CENSUS AND SAMPLE SURVEY

 

 

 

 

Chapter-4

CENSUS AND SAMPLE SURVEY

 

CENSUS AND SAMPLE SURVEY

There are two methods of survey to collect data :-

1. CENSUS METHOD

It is also known as census survey. In this method, every element of the population is included in the investigation. When complete enumeration of all the items of the universe is made, it is known as census survey. e.g. population census is carried out once in every 10 years in India. This is done by using census method.

2. SAMPLE METHOD

When enumeration of few units of the population or universe is done it is called. Sample method of collection of data.

For example

(a)        Checking the quality of rice

(b)        A blood test

IMPORTANT DEFINITIONS

POPULATION OR UNIVERSE

It means the total number of data in a particulars enquiry.

 

SAMPLE

A small population of items selected from the universe for statistical studies.

 

 

METHODS OF SAMPLING

There are 2 methods

1. Random Sampling : – It is the method used for selecting a sample. It means that chances of each item of the universe being included in the sample are equal. There are 2 types of random sampling.

(a)        Simple and Unrestricted Random Sampling :In this method, each item of the universe has an equal chance of being selected in a sample. It is free from discrimination and human judgement. It depends on the law of probability which decides the inclusion of items in a sample. There are two methods of obtaining the simple random sample. These are : –

(i)         Lottery Method :This is the simplest method of simple random sampling. All the items of the universe are numbered and these numbers are written on the identical papers. These are mixed in a bowl and then these starts the selection by draw one by one. These numbers are picked out blind folded. All the slips must be identical in shape, size and colour.

(ii)        Table of Random numbers :A table of random digits is simply a table of digits which have been generated by random process. Following tables of random digits are available.

• Tipet’s random sampling table

There are 10,400 numbers

• R and corporations : a million random digits.
• Fisher’s and Yate’s table having 15000 digits.

Includes 1000 random numbers

Merits: of Simple random Sampling

(a)        It is a scientific method of taking out a sample from the universe. Every item in the universe haveanofsimple random samplingequal chance of being selected.

(b)        It is more representative of population when size of the sample increases.

(c)        This method is economical as it saves, time labour and money.

(d)        Sampling error cannot be measured

(e)        Theory of probability is applicable, if the sample is random.

Demerits :

(a)        This requires complete list of population but upto date list are not available in many enquires.

(b)        If the sample is small, than it will not be a representative of population.

(c)        The numbering of the list and preparation of the list is time consuming if the population is large.

(b)        Restricted Random Sampling :following are the methods :-

(i)         Sratified Random Sampling or Mixed Sampling :In this method, the universe is divided into stratas or homogeneous groups and an equal sample is drawn from stratum at random. This method is useful when population is not fully homogeneous.

Merits :

(a)        The sample taken under this method is more representative of population.

(b)        It ensures greater efficiency as each group consist of homogenous items.

(c)        For non-homogenous population, it is more reliable.

(d)        It is easy to administor sub-divided data.

Demerits :

(a)        This method is not useful unless some information concerning the population and the strata is available.

(b)        If proper stratification is not done, the sample will have an effect of bias.

 

(ii)        Systematic Sampling or Quasi Random Sampling :This method is used when complete list of the population is available.

Merits :

(a)        It is systematic, very simple and convenient.

(b)        In this method, time and work is reduced much.

(c)        The results are also satisfactory.

Demerits :

(a)        It is feasible only when units are systematically managed.

(b)        If the universe is arranged in wrong manner, the results will be misleading.

 

(c)        Cluster or Multi-Stage Sampling :In this method, sampling is carried out in the number of stages. In this method, universe is divided and sub-divided according to the observations. At each stage sampling is done by suitable method.

2. Non-Random Sampling :Non-Random Sampling is done on the basis of convenience and judgment of the investigator and not on the basis of probability as against random sampling. These are the methods of non-random sampling.

(a)        JUDGEMENT AND PURPOSIVE SAMPLING

In this method, particular item of sample is selected by the investigator consciously by his judgement.

In order for the judgement sampling to be reliable, it should be free from individual bias and prejudice.

Suitability

1. The number of items in the universe should be small.
2. When small size sample is to be drawn.

 

 

(b)        QUOTA SAMPLING

In the quota sampling, quotas are set up according to the given criteria. But within the quota, the selection of sample items depends on personal judgment. This method is commonly used in the survey of political, religious and social opinion.

e.g. An investigator is asked to interview 50 students taking 10 students each from VIII,IX, X, XI, XII. The investigator can select any 10 students according to his own judgment.

(c)        CONVINIENCE SAMPLING

            In this method, sample is selected, purely on the basis of convenience.

e.g. for interviewing the students of the school, the interviewer or may select a school nearby to his house because it is convenient for him. But results obtained are not truly representative of the universe.

ADVANTAGES OF SAMPLING

1. Economy In Time : –

Time is saved not only in the collection of data but also because of the smaller size in analysing it.

2. Economy in Cost : –

            Less efforts are required so it is much more commercial than census method.

3. Minimum Error : –

            The total error is much lower in the sampling method.

4. Follow up : –

With the help of trained investigator and small size of data, it is possible to cross check the reliability of the data.

5. Indispensability : –

In certain circumstances, sampling method