Sampling class phd aku

DR RAJESH G KONNUR
PROFESSOR
Sampling

Terms
 Research : “Scientific research is systematic ,
controlled , empirical & critical investigation of
natural phenomena guided by theory & hypotheses
about the presumed relations among such phenomena.
(Kerlinger 1986)
 Sample : A sample is a smaller (but hopefully
representative) collection of units from a population
used to determine truths about the population .(Shield
2005)
- Sample is a subset of the population.

 Sampling :
- The process of drawing a number of individual cases
from a larger population.
- A way to learn about a larger population by obtaining
information from a subset of a larger population.
Ex. Election – pre result survey sample.
Population : a set which includes all measurements of
interest to the researcher ( the collection of all
responses, counts that are of interest).

Target Population : A population of
theoretical interest.
Study Population : The group of elements
from which a sample is actually selected.
Element : That unit about which
information is collected & which
provides the basis for analysis.

 Representativeness : The extent to
which the sample “mirrors” the
population.
 Sampling Unit – That element or set
of elements considered for selection in
some stage of sampling.

 Sampling Frame : The actual list of sampling
units from which the sample or some stage of the
sample , is selected.
 Sample size : The number of elements selected.
 Parameter: Numerical characteristic of a
population.
 Statistic: Numerical characteristic of a sample.

Why Sample ?
 -To learn something about a large group without
having to study every member of that group.
 -Time and Cost :
- Studying every single instance of a thing is
impractical or too expensive.
Ex. Census
 Improve data quality :
- Obtain in-depth information about each subject
rather than superficial data on all.

Sample Size
 The more heterogeneous a population is,
the larger the sample needs to be.
 Depends on topic .
 For probability sampling, the larger the
sample size, the better.
 With nonprobability samples, not
generalizable regardless – still consider
stability of results.

Types of sampling
 Probability (Random) Samples
 Simple random sampling
 Systematic random sampling
 Stratified random sampling
 Multistage sampling
 Multiphase sampling
 Cluster sampling
 Non-Probability Sampling
 Convenience sampling
 Purposive sampling
 Quota sampling

Sampling Frames, Probability versus
Nonprobability Samples
 Probability Sampling :
A sample drawn in a way to give every member
of the population a known (nonzero) chance of
inclusion .
Probability samples are usually more
representative than nonprobability samples of the
populations from which they are drawn.

Probability Sampling
Probability sampling :
 A sample of a population in which each person has a
known chance of being selected.
 Basically an equal chance at the start.
Pro: it’s a simple process .
Con: a complete list of the population might not be
available.
It may include some “outside”.

Simple Random Sampling
 Elements are selected at random from the sampling
frame.
 It can be used by computer , table numbers , by writing
characteristics.
 Ex . Lottery
Advantages:
 Estimates are easy to calculate.
 Simple random sampling is always an Equal
Probability of Selection design, but not all EPS
designs are simple random sampling.

Disadvantages :
 If sampling frame large, this method
impracticable.
 Minority subgroups of interest in population
may not be present in sample in sufficient
numbers for study.

Stratified Random Sampling
 Where population embraces a number of distinct
categories , the frame can be organized into separate
“strata” . Each stratum is then sampled as an
independent sub-population , out of which individual
elements can be randomly selected.
 Every unit in a stratum has same chance of being
selected.
 Using same sampling fraction for all strata ensures
proportionate representation in the sample.
 Used in situations in which the researcher knows
some of the variables in the population that are
critical to achieving representativeness.
 Ex. Age, Gender, SES, Diagnosis, type of care….

 Finally, since each stratum is treated as an independent
population, different sampling approaches can be applied
to different strata.
Drawbacks:
 First, sampling frame of entire population has to be
prepared separately for each stratum.
 Second, when examining multiple criteria, stratifying
variables may be related to some, but not to others, further
complicating the design, and potentially reducing the
utility of the strata.

Stratified random sampling
Finally, in some cases (such as designs with a large
number of strata, or those with a specified minimum
sample size per group), stratified sampling can
potentially require a larger sample than would other
methods.

Multi Stage Sampling
 A probability sampling procedure that involves
several stages, such as randomly selecting clusters
from a population, then randomly selecting elements
from each of the clusters.
 Ex. – Random Digit Dial.
- Stage 1 : Areas codes randomly sampled
- Stages 2 : Three digit local exchanges randomly
sampled
- Stage 3: Last four digits randomly sampled
- Stage 4 : Asking the person who answers the
phone for the appropriate person you want to

Systematic Sampling
A probability sampling procedure that
involves selecting every kth element from a
list of population elements, after the first
element has been randomly selected .
Example.
Divide the total number of elements by
the number you want in your sample 24/6
= 4

 Randomly select a number between 1 and 4
and then select every 4th element from that
number.
 Selection Interval : The distance between the
elements selected in a sample.
Selection Interval = Population Size
Sample size

Cluster Sampling
 A probability sampling procedure that involves randomly selecting
clusters of elements from a population and subsequently selecting
every element in each selected cluster for inclusion in the sample .
 Cluster sampling is an option if data collection involves visits to
sites that are far apart.
 Example
 You are conducting a study of Bihar high school students.
 You could obtain a list of all high school students in the state
and complete random sampling.
 A cluster sample would be more practical
 Obtain a list of all high schools in Bihar.
 Random sample the high schools from the list.
 Obtain a list of students for each high school selected and
then contact each of those students.

