Section 3 Vocabulary

Question 1

bias

Answer 1

any systematic failure of a sampling method to represent its population (over/underestimating parameters)
1. too reliant on voluntary response
2. undercoverage of pop.
3. nonresponse bias
4. response bias

Question 2

nonresponse bias

Answer 2

when large fraction of those sampled fails to respond
1. voluntary response bias

Question 3

voluntary response bias

Answer 3

individuals can decide whether or not to participate; sample is typically constituted of strongly opinionated ppl.

Question 4

response bias

Answer 4

anything that influences responses
1. wording of the questions asked

Question 5

undercoverage

Answer 5

gives part of population less representation than this group really has in the population

Question 6

population parameter

Answer 6

any number computed for the entire population
ex: # of adults who think that drinking + driving is a problem
(# of people who actually have belief that experiment is trying to prove)

Question 7

population

Answer 7

entire group of individuals
ex: all adults 21+ in given city

Question 8

census

Answer 8

sample that consists of entire population

Question 9

sample

Answer 9

representative subset of a population, examined in the hope of learning about entire population
ex: adults interviewed at the particular bar

Question 10

sample size

Answer 10

of individuals in a sample; testament to how well sample represents population

Question 11

sample surveys

Answer 11

asks questions of a sample drawn from a population in hops of learning something about the entire population
(issues w/response + nonresponse bias)

Question 12

sampling frame

Answer 12

list of individuals from whom sample is drawn; if someone is not in the sampling frame, they are not in the sample
ex: all individuals interviewed at the bar

Question 13

sampling variability

Answer 13

natural tendency of randomly drawn samples to differ from one another

Question 14

cluster sampling

Answer 14

entire groups are chosen at random; random clusters are more representative of entire population
ex: using a list of all classes, randomly choosing a few classes (clusters) then surveying all students in each class

Question 15

convenience sampling

Answer 15

sample consists of individuals who are conveniently available
ex: student organization wanting to get signatures for a petition by camping out in front of a lecture hall

Question 16

multistage sampling

Answer 16

combines several different sampling methods
ex: randomly selecting some hospital facilities from each of 5 geographic regions and then taking samples from each of these hospitals’ discharge lists

Question 17

simple random sample

Answer 17

each set of x elements in population has an equal chance of being selected
ex: select 3 students to form a class to receive ice cream by putting all students’ names into a hat and picking out 3 names randomly

Question 18

stratified random sampling

Answer 18

population is divided into strata (several subpopulations) and then random samples are drawn from each stratum; usually more consistent than simple random samples
ex: dividing a university into departments and then randomly selecting a few students from classes in each department

Question 19

systematic sampling

Answer 19

selecting individuals systematically from a sampling frame
ex: instruct each hospital facility to survey every 500th patient discharged

Question 20

strata

Answer 20

several subpopulations
ex: university=population; departments=subpopulations