Sampling

Sampling is a process used in statistical analysis in which a predetermined number of
observations are taken from a larger population.
Types of sampling
Probability
- Simple Random sampling
- Stratified
- Area or Cluster
Non Probability
- Quota
- Purposive or judgement
- Convenience or accidental or incidental
- Snowball
- Systematic
- Saturation
- Dense
Population and sample size must be clearly defined prior to data collection
Goal of sampling is to have a sample that represents a population as accurately as possible.
1. Double sampling example- Step 1: n' initial simple random samples are selected from a
population of N units. ...
2. Step 2: A second sample is then selected by stratified random sampling from the first sample. ...
Advantage of Probability technique

1) They can use powerful statistical methods also known as parametric stats.
2) Non- Probability have to use less powerful statistical method called as non-parametric
stats.
Parametric statistics is a branch of statistics which assumes that sample data comes from a
population. Based on assumption.
Nonparametric statistics make no assumptions about the probability distributions of the
variables being assessed. include both descriptive and inferential statistics.
Probability
Simple Random sampling

- Every person has an equal chance to represent population
- Also called unrestricted random sampling
Four to five chairs are arranged in an inner circle. This is the fishbowl. The remaining chairs are
arranged in concentric circles outside the fishbowl. A few participants are selected to fill the
fishbowl, while the rest of the group sit on the chairs outside the fishbowl.
Simple random without replacement

In this method when selecting from a population the sample taken in the first selection will be
replaced and not chosen again. So the total count will keep on decreasing
For example: If u draw the number 4 from a jar of numbers from 1 to 10 u will take the 4 out
and not use it again in the next draw.
Simple random with replacement
In this method when selecting from a population the sample taken in the first selection will not
replace and can be chosen again. So the total count will remain the same.
For example: If u draw the number 4 from a jar of numbers from 1 to 10 u will keep the 4 back
inside the jar and use it again in the next draw.
Advantages
 It is considered representative of the population

 Researcher need not know the composition of population beforehand.
 Sampling error can be easily measured
Disadvantages
 Small group in the population might not be represented in the sample.

 Sampling error is greater than stratified random sample.
Stratified Random Sample
In stratified random sampling, the strata are formed based on members' shared attributes or
characteristics.
Strata---- layers or sections
It is used with a sample population easily broken into distinct subgroups and they collect each
data from each sample and pool it together.
These subpopulations have individuals selected randomly.
In this method the population is defined in terms of strata for example gender, educational
level, socio- economic status etc.
Types of Stratified Random Sampling
1) Proportionate- The sample ratio is proportionate to the population strata ratio.
Example: Reservation seats (each state gets number of seats according to their
population not size of area like Maharashtra get more seats as compared to Madhya
Pradesh even though it is bigger compared to Maharashtra)
2) Disproportionate- The sample ratio is not proportionate to the population strata ratio.
Advantage of proportionate over disproportionate
- Proportionate is more representative as compared to disproportionate.
- Sampling error is minimum
Disadvantage
1) It is time consuming
2) Researcher might give importance to factors that are not significant to the study.
Advantage of Disproportionate
1) Less time consuming
Disadvantage of Disproportionate
1) It is less representative as compared to proportionate
2) Sampling error is maximum or more compared to proportionate
Area or Cluster sampling
Cluster sampling is a sampling technique used when "natural" but relatively heterogeneous
groupings are evident in a statistical population.
- It is important for geography
- It is used in psychology
Advantages
1) A number of geography locations of the population are chosen at random and majority of
that area is interviewed or surveyed.
2) Incase the individual are heterogenous random selections of subdivision take place, also
called multistage sampling.
3) It is simpler than stratified proportionate sample.
4) Cost and time effective
5) Can be useful for places where census data is missing.

A census is the procedure of systematically acquiring and recording information about the
members of a given population.
Disadvantages
1) Sampling error is higher as compared to the other two methods
Non Probability
Quota Sampling
A quota sample a type of non-probability sample in which the researcher selects people
according to some fixed quota. For example gender
Researchers identifies different quotas of population and then selects individuals arbitrary
(chosen randomly)
Advantages
1) It gives a better representation of certain groups within the population without over
representing them.
2) It makes comparison of the group easier.
Disadvantages
1) Giving some weightage to strata of society
2) Time consuming
3) Higher sampling error because of inaccurate representation.
3) Researchers bias comes into play
Purposive or Judgemental Sampling

is a sampling technique in which researcher relies on his or her own judgment when choosing
members of population to participate in the study.
For example- TV reporters stopping certain individuals on the street in order to ask their
opinions about certain political changes
Advantage
1) The approach is understood as well as refined through experience over many years.
(the researcher has many experience over so many years to pick members for study)
2) Opportunity to bring this judgment and expertise to play
- Researchers bias might differ the result of something

