Sampling and Sampling

Distribution
OBJECTIVES:
At the end of the lecture, students will be able to:
❑ Describe sample design and sampling frame,
❑ Bias,
❑ Sampling and non-sampling errors.
❑ Sampling with and without replacement,
❑ probability and non-probability sampling,
❑ Sampling distributions for single mean and proportion,
 POPULATION 2
► A population is the total group of interest.

SAMPLE
► A sample is a subgroup of the group of interest

TARGET
POPULATION
The group to
whom researchers
hope to generalize
their findings.
3
 What is the
relevant 4
population?

How much What type

will the of sample
sample shall we
cost? draw?

SAMP
LE
DESIG
N
What
What
sampling
sample size
frame shall
is needed? [
we use?]

What are
the
parameters
of interest?
SAMPLING FRAME 5

► The sampling frame is the list from which units are drawn for the sample.
The 'list' may be an actual listing of units, as in a phone book from which
phone numbers will be sampled, or some other description of the
population, such as a map from which areas will be sampled.

BIAS
A term used to describe statistics that don't provide an accurate
representation of the population

Any sampling procedure that creates inference that overestimate or under estimate
the characteristics is said to biased
 ERROR 6

► The Difference between the estimated value and population true

value is called error

► The errors are of two types :

1. Sampling error
2. Non-sampling error
 SAMPLING ERROR 7

“Refers to the fact that the chance will nearly always ensure that one sample will differ
from another, even if the two samples are drawn from exactly the same target population
in exactly the same random way.”

► Sampling error is the degree to which a single estimate might differ from population.

► The Sampling can be reduced by increasing the sample size, by improving the sample
design .

Sampling error is inherent and unavoidable in every sampling scheme

 NON-SAMPLING ERROR 8

► The error that are caused by sampling the wrong population of interest and by
response bias, as well as those made by an investigator in collecting and
reporting the data.
Sampling With and Without 9
Replacement
► Sampling without replacement is where items are chosen
randomly, and once an observation is chosen it cannot be chosen
again.
► On the other hand, when you sample with replacement, you
also choose randomly but an item can be chosen more than once
🞂​With small population, researcher usually includes all the
individuals in the population: Census study

when the population is too large then a small, but carefully

🞂​But,

chosen sample can be used to represent the population.

◦ The sample reflects the characteristics of the population from which it is
drawn.

◦ ​ .
The process of selecting representative subset from
population is called Sampling in Statistics
SAMPLING METHODS 11

Sampling
techniques

Non
Random Random
sampling Sampling
PROBABLITIY METHODS 12

“Generation of probability samples involves Randomization at some point in the

process”

NO RANDOMIZATION in Non-probability samples.

WHY PROBABLITY 13

METHODS???
I. Generalization
II. Less sampling error than Non-probability samples.
14
 15
SIMPLE RANDOM
SAMPLING(SRS)

“Simple random sampling is a procedure in which each

member of the population has an equal chance of being
selected for the sample, and selection of each subject is
independent of selection of other participants” [1,2]
METHODS FOR SIMPLE RANDOM SAMPLING 16

► In MS Excel you can

► Type a Command in any of the cell

► =N*rand()

N= total no of sample
/Population.
 17
► LOTTERY METHOD/HAT METHOD.
 18
RANDOM NUMBER TABLE:
• The preferred method for generating a simple random
sample is to use random numbers that are provided in
a table or generated by a computer.

• Moving in either a horizontal or vertical direction,

 SYSTEMATIC RANDOM SAMPLING 19

“Systematic sampling is a process by which the researcher selects every nth

person on a list.”

FORMULA:

K= N/n
STEPS 20

► Number the units in the population from 1 to N decide on the n

(sample size)that you want or need.

► k = N/n.

► The interval size randomly select an integer between 1 to k then

take every kth unit
21
STRATIFED RANDOM SAMPLING 22

“When certain subgroups must be represented in adequate numbers within the

sample or when it is important to preserve the proportions of subgroups in the
population within the sample.”

