Sampling

 SAMPLING

Description also available in video format (attached below), for better experience use your desktop.

Introduction

·       Sampling: A process of selecting observations to give an adequate description of the population.

·       Sample: A unit that is selected from population & represent the whole population.

·       Sampling Frame: Listing of the population from which a sample is taken out or chosen.

 

Sampling Design & Procedure

Types of Sampling

Probability Sampling

 

1.    Simple Random Sampling

 

a.     A subset of individual persons is chosen from a population randomly, all with the same probability

b.    Benefits includes the minimal knowledge of population need and easy to analyze the data

c.     Limitations includes the low frequency of use and large risk of random error

2.    Stratified Random Sampling

a.     The population is divided into 2 or more groups known as strata

b.    And then subsamples are randomly selected from each strata

c.     Advantage includes the assurance of representation of all group in sample population

d.    Limitation includes the high cost to prepare the strata list

3.    Cluster Sampling

a.     The population is divided into subgroup known as clusters like families

b.    A simple random sample is taken from each cluster

c.     Advantage includes the characteristics of both cluster and population

d.    Limitations include the cost to reach an element to sample is very high and each stage in cluster sampling introduce an error

4.    Systematic Random Sampling

a.     It orders all units in the sampling frame and then every n^th number on the list is selected

b.    Just like in the above mentioned image every 3^rd person is present in the sample

c.     Benefits includes the simple way to draw the sample and easy to verify

d.    Limitation includes the requirement of periodic ordering

 

Non-Probability Sampling

 

1.     Convenience Sampling

a.      It involves the choosing of sample as per the convenience of the researcher

b.     Main advantages includes the low cost and extensively used

c.      Limitations includes the failure of Bias & Variability measurement and restriction of generalization

2.     Quota Sampling

a.      Population is first segmented into mutually sub groups, just as in stratified sampling

b.     Used when the budget of researcher is limited

c.      No need for the list of population element

d.     The main drawback is the failure of justification of projecting data beyond the sample

3.     Judgmental Sampling

a.      In this sampling the researchers employs his or her own expert judgement to choose the sample from the population

b.     Benefits includes the assurance of quality and meet the specific objective

c.      Limitation includes the bias selection and time consuming

4.     Snowball Sampling

a.      Research starts with a single & key person and introduce the next one to become a chain

b.     Advantages includes the low cost and for locating rare populations

c.      Limitation is the failure of justification of projecting data beyond the sample

 

Sample Size Determination

·       Sample size: the sub-population to be studied in order to make an inference to a reference population.

·       Sample size determination: Mathematical estimation of the number of subject which are needed to be include in the study to

·       To allow for appropriate analysis

·       To provide the desired level of accuracy

·       To allow the validity of significance tests

·       Approaches to calculate Sample size:

o   The Study Design

§  Case Control design

§  Cohort design

§  Cross sectional studies

§  Clinical trials

§  Diagnostic test studies

o   Primary Outcome Measure

§  Odds ratio

§  Relative ratio

§  Proportions

·       Procedures to calculate Sample size

o   Use of formulae

o   Readymade tables

o   Nomograms

o   Computer Software

 

Sampling Errors

·       These are the errors which arises due to the use of sampling surveys

·       Generally classified into two types

o   Biased Errors: Due to the selection of sampling techniques and size of the sample

o   Unbiased/Random sampling errors: Differences between the members included or not included in the sample size

·       Methods that can be used to reduce the sampling errors are

o   Specific problem selection

o   Systematic documentation

o   Effective enumeration

o   Effective pre testing

o   Controlling methodological bias

o   Selection of appropriate sampling techniques

 

Video Description 

     Don’t forget to do these things if you get benefitted from this article

o   Visit our Let’s contribute page https://keedainformation.blogspot.com/p/lets-contribute.html

o   Follow our  page

o   Like & comment on our post