Sampling and sampling distribution
SAMPLING
It is the process of selecting a sample from the population
for study
A Finite subset of population, selected from it, with the
objective of investigating its properties is called ‘sample’.
The number of units in the sample is known as sample size.
Why is SAMPLING Essential?
Sampling saves time
Sampling saves money.
In most cases complete census is not possible
and hence, sampling is the only option left.
Sampling broadens the scope of the study in
light of the scarcity of the resources.
Sampling principles
• Principle or Law of statistical regularity: this law is based
upon mathematical theory of probability. It is based upon the
following two conditions.
i) Large sample size: as the sample size increases,
the true characteristics of the population are more likely to
reveal.
ii) Random selection: the sample should be selected
randomly in which each and every unit of the universe has
an equal chance of being selected.
• Principle of inertia of large numbers: – it is based upon
the concept that as the sample size increases the better results
we will get.
Probability Methods
Simple random sampling: In which each element of
the population has an equal and independent chance of
being selected.
Lottery method and Round table numbers
Stratified sampling: It involves developing strata
which are distinctly different from each other but are
homogeneous within. Thereafter units are chosen from
each strata randomly.
Systematic sampling: Every nth item within a
defined population is selected. It involves beginning
with a random start of an element range of 1 to n.
Cluster sampling: Groups of elements that are
heterogeneous in nature form a group and then the
groups are choosen randomly.
Multistage sampling
Ist Stage--------------- States
IInd Stage------------- Districts
IIIrd Stage-------------- Cities
Ivth Stage-------------- Blocks
Sampling techniques
Non probability
Convenience sampling: The researcher has the
freedom to choose any respondent based on his
convenience.
Judgement sampling: The respondents are
deliberately made a part of a sample by virtue of their
position, knowledge or any other criteria meet the
research purpose.
Quota sampling: Groups are created which are
homogeneous within and a quota is fixed for each
group.
Snow ball sampling: A set of respondents are selected
initially and interviewed and they further recommend
the names of other people.
SNOWBALLING
Choosing the right size of sample for your survey or
experiment depends on the answers to 3 important
questions.
1. How much variability is there in the population? The
more variability, the larger the sample needed.
2. What proportion of the population has the attribute you
are interested in?
3. How large a margin of error will you tolerate, and what
confidence level will you be using? 95%? Smaller margins
of error require larger samples.
Design of experiments
Health service boss Mr. vinod is delighted with the latest
scheme at Avonford Hospital that aims to cut the number of
days lost through absenteeism. Cleaning, catering and some
care staff who have six months unbroken attendance record
are being given 25000 Rs high street vouchers to spend on
whatever they like as a reward for simply turning up to
work.
Mr. Vinod says “This may sound a crazy and expensive
idea, but I have seen it work in other sectors so wanted to
give it a try here – and it is actually saving us money, leave
alone reducing the hassle of having to find cover at the last
minute.”
Health Boss,Mr Vinod, has conducted an experiment. He
has done something, by introducing vouchers for good
attendance, observed what happened, compared the outcome
with what happened before the voucher scheme was
introduced, and evaluated the result.
The reason for conducting an experiment is to test a
hypothesis; in the Avonford Star article the hypothesis was
that introducing High Street Vouchers as a reward for good
attendance would reduce absenteeism.
Sampling distributions
Researcher selects a sample and computes the statistics to
make inferences. On the basis of these inferences the
researcher makes inferences about population parameter. So
it is important to have clear understanding about the
distribution of the sample statistics.
Sample mean is a commonly used statistics in the inferential
process.
A sampling distribution is a probability distribution of a
statistic that is obtained by drawing a large number of
samples from a specific population.
For making sampling distribution clearer, we will
take a population with a particular distribution after
this we will randomly select a sample of given size
The sample mean will be calculated and distribution
of sample mean will be determined
What is a survey?
• A Survey is defined as a research method used
for collecting data from a pre-defined group of
respondents to gain information and insights on
various topics of interest.
• The need to observe or research facts about a
situation leads us to conduct a survey. As we
mentioned at the beginning
The steps of a survey are:
• Formulation of the Statement of Objectives
• Selection of a survey frame
• Determination of the sample design
• Questionnaire design
• Data collection
• Data capture and coding
• Editing and imputation
• Data analysis
• Data dissemination
• Documentation
Usefulness in Survey Analysis
• Presents the data collected through survey in the
meaningful form.
• Simplifies the raw data collected through survey.
• Helps in formulating and testing of hypothesis.
• Reaching on valid inferences and enhance decision
making.