Sample Design

and
Sampling Procedures

By Group 8

Gaurav Gupta (10P138)

Kandarp Suchak (10P144)
Mayank Bukrediwala (10P149)
Patel Jigishkumar ( 10P157)
Prerna Sharma (10P160)
Ramya M( 10P165)
Page 1
 Sampling
• Sampling: It is the process of
using subsets of a larger
population (samples) to make
conclusions about the entire
population
• Faster and cheaper than
asking the entire population
• Two keys :
– Selecting the right people
– Selecting the right number of
people

Page 2
 Sampling Terminology
• Population or universe – any complete group that share some
set of characteristics
• Population element – individual member of population
• Sampling frame – Listing of population from which a sample is
taken
• Census – investigation of all individual elements making up a
population
• Survey- A polling/investigation of the sample
• Target Population : Specific, complete group relevant to the
research project

Page 3
 Why Sample?
• Properly selected samples yield accurate and
reliable results.
– If elements are similar smaller sample is needed
• May even be more accurate than census
– Bureau of Census uses samples to check
accuracy of the U. S. Census
• Saves resources

Page 4
Stages in the Selection of a Sample
Define the target population

Select a sampling frame

Determine if a probability or nonprobability

sampling method will be chosen

Plan procedure
for selecting sampling units

Determine sample size

Select actual sampling units

Conduct fieldwork
Page 5
 Sampling Frame ( Working population)

• A list of elements from which the sample may be

drawn
• Provides the list that can be worked with operationally
• Mailing lists – List of the names, addresses and
phone numbers of specific populations
• Sampling frame error – occurs when population is not
accurately represented in the sampling frame.

Page 6
Random Sampling Error
• The difference between the sample results and the
result of a census conducted using identical
procedures
• Statistical fluctuation due to chance variations

Systematic Errors
 Nonsampling errors
 Unrepresentative sample results – (e.g.,
educated vs. uneducated respondents in
mail survey)
 Not due to chance
 Due to study design or imperfections in
execution Page 7
Errors Associated with Sampling
Sampling frame error
 Occurs when certain sample elements are not listed or
available and are not represented in the sampling frame
 Random sampling error
 Difference between sample resut and the result of a census
conducted using similar procedures
Nonresponse error
 Statistical difference between a survey that includes only
those who responded and a survey that also includes those
who failed to respond

Page 8
 Two Major Categories of Sampling

• Probability sampling
• Known, nonzero probability for every element
• Equal chance of being included in the sample
• Nonprobability sampling
• Probability of selecting any particular member is
unknown, i.e. unequal chance of being selected in the
sample
• Technically, inappropriate to apply statistical techniques
to project beyond the sample
• Cheaper and faster than probability

Page 9
Two Major Categories of Sampling

Page 10
 Convenience Sampling
• Also called haphazard or accidental sampling
• The sampling procedure of obtaining the people or
units that are most conveniently available
• Eg. People in the classroom, friends, relatives
• Advantage : Very low cost , No need for list of
population
• Disadvantage:
– Bias associated with estimates cannot be measured or
controlled
– Projecting data beyond sample is inappropriate

Page 11
 Judgment Sampling
 Also called purposive sampling
 An experienced individual selects the sample based on his or her
judgment about some appropriate characteristics required of the sample
member
 E.g. a research is conducted on toy stores, selecting three stores
purposively from where accurate results can be found.
 Advantage:
 Moderate cost and average use by the researcher
 Useful for certain types of forecasting
 Sample guaranteed to meet a specific objective
 Disadvantage:
 Bias due to researchers’ beliefs may make sample unrepresentative
Projecting data beyond sample is inappropriate

Page 12
 Quota Sampling
 Ensures that the various subgroups in a population are represented on
pertinent sample characteristics
 To the exact extent that the investigators desire
 Advantages:
 Moderate cost and extensively used by the researcher
 Introduces some stratification of population
 Requires no list of population

 Disadvantages
 Introduces bias in researcher’s classification of subjects
 Non-probability selection within classes/groups means
 Error from population cannot be estimated

Page 13
 Snowball Sampling
 A variety of procedures
 Initial respondents are selected by probability methods if possible
 Additional respondents are obtained from information provided by the
initial respondents
 Eg. Friends of friends
 Advantages
 Low cost and used in special situations
 It is useful in locating members of rare populations.
 Disadvantages
 High bias because sample unit is not independent
 Projecting data beyond sample is inappropriate

Page 14
 Simple Random Sampling
• A sampling procedure that ensures that each element in the
population will have an equal chance of being included in the
sample
• Advantages
– Only minimum advance knowledge of population is neede
– Easy to analyze data and compute error.
• Disadvantages
– Requires sampling frame
– High cost
– Not frequently used in practice

