Transcript Types of Samples

CHAPTER 4:
Data Collection
and Sampling Methods
to accompany
Introduction to Business Statistics
fourth edition, by Ronald M. Weiers
Presentation by Priscilla Chaffe-Stengel
Donald N. Stengel
© 2002 The Wadsworth Group
Chapter 4 - Learning Objectives
• Describe the types of studies and their purposes:
–
–
–
–
Exploratory
Descriptive
Causal
Predictive
• Explain the differences between data sources:
– Primary
– Secondary: Internal versus External
• Differentiate between sampling designs:
– Probabilitistic versus nonprobabilitistic designs
• Differentiate between sampling and nonsampling
errors.
© 2002 The Wadsworth Group
Chapter 4 - Key Terms
• Error
– Sampling
– Nonsampling
• Types of Studies
–
–
–
–
Exploratory
Descriptive
Causal
Predictive
• Types of Data
– Primary
– Secondary
» Internal vs External
• Probability sample
–
–
–
–
Simple random sample
Systematic sample
Stratified sample
Cluster sample
• Nonprobability sample
–
–
–
–
Convenience sample
Quota sample
Purposive sample
Judgment sample
© 2002 The Wadsworth Group
Types of Studies
• Exploratory
– Understand a problem, identify relevant
variables, formulate hypotheses
• Descriptive
– Establish reliable measurements
• Causal
– Determine relationships among variables
• Predictive
– Use analysis to forecast
© 2002 The Wadsworth Group
Sources of Data
• Primary
– Data generated by the researcher for this study
– Survey, experimental, observational research most
popular
– Tend to require more time and expense than
secondary data
• Secondary
– Data gathered from another source or for another
purpose
» Internal sources within the researcher’s
organization
» External sources, including governmental, trade,
commercial and internet sources
© 2002 The Wadsworth Group
Types and Sources of Error
• Sampling Error
– Random, nondirectional
– When a sample is used instead of a census
• Nonsampling Error
– Directional bias overstating or understating
the true population parameter
– Potential sources:
» Poor sample design
» Poor measurement
» Poor instrumentation
© 2002 The Wadsworth Group
Types of Samples
Probability, or Scientific, Samples: Each element to be
sampled has a known (or calculable) chance of being selected.
• Simple random • Every person has an equal
chance of being selected.
Best when roster of the
population exists.
• Randomly enter a stream of
• Systematic
elements and sample every
kth element. Best when
elements are randomly
ordered, no cyclic variation.
© 2002 The Wadsworth Group
Types of Samples
Probability, or Scientific, Samples: Each element to be
sampled has a known (or calculable) chance of being selected.
• Stratified
• Cluster
• Randomly sample elements
from every layer, or
stratum, of the population.
Best when elements within
strata are homogeneous.
• Randomly sample elements
within some of the strata.
Best when elements within
strata are heterogeneous.
© 2002 The Wadsworth Group
Types of Samples
Nonprobability Samples: Not every element has a chance to
be sampled. Selection process usually involves subjectivity.
• Convenience
• Quota
• Elements are sampled
because of ease and
availability.
• Elements are sampled, but
not randomly, from every
layer, or stratum, of the
population.
© 2002 The Wadsworth Group
Types of Samples
Nonprobability Samples: Not every element has a chance to
be sampled. Selection process usually involves subjectivity.
• Purposive
• Judgment
• Elements are sampled
because they are atypical,
not representative of the
population.
• Elements are sampled
because the researcher
believes the members are
representative of the
population.
© 2002 The Wadsworth Group
Example: Types of Data
Problem 4.46
• The Sonic Travel Agency has not kept
pace with the current boom in family
vacation air travel between the agency’s
northern city and sunny Florida. For
each of the following, indicate whether
they are primary or secondary. If
secondary, further indicate whether
they are internal or external.
© 2002 The Wadsworth Group
Problem 4.46
• a) Sonic’s research • a) Primary data
department conducts – The data were generated
by the researcher to study
a survey of past
this problem.
customers to
determine their level
of satisfaction with
the agency’s service
© 2002 The Wadsworth Group
Problem 4.46, continued
• b) The Florida
• b) Secondary data
Bureau of Tourism
– The data were not generated
sends travel
for the purpose of examining
agencies across the
the question at hand.
nation the results of
External data
a study describing
the characteristics of – The data were provided by an
outside agency, the Florida
winter vacationers
Bureau of Tourism.
to the state,
including where and
how long people
intended to stay.
© 2002 The Wadsworth Group
Problem 4.46, continued
• c) To get a better
• c) Secondary data
grasp of the
– The data were not generated
for the purpose of examining
problem, the
the problem at hand.
president of Sonic
Internal data
examines the
company’s annual
– The data were produced
from within the firm, so they
reports for the
are internal data.
past ten years.
© 2002 The Wadsworth Group
Example: Types of Studies
Problem 4.48
• For each of the report titles below,
indicate whether the study involved
was exploratory, descriptive, causal, or
predictive.
© 2002 The Wadsworth Group
Problem 4.48
• a) “The Popularity of • a) Descriptive study
Bowling as a
– This is probably a
descriptive study
Participant Sport in
showing the number of
Ohio.”
people who bowl in Ohio.
• b) “The Effects of TV • b) Causal study
– This is probably a causal
Violence on Crime in
study that examines the
Small Towns.”
effect of TV violence (the
cause) on crimes in small
towns (the outcomes).
© 2002 The Wadsworth Group
Problem 4.48, continued
• c) “A Preliminary • c) Exploratory study
Investigation of
– This is probably an
exploratory study that points
the Market for Pet
out relationships and
Foods.”
possible hypotheses for
further study.
• d) Predictive study
• d) “Family
– This is probably a predictive
Vacationing in the
study to forecast vacationing
21st Century.”
patterns in the 21st century.
© 2002 The Wadsworth Group
Example: Types of Errors
Problem 4.64
• Researchers at a university with an
enrollment of 12,000 take a census of the
entire student population, asking:
“Shoplifting is not only illegal, it raises
the prices that we all have to pay. Have
you ever shoplifted from the university
bookstore?” Based on this, discuss the
possible presence of a) sampling and
b) nonsampling error.
© 2002 The Wadsworth Group
Problem 4.64
• a) Sampling error not • a) Since the study
includes the entire
present
student population,
there will be no
sampling error present.
• b) Nonsampling
error present
• b) Because the question’s
wording invites biased
responses, students may
be afraid to admit to
shoplifting, so there will
be nonsampling errors
present.
© 2002 The Wadsworth Group