Types of Non- probability Sampling
 Nonprobability Sampling.
 A sample that has been drawn in a way that doesn’t
give every member of the population a known chance
of being selected .
 Purposive Sampling:
 A nonprobability sampling procedure that involves
selecting elements/ samples based on a researcher's
judgment about which elements will facilitate his or her
investigation.

Types of Purposive sampling
1. Extreme or deviant cases:
The selection of the cases might be to choose
individuals or sites that are unusual or special in some
way.
Ex. Analyze the highly successful program and compare
with the fail one.
2. Intensity sampling
-It’s similar to the extreme – case strategy except
there is less emphasis on extreme.
-Explore rich information on the typical cases.

3. Maximum-variation sampling:
Maximize the variation within the sample and indicate
their major difference.
EX: The study of students’ English ability in different
location( rural, urban).
4. Homogeneous sampling :
The researcher seeks to describe the experience of
subgroups of people who share similar characteristics.
EX: rural women’s attitude toward surrogacy.

5. Typical Case Sampling :
- Choose the case in which a program has been
implemented to show this case is indeed average.
- It is like Intensity sampling.
6. Stratified Purposive Sampling :
- It’s a combination of sampling strategies.
- Subgroups are chosen on specified criteria – a
sample of cases is selected within these strata.
7. Critical Case Sampling :
- Study a very important critical case.
- The effect should be representative.

 Quota Sampling :
A nonprobability sampling procedure
that involves describing the target
population in terms of what are
thought to be relevant criteria and
then selecting sample elements to
represent the “relevant” subgroups
in proportion to their presence in the
target population .

Proportional Quota Sampling
• Objective: Represent major characteristics of
population by sampling a proportional amount
of each. For example, if you know the
population has 40% women and 60% men,
you want your sample to meet that quota.
• Problem: How do you pick the characteristics?
How do you know their proportion in
population?

Non proportional Quota
Sampling
• Making sure you have enough units
from each target group of interest
(even if not proportional).
• As with stratified random sampling,
you might do this to assure that you
have good representation of smaller
population groups.

Snowball Sampling:
 A nonprobability sampling procedure that involves using
members of the group of interest to identify other
members of the group.
 One person recommends another, who recommends
another, who recommends another, etc.
 Good way to identify hard-to-reach populations, for
example, homeless persons.

Convenience Sampling /Grab/
Opportunity /Haphazard Sampling
 A nonprobability sampling procedure that
involves selecting elements that are readily
accessible to the researcher
 Sometimes called an available-subjects sample .
 A type of nonprobability sampling which involves the
sample being drawn from that part of the population
which is close to hand. That is, readily available and
convenient.

 -The researcher using such a sample cannot
scientifically make generalizations about the total
population from this sample because it would not
be representative enough.
 For ex, if the interviewer was to conduct a survey
at a shopping center early in the morning on a
given day, the people that he / she could interview
would be limited to those given there at that given
time, which would not represent the views of other
members of society in such an area, if the survey
was to be conducted at different times of the day &
several times per week.
 Useful for pilot testing.

Event Sampling :
 Event Sampling Methodology (ESM) is a new
form of sampling method that allows researchers
to study ongoing experiences and events that vary
across and within days in its naturally-occurring
environment. Because of the frequent sampling of
events inherent in ESM, it enables researchers to
measure the typology of activity and detect the
temporal and dynamic fluctuations of work
experiences. Popularity of ESM as a new form of
research design increased over the recent years
because it addresses the shortcomings of cross-
sectional research, where once unable to,
researchers can now detect intra-individual
variances across time. In ESM, participants are
asked to record their experiences and perceptions
in a paper or diary.

Stages in Selection of Sample
 Select a sampling frame.
 Determine sampling method.
 Plan procedure fro selecting
elements.
 Define the target population.
 Estimate sampling size
 Draw sample
 Conduct field word

Errors in Sample
 Biased Selection – misses and/or over
represents categories of elements.
 Chance Variability – a sample deviates from
the population value as a result of chance –
increasingly problematic as sample size
decreases.
 Systematic Error (or bias) : Inaccurate
response (information bias).
 Sampling Error (random error).

Type 1 Error
 The probability of finding a difference with our
sample compared to population, and there really
isn’t one….
 Usually set at 5% (or 0.05)
 Type I Errors
 You reject a null hypothesis when you
shouldn't’t.
 You conclude that you have an effect when
you really do not.
 The alpha level determines the probability of
a Type I Error (hence, called an “alpha error”).

Type 2 Error
 The probability of not finding a difference
that actually exists between our sample
compared to the population…
 Power is (1- β) and is usually 80%.
 Type II Errors
Failure to reject a false null hypothesis
Sometimes called a “Beta” Error.

Sample Size
for a mean –
n = z2 2 / e2
where:
 e, the sampling error, is the difference between
sample mean and population mean .
[e is expressed in units]

for a proportion -
n = [z2 p (1 – p)] / e2
where:
 e, the sampling error, is the difference between
sample proportion and population proportion
[ e is expressed in percentage points].

How large should a sample be ?
 ..... It depends on...
 population size , characteristics
(homogeneousless / heterogeneousmore)
 spread of the data (sd)
 Purposes (descriptive? testing? (power))
Population sample Sampling Ratio
200 171 85.5%
500 352 70.4%
1000 543 54.3%
2000 745 37.2%
5000 960 19.2%
10000 1061 10.6%
20000 1121 5.6%
50000 1160 2.3%

Sampling class phd aku

More Related Content

What's hot (20)

Similar to Sampling class phd aku (20)

More from sangita singh (14)

Recently uploaded (20)

Sampling class phd aku