randomization
Terms
undefined, object
copy deck
- confounding
- when many variables confuse the ability to determine the cause of a characteristic.
- census
- an attempt to include the entire population
- bias
- when a study favors an outcome
- randomization
- when chance is used to give subjects treatments to aviod bias
- srs
- when every unit has an equal chance to be examined
- stratified random sample
- when groups are made and then srs is used
- systematic random sample
- numbering off, choosing every other person
- response variable
- measures an outcome of a study
- explanetory variable
- explains the outcome
- lurking variable
- has an important effect on the outcome but isn't used in the study
- experimental units
- thing on which an experiement is performed
- subjects
- human experimental units
- anecdotal evidence
- based on haphazardly selected individual cases that are striking in some way
- treatment
- specific experiemental process applied to each case
- control group
- the group without treatment and no sham treatment
- placebo
- a dummy or sham treatment
- block design
- a bock is a group know to be similar before the experiment in some way that is expected to effect the response to the teatement. (strat random sample)
- princeples of experimental design
- control, randomization, replication
- double blind experiement
- when the subject nor the evaluator know which treatment the subject is recieving
- statistical significance
- a difference too large to be attributed to chance
- cautions about experimentation
- hidden bias, lack of realism
- voluntary response samples
- only those who choose to respond are evaluated
- cautions about sample surveys
- undercoverage, nonresponse, response bias, wording of question
- descriptive statistics
- methods of organizing, displaying, and describing data using tables, graphs etc.
- inferential statistics
- uses collected data to make generalizations about the entire group