INTRODUCTION
DEFINITION
Sampling is a process used in statistical analysis
in which a predetermined number of observations
are taken from a larger population .
METHODOLOGY
The methodology used to sample from a larger
population depends on the type of analysis being
performed , but may include simple random sampl
ing or systemic sampling.
BASIC TERMS
POPULATION
Any complete group of entities that shares some common set of characteristics.
POPULATION ELEMENT
An individual member of population .
TARGET POPULATION
Refers to the entire group of Individuals or objects to which researchers are
interested in generalizing the conclusions.
CENSUS
An investigation of all the individual elements that make up a population
CONT…..
SAMPLE
A subset of population used to study the population, means a smaller manageable
version of large group of population.
SAMPLING UNIT
An individual member of population.
SAMPLING FRAME
List of all elements.
SAMPLE DESIGN
The method we use to select our sample.
SAMPLE STATISTICS
The information obtained from our respondents .
PURPOSE OF SAMPLING
To estimate a population parameter.
To gain an impression of an area or collection of things.
To test hypothesis.
It is used when the data is unlimited.
Time , money and energy is saved.
It is impractical to study the entire population.
STAGES IN SAMPLE
SELECTION
Determine which sampling method will be chosen.
Select a sampling frame.
Define the target population.
Plan procedures for selecting sampling unit.
Determine sample size.
Select actual sampling units.
Conduct field work.
PROBABILITY
SAMPLING
In probability sampling a sample from larger population
is chosen using a method based on theory of probab
ility.
It relies on random judgment.
It can be used to estimate the distribution of an opinion
in the entire population.
All persons have chances of being selected.
Results are more likely to reflect the entire population.
Probability sampling requires a sampling frame, and w
hen a sampling frame is not possible, non probability s
ampling is used.
1. SIMPLE RANDOM SAMPLING
Simple random sampling is a process in which
selection of sampling unit from population is based on
chance.
Each member of the subset has an equal probability of
being chosen.
It is meant to be an unbiased representation of a group.
Selection process:
• Identify and define the population
• Determine the desires sample size
• List all the members of population
• Assign all the members on the list a number
• Ensure that number are chosen randomly by random n
umber table, lottery method, or using the number on
currency notes.
2. SYSTEMATIC SAMPLING
Systematic sampling involve the method by which larger population are
selected according to a random starting point but with fixed periodic
interval.
Selection process:
• Defining the population
• Decide the sample size
• Listing population and assigning number to cases
• Calculating sampling interval
• Select the first unit randomly
• Select every kth unit from there
3. STRATIFIED SAMPLING
The population is divided into two or more then two groups called strata,
according to some criterion, such as geographic location, age or income
and subsamples are randomly selected from each strata.
Selection process:
• To split the population into sections.
• The strata are chosen to divide a population into important categories
relevant to the research interest.
• Then draw the sample in proportion to its size.
4. CLUSTER SAMPLING
The process of randomly selecting intact groups, not individuals, within the
defined population sharing similar characteristics.
The population of each cluster must be known.
The complete list of all the individual in the country is not necessary.
Selection process:
• Define the population.
• Determine the desired sample size.
• Identify and define a logical cluster.
• Estimate the average number of population members.
• Randomly select the needed number of cluster
NONPROBABLITY SA
MPLING
In nonprobability sampling, a particular member of
the population being
chosen is unknown.
It relies on personal judgment.
Each element of the population does not have an e
qual chance of being included in sample.
The researcher cannot estimate the error.
1 . CONVENIENCE SAMPLING
Convenience sampling is made up of people who are easy to reach
.
There is no other criteria to the sampling method except the people
to be available and willing to participate.
It can also be called as accidental or haphazard sampling.
Selection process:
• Use of students.
• Members of social organization.
• People on the street interviews.
2. JUDGMENTAL SAMPLING
Judgmental sampling is a technique in which researcher select sample
units to be sampled based on their knowledge and professional
judgment.
It is used when a limited number of individuals possess the trait of
interest.
It can be used when the researcher knows a reliable professional or
authority that he thinks is capable of assembling a representative
sample.
It is used when researcher feels that other sampling techniques will
consume more time.
3. QUOTA SAMPLING
Quota sampling is a technique in which the assembled sample has the
same proportion of individuals as the entire population with respect to
known characteristics.
Selection process:
• Divide the population into subgroups.
• Identify the proportion of subgroups in the population .
• Researcher select subjects from various subgroups.
• Ensure that the sample is representative of the entire population.
4. SNOWBALL SAMPLING
Snowball sampling is a technique where existing study subjects recruit f
uture subjects
Data that is gathered is useful in research.
It is often used in hidden populations such as drug users which are diffic
ult for researcher to access.
Selection process:
• Identify the potential subjects in the population.
• Only one or two subjects can be found initially.
• Participants should be made aware that they don't have to provide any o
ther names.
• These steps are repeated until the needed sample size is found.
MERITS AND DEMERITS OF
SAMPLING
MERITS:
Less time consuming.
No repetition of query.
Accuracy of data is high.
Scope of sampling is high.
Better rapport.
Detailed information.
DEMERITS
Chances of biasness.
Improper selection.
Exclusion of data.
Proper size of sample is a difficult job.
Lack of knowledge regarding the topic.
Without proper planning, results would be unreliable.
PROBABILITY SAMPLING:
MERITS: DEMERITS:
Cost effective.
Involves lesser degree of judgement. Only specific samples types are collected.
Easier way of sampling. Redundant and monotonous work.
Quick to gather, takes less time.
Easily compiled by non experience p
erson.