🞂​Procedure: Divide the population into non- overlapping groups (i.e., strata) N1, N2,

N3, ... Ni, such that N1 + N2 + N3 + ... + Ni = N. Then do a simple random sample of f =

n/N in each strata.

STEPS 23

► Divide the population into non-overlapping groups (i.e., strata)

N1, N2, N3, ...Ni, such that N1 + N2 + N3 + ... + Ni = N.

► Then, select from each stratum through simple random sampling

CLUSTER SAMPLING 24

► Cluster sampling is the use of naturally occurring groups as the

sampling units. It is used when an appropriate sampling frame
does not exist or when logistical constraints limit the researcher’s
ability to travel widely.

► Procedure:
► Divide population into clusters (usually along geographic
boundaries) Randomly sample clusters. Measure all units within
sampled clusters
25
STEPS 26

► Each step of the cluster sampling procedure can be implemented

through simple random, systematic, or stratified sampling.

► Select ‘mi’ cluster from “M” sets of cluster in entire population then
complete ‘n’ sample from those

► Selected (Heterogeneous) clusters.

 27
NON-PROBABILITY SAMPLING

“Absence of randomization”
 WHY???? 28

► Limited Funding.

Because many studies are self funded, subject selection is confined

to a single setting with a limited number of available patients, so
the researcher often chooses to study the entire accessible
population.
NON-PROBABILITY METHODS 29

► Four types: Convenience, snowball , Quota and Purposive sampling.

CONVENIENT SAMPLING 30

“Samples of convenience involve the use of readily available participants”

SNOWBALL SAMPLING 31

“A snowball sample may be used when the potential members of the

sample are difficult to identify”

► Selection of participants driven from preceding respondent (Rare

Disaeses)
32
 PURPOSIVE SAMPLING 33

Researcher has a specific reason for selecting particular participants for

study.

► Purposive sampling uses handpicked units that meet the researcher’s

needs.
34
​ Random Sampling Methods:
1. Simple Random Sampling
◦ Select ‘n’ sample from a frame “N” population
◦ By random number dialing, lottery method
2. Stratified Random Sampling
◦ Select ‘ni’ sample from a each “Ni” homogenous populations
◦ Advantage: Estimate from each population could be obtained
3. Cluster Random Sampling
◦ Select ‘mi’ cluster from “M” sets of cluster in entire population then
complete ‘n’ sample from those selected (heterogeneous) clusters
◦ Advantage: Reduced cost and time
Random Sampling Methods:
4. Systematic Sampling
◦ Select ‘kth’ observations until to reach ‘n’ sample size
◦ The first selection is random then choose each k
= N/n sample observation
5. Multistage (Random) Sampling
◦ Combination of multiple sampling techniques

3
6
Non-Random Sampling Methods:
1. Convenient Sampling
◦ Selection of participants with easiest approach
2. Snowball sampling
◦ Selection of participants driven from preceding respondent
3. Quota sampling
◦ Selection of participants from known aspects of population
4. Purposive sampling
◦ Selection of participants from unique aspect

24
 SAMPLING DISTRIBUTION 38

► We discussed several different measures of central tendency and

variability distinguished between numerical descriptive measures
of a population (parameters) and numerical descriptive measures
of a sample (statistics).

► The numerical value of a sample statistic cannot be predicted

exactly in advance.
► Even if we knew that a population mean was $216.37 and that
the population standard deviation was $32.90—even if we knew
the complete population distribution—we could not say that the
sample mean y would be exactly equal to $216.37.
 39

► A sample statistic is a random variable; it is subject to random

variation because it is based on a random sample of
measurements selected from the population of interest.
► Also, like any other random variable, a sample statistic has
probability distribution.
► We call the probability distribution of a sample statistic the
sampling distribution of that statistic.
► Stated differently, the sampling distribution of a statistic is the
population of all possible values for that statistic.
REFERNCES: 40

► R. L Ott, Micheal T longnecker. An introduction to statistical

methods and data analysis, 7th ed. Brooks/Cole, Cengage
Learning 2015
► Walpole RE. Students study guide: introduction to statistics. 3 rd ed.
1982.

► Images by Google.com