Disadvantage
1) Researchers bias comes into play
2) Sampling error is high as it is not representative of the population.
Accidental or Incidental Sampling
(sometimes known as grab, convenience sampling or opportunity sampling) is a type of non-
probability sampling that involves the sample being drawn from that part of the population
that is close to hand. Example: To ask about the culture of spain u would choose people next to
you.
Advantage
- It is the easiest, least time consuming or most time effective.
Disadvantage
- Sampling error is highest because it is not representative of the population
Snowball Sampling
It is a sampling technique used when the members of the population are difficult to locate.
Collect data from few members of target population that he or she can locate and then ask
those individuals to provide information needed to locate other members.
The sample is said to grow like rolling snowball.
Advantage
1) It is representative of the population.
2) Can have a huge sample size
3) Useful in studying social constructs and structures in organization
4) Locate hidden population.
5) Cost effective
Disadvantage
1) Small sample size if not done correctly
2) Difficult to control bias of researcher and of the sample
( For example while collecting data if you ask a doctor and they ask an intern and not a
doctor then data collection can be wrong.
Complex and time consuming

Systematic Sampling
You select every nth individual from a predetermined list, where n equals the population size
divided by the sample size.
(For example- In the group of 100, you divide it into smaller groups of 10 in which if you pick 7 th
sample in the first group then you will have to choose 17th in second and 27 in the third etc)
Has some level of randomness but it is not
To choose the nth element
Divide the population by sample size
For example- When there is population of 500 and we are taking groups of 50 then we have
divide
500/50= 10
So every 10th element you will choose like from 1 to 50 u will choose 10 then from 60 to 100
you will choose 60 so on and so forth
Advantage
1) Quick and easier to extract the sample than simple random
2) Ensures cases are spread across the population
3) Time effective
Disadvantage
1) Moderate level of sampling error
2) Not representative of the population
(If the list is not in order the sampling error is high and if list is in order the sampling error is
low)
Example- If you want to pick the 5th tallest person in group they should be first put height
wise and then 5th tallest is picked. The sample error is low as compared to randomly picking
the 5th person in each group for their height)
Saturation Sampling
Any person who possess a trait of interest is taken in sample and in this sampling technique you
need majority of the population.
For example in a population of 10,000 you need to take 4,000
Dense sampling
More than 50% of the population becomes your sample
For example in a population of 10,000 you need to take 6,000
Advantage for Dense and saturation
1) More representative of the population

2) Sampling error is lower as compared to other non-probability methods.
Disadvantage
1) Not quick, time consuming
2) Costly
These two methods are used when the population is very specific or very small
Example- Out of 2000 patients 100 have mood disorder so only 100 is the target population and
sample can be 50 or more
Double Sampling
Initially sample of unit is selected for obtaining axillary or basic information only and then a
second sampling is selected in which the additional information is selected.
In the amity population AIBAS is selected initially to get the basic information then the first year
students are taken to get the additional information. This is double sampling is the concept.
Mixture of more than 1 method or sample to create a smaller sample
Advantage
1) Representative of the population because sampling error is lower.
2) It is usually the most effective and practical way to carry out large surveys in the public
Disadvantage
1) Consists of complex calculations
2) Time consuming
3) Costly
4) Complex
First take quota then dense because quota you are choosing according to gender or socio
economic status etc then you will get dense most of the population will be represented
If there is not a clear accept or regret decision of the first sample then a second sample is taken
and the decision is made based result of the combined sample
Preferred is first is probability then non probability

From random to specific or low sampling to high sampling error
Like from representing population to sample hence ^

Sampling

Uploaded by

Copyright:

Available Formats

Sampling

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sampling

Uploaded by

Copyright:

Available Formats

Sampling is a process used in statistical analysis in which a predetermined number of

observations are taken from a larger population.

Advantage of Probability technique

Simple Random sampling

Simple random without replacement

 It is considered representative of the population

 Small group in the population might not be represented in the sample.

3) It is simpler than stratified proportionate sample.

4) Cost and time effective

5) Can be useful for places where census data is missing.

Purposive or Judgemental Sampling

- Researchers bias might differ the result of something

Complex and time consuming

For example in a population of 10,000 you need to take 4,000

Advantage for Dense and saturation

1) More representative of the population

Preferred is first is probability then non probability

You might also like