Professional Documents
Culture Documents
BA035IU
• Lecturer: Dr. Alang Thớ
• Email: alangtho@hcmiu.edu.vn
• Workshops: Monday, 8.00– 10.30
• Consultation appointment: Skype
Summarizing main points
from previous weeks
Sampling fundamentals
This week 8
Sampling size and statistical
theory
Summary of this week’s
lesson
• Distinguishing among a population, a census,
and a sample
• Understanding sampling process
• Describing probability and non-probability
sampling procedure
• Determining sample size with ed hoc methods
• Understanding confidence intervals and
Learning interval estimate
objectives • Calculating sample size for a simple random
sample
• Calculating sample sizes for stratified sample
and for multistage designs
Summarizing main points from previous
weeks
Sampling fundamentals
Differentiating a
Probability sampling
population, a Sampling process
techniques
sample, and a census
Determining sample
Non-probability
size with ad hoc
sampling techniques
methods
Differentiating population, sample, and
census
1 2 3
Population is the set of Sample is a subset, or Census is an
all objects that possess some part, of a larger investigation of all the
some common set of population (Zikmund individual elements that
characteristics (Aaker et and Babin, 2010, p. 301) make up a population
al., 2002, p. 363) (Zikmund and Babin,
2010, p. 301)
When is census appropriate?
Information is needed
Population size itself is
from every individual in
quite small
the population
Cost of making an
Sampling errors are high
incorrect decision is high
When is sample appropriate?
Easier to manage
Increasing response surveys and control in Population being dealt
quality with is homogeneous
the interview process
Used if census is
impossible
Error in sampling
Total Error • Difference between the true value and the observed
value of a variable
Execute sampling
Specify clearly
Should be
what is Convenience
reproducible
excluded
Determining sampling frame
Sample selection
Target population
methods
Addressing nonresponse
Sample size
problem
Probability sampling techniques (cont’d)
Stratified Sampling:
• The chosen sample is forced to contain units from each of the segments or strata of the population
• Proportionate Stratified Sampling:
• Number of objects/sampling units chosen from each group is proportional to number in
population;
• Can be classified as directly proportional or indirectly proportional stratified sampling
• Disproportionate Stratified Sampling:
• Sample size in each group is not proportional to the respective group sizes;
• Used when multiple groups are compared and respective group sizes are small
Probability sampling techniques (cont’d)
Cluster Sampling
• Involves dividing population into subgroups
• Random sample of subgroups/clusters is selected and all members of subgroups are
interviewed
• Very cost effective
• Useful when subgroups can be identified that are representative of entire population
Systematic Sampling
• Involves systematically spreading the sample through the list of population members
• Commonly used in telephone surveys
Cluster sampling and systematic sampling
Selecting sampling frames: Non-probability
sampling techniques
Costs and trouble of developing sampling frame are eliminated
Judgmental Snowball
Convenience Quota
• Used to obtain information • Minimum number from each
quickly and inexpensively specified subgroup in the
population
• Often based on demographic
data
An example of quota sampling
Non-response problems
Refuse to respond
Be inaccessible
Those who respond may differ from non respondents in a meaningful way, creating biases
1 2 3
Improve research Repeat the contact Attempt to estimate
design to reduce the one or more times the nonresponse
number of (call back) to try to bias
nonresponses reduce
nonresponses
Determining sample size
Use of statistical
techniques or ad hoc
methods Used when a person
knows from
experience what
sample size to adopt
Ad Hoc Methods
Used when budgetary
constraints dictate the
size of the sample
Ad hoc methods: Rules of thumb
Sample
Sample Reliability
Characteristics/Statistics
• Population mean is not • X will vary from sample to
known and must be sample
estimated from a sample • As sample size (n)
• Sample mean (X) is used increases, variation in X
to estimate the unknown will decrease
population mean
Sample characteristics/statistics (cont’d)
• Variation of X
• Standard error depends on sample size
• Assume that variation of X follows normal distribution
• Sampling distribution
• Indicates probability of getting a particular sample mean
Interval estimation
• X varies from sample to sample
• The difference between the sample mean (X) and the population
mean is the sampling error
• X + sampling error = interval estimate of sample mean
• Interval size depends on the confidence level the researcher wants for
the interval to contain the true population mean
• If the population standard deviation is not known, it is necessary to
estimate it with the sample standard deviation
Interval estimate of the population mean
• X + sampling error
sx - population standard deviation
Or
x + z sx / Ön
N - sample size
Size of interval estimate
• Two methods to get information from population. Census is contacting every member in the
population, while sampling is estimating a population parameter by contacting only a subset of
the population.
• Sampling process includes defining the target population; determining the sampling frames;
selecting the sample size; etc. For sample selection methods, it consists of probability sampling
methods and non probability sampling methods.
• Sample size is determined by ad hoc methods and statistical techniques. With statistical
techniques, sample size is determined by the confidence level desired, the allowed estimate
error, and the variance of the population.