimental units are assigned to treatments at random
Completely randomized design
Treatments are assigned to all the experimental units by chance
3 principles of experimental design
1) Control 2) Random Assignment 3) Replication
replication is
using enough experimental units
Double blind
Neither subjects nor the doctors know which treatment is being used
statistically significant
effect is too large to happen by chance
Block
a group of experimental units known before the experiment to be similar
Randomized block design
Random assignment is used within each block
Matched pairs
comparing two similar experimental units with two treatments
lack of realism
experiments that are difficult to test
confidentiality
information is kept secret and private