Chapter 3 -- Designing the Sample

1			The term ________ refers to the degree of bias that exists in the data.
			A)	accuracy
			B)	reliability
			C)	validity
			D)	error
2			The reliability of a measurement is reduced by:
			A)	error.
			B)	bias.
			C)	invalidity.
			D)	extraneous effects.
3			The greater the population variance, the ________ the reliability of the measurement for a given sample size.
			A)	greater
			B)	lesser
4			The greater the sample size, the ________ the reliability of the measurement.
			A)	greater
			B)	lesser
5			The greater the reliability of the measurement, the ________ the confidence intervals will be.
			A)	narrower
			B)	wider
6			The 95% confidence interval is within plus or minus ________ standard errors.
			A)	one
			B)	two
			C)	three
7			If the population is large, the selection of the sample size is ________ the population size.
			A)	determined by
			B)	highly dependent on
			C)	slightly dependent on
			D)	somewhat independent of
			E)	entirely independent of
8			If the population to be sampled is very homogeneous, with little variation from one sample unit to the next, this calls for a relatively ________ sample.
			A)	small
			B)	large
9			Continuing to collect data and submit them to analysis until the desired level of confident is reached is called:
			A)	systematic sampling.
			B)	progressive sampling.
			C)	preliminary sampling.
			D)	perpetual sampling.
10			Which is NOT an advantage of random sampling?
			A)	It is relatively inexpensive.
			B)	It best represents the population.
			C)	It has desirable statistical properties.
			D)	It permits computation of confidence intervals.
11			Dividing the population into two or more segments and sampling different proportions of each is called ________ sampling.
			A)	cluster
			B)	convenience
			C)	stratified
			D)	differential
			E)	logistical
12			A clustered sampling design can be used:
			A)	only over space or geographic area.
			B)	only over time periods.
			C)	over either space or over time.
			D)	over neither space nor time.
13			With what kind of data collection would cluster sampling most likely be used?
			A)	Mail data collection
			B)	Online data collection
			C)	Telephone interviewing
			D)	Personal interviewing
14			The need to qualify potential respondents is the major factor that distinguishes a ________ from other types.
			A)	clustered
			B)	random
			C)	quota
			D)	convenience
			E)	progressive
15			Quota sampling can be based on:
			A)	observation or questioning.
			B)	a sample frame and sampling units.
			C)	group membership or participation.
			D)	interviewer discretion or judgment.
16			The degree to which those within one area are likely to be similar to one another or to interact is an important consideration when designing a ________ sample.
			A)	random
			B)	stratified
			C)	clustered
			D)	convenience
			E)	quota
17			Nth name sampling is a method for selecting a ________ sample.
			A)	convenience
			B)	quota
			C)	clustered
			D)	stratified
			E)	random
18			With nth name sampling, the value of n is determined by:
			A)	the sample size and population variance.
			B)	population size divided by sample size.
			C)	the number and size of sampling clusters.
			D)	the number and size of population strata.
19			Nonresponse bias is likely to be most severe for:
			A)	self-administered surveys.
			B)	telephone interview surveys.
			C)	personal interview surveys.
20			The most likely kind of bias that would result from a sampling design that invites all visitors to a Website to complete an online survey ________ bias.
			A)	visibility
			B)	termination
			C)	order
			D)	self-selection
			E)	nonresponse