Page 15
 Systematic Sampling
• A simple process
• Every nth name from the list will be drawn
• Initial starting point is selected by a random process

• Sampling Interval : The number of population

elements between units selected for the sample
• Periodicity: A problem that occurs if original list has a
systematic pattern

Page 16
 Stratified Sampling
 Probability sample
 Random Subsamples are
drawn within different strata
 Each stratum is more or less
equal on some characteristic
 Ensures a more efficient
sample
 Sample accurately reflects
the population on the basis
of the criterion used for
stratification

Page 17
 Cluster Sampling
• The purpose of cluster sampling is to sample
economically while retaining the characteristics of a
probability sample.
• The primary sampling unit is no longer the individual
element in the population
• The primary sampling unit is a larger cluster of
elements located in proximity to one another
• Eg. Area sample
• Frequently used when no lists of the sample
population are available

Page 18
 What is the
Appropriate Sample Design?

• Representativeness is Always Important

• Degree of accuracy
• Resources
• Time
• Advanced knowledge of the population
• National versus local
• Need for statistical analysis

Page 19
Determination
of
Sample Size

Page 20
 What does statistics mean?
• Descriptive statistics
– Number of people
– Trends in employment
– Data
• Inferential statistics
– Make an inference about a population from a
sample

Page 21
Population Parameter
• Variables in a population
• Measured characteristics of a population
• Greek lower-case letters as notation

Sample Statistics
• Variables in a sample
• Measures computed from data
• English letters for notation

Page 22
 Making Data Usable
• Frequency distributions
• Proportions
• Central tendency
– Mean
– Median
– Mode
• Measures of dispersion

Page 23
 Measures of Central
Tendency
• Mean - arithmetic average

 Xi
 
N
• Median - midpoint of the distribution
• Mode - the value that occurs most often

Page 24
 Measures of Dispersion
or Spread
• Range
• Mean absolute deviation
• Variance
• Standard deviation

Page 25
 The Range
as a Measure of Spread
• The range is the distance between the smallest
and the largest value in the set.
• Range = largest value – smallest value

Average Deviation
(X i X)
0
n

Page 26
 Low Dispersion Verses High
Dispersion
5
Low Dispersion
4
Frequency

150 160 170 180 190 200 210

Value on Variable
Page 27
 Low Dispersion Verses High
Dispersion
5

4
High dispersion
Frequency

150 160 170 180 190 200 210

Value on Variable
Page 28
 Variance
• The variance is given in
 X  X ) 2
squared units
S 
2

n 1
• The standard deviation
is the square root of

  Xi X 
S 2
variance:

n 1

Page 29
 Normal Curve: IQ Example
 Normal curve
 Bell shaped
 Almost all of its values are
within plus or minus 3
standard deviation

13.59% 34.13% 34.13% 13.59%

2.14%
2.14%

70 85 100 115 145

Page 30
Standardized Normal Distribution
• Symmetrical about its mean
• Mean identifies highest point
• Infinite number of cases - a continuous distribution
• Area under curve has a probability density = 1.0
• Mean of zero, standard deviation of 1

Page 31
 Standard Normal Curve
• The curve is bell-shaped or symmetrical
• About 68% of the observations will fall within 1
standard deviation of the mean
• About 95% of the observations will fall within
approximately 2 (1.96) standard deviations of the
mean
• Almost all of the observations will fall within 3
standard deviations of the mean

Page 32
The Standardized Normal is the
Distribution of Z

–z +z
Page 33
 Standardized Values
• Used to compare an individual value to the
population mean in units of the standard
deviation

x
z

Page 34
 Linear Transformation of Any Normal
Variable Into a Standardized Normal
Variable

 

  X

Sometimes the Sometimes the

scale is stretched scale is shrunk

x
z
-2 -1 0 1 2


Page 35
 Central Limit Theorem

•As a sample size increases, the distribution of

sample means of size n, randomly selected,
approaches a normal distribution

Page 36
Distribution Mean Standard
Deviation
Population  
Sample S
X

Sampling X SX

Page 37
 Confidence Interval

  X  small sampling error E  Z cl S X

S
Sx 
n

S
  X  Z cl
n

Page 38
 Sample Size
• Variance (standard
deviation)
• Magnitude of error
• Confidence level

2
 zs 
n 
E

Page 39
Standard Error of the
Proportion
sp 
pq
n

or

p ( 1 p )
n
Page 40
Confidence Interval for a
Proportion

p  ZclSp

Page 41
Sample Size for a Proportion
Z pq 2

n
E 2

Where:
n = Number of items in samples

Z2 = The square of the confidence interval

in standard error units.

p = Estimated proportion of success

q = (1-p) or estimated the proportion of failures

E2 = The square of the maximum allowance for error

between the true proportion and sample proportion
or zsp squared.

Page 42
Page 43