1. SIMPLE RANDOM SAMPLING:
MERITS: DEMERITS:
Whole population represented by the
sample.
Less range of variation.
Conclusion in less time. Redundant and monotonous work.
Less costly.
Lesser degree of judgement, can be done
by nontechnical person.
Easier way of sampling.
2. SYSTAMATIC RANDOM
SAMPLING
MERITS: DEMERITS:
More convenient. Size of population may not be known
before sampling.
Easy to understand. Assumes that the population is uniform.
Cost effective. Difficulty in gathering people in a crowd.
Higher degree of control.
Less time consuming.
3. STRATIFIED RANDOM
SAMPLING
MERITS: DEMERITS:
More precise. Requires more work.
Minimizes the biasness. Hard to classify each kind of population.
Greater accuracy.
No over-represented or under-represented.
4 . CLUSTER SAMPLING
MERITS: DEMERITS:
Less costly for surveys. Least representative of population.
It can apprehend both population and clust
er.
Statistically less efficient.
Very useful when population are large and
spread over a large area .
Sampling error is high.
Do not need specific names within populati
on.
NON POBABILITY SAMPLING
MERITS: DEMERITS:
Cost effective. Excessive dependency on judgement.
Works best if the exhaustive population is
not defined.
Needs much purpose-oriented pollsters.
Best method of sampling if sound knowled
ge on the subject.
Focuses on simplicity over effectiveness.
1. CONVENIENCE SAMPLING
MERITS: DEMERITS:
Easy method. Fails to represent whole population.
Less time consuming. Whole system may become useless.
Economic way of sampling
2. SNOWBALL SAMPLING:
MERITS: DEMERITS:
Referral system helps find sample quickly. Potential sampling bias.
Works for hesitant subject. Chance of lack of co-operation.
Secretive groups can be identified easily. Peer network might not exist.
3. QUOTA SAMPLING:
MERITS: DEMERITS:
Saves time. Non-random
Extra information speeds up sampling proc
ess.
Ignoring important characteristics for ease.
High accuracy.
4. JUDGEMENTAL SAMPLING:
MERITS: DEMERITS:
Less costly,
More accessible,
More convenient.
No guarantee that chosen sample are true.
Select only those individual whom are relev
ant to research purpose.
Potential for inaccuracy in researcher’s crit
eria and resulting in sample selection.
SAMPLE SIZE:
DEFINITION:
Sample size measures the number of individual sample and
observations used in a survey or experiment.
CHOSING SAMPLE SIZE:
Experience.
Target variance.
Statistical test.
Confidence level.
DETERMINING OF SAMPLE SIZE:
It is the mathematical determination of the number of subjects
that are to be included in the study.
When a sample is taken from a population the findings are gen
eralized to the whole population.
Optimum sample size determination is required for the
following:
1. To allow appropriate analysis.
2. To provide the desired level of accuracy.
3. To allow validity of significance test.
FACTORS FOR DETERMINING
SAMPLE SIZE:
1. Number of groups and subgroups within a sample.
2. Value of information in the study.
3. Accuracy level required in studies.
4. Cost of sample.
5. Variability of the population.
6. Nature of data and data size.
7. Kind and number of comparisons.
8. Homogeneity of samples.
CONT…..
ADVANTAGES OF HAVING A
LARGE SAMPLE SIZE:
DISADVANTAGES OF HAVI
NG SMALL SAMPLE SIZE:
Low Sampling Error. Variability.
Precision. Undercoverage Bias.
Confidence Intervals. Voluntary Response Bias.
Margin of error decreases. Less accuracy.
ERRORS IN SAMPLING:
1. Sampling error.
2. Non-sampling error.
3. Processing error.
4. Response error.
5. Open and closed end question error.
6. Lying in sampling error.
7. Non-responsive error.
8. Dropout.
9. Result error.
CHARACTERISTICS OF A SAMPLE:
A GOOD SAMPLE CONTAINS:
Gender
Age
Income
Education level
Geographical location
Sample size
CHARACTERISTICS OF SAMPLE
DESIGN:
Sample design must result in a truly representative sample.
Sample design must be such which results in a small sample error.
Sample design must be viable in the context of funds available for the research
study.
Sample design must be such so that systematic bias can be controlled in a better
way.
Sample should be such that the results of sample study can be applied in general
for the universe with a reasonable level of confidence.
Representative.
Appropriate size.
Unbiased.
Random.
SAMPLING CRITERIA :
DEFINITION:
A complete set of elements that possess some common
characteristics defined by the researcher.
Merits of criterion sampling:
It is useful for identifying and understanding cases that are
information rich .
It can provide important qualitative component to qualitative
data
SAMPLING CRITERIA ERRORS:
The errors in sampling criteria are as follows:
Inappropriate sampling frame.
Defective measuring device.
Non-respondents.
Indeterminancy principle.
Natural bias in reporting the data.
ETHICS:
Researcher’s ethical responsibility to safeguard the story teller by maintai
ning the understood purpose of research.
The relationship should be based on trust between the researcher and pa
rticipants
Inform participants of the purpose of the study.
Research participants should not be subjected to harm in any way.
Respect for the dignity of research participants should be prioritized.
Full consent should be obtained from the participants prior to the study.
Protection of the privacy of the research participant has to be ensured.
CONT….
• Adequate level of confidentiality of the research data should
be ensured.
• Anonymity of the individual and organizations participating in
the research has to be ensured.
• Any deception or exaggeration about aims and objective of
the research much be avoided.
• Communication in regards of research should be done with
honesty.
• Avoid using any biased data, false information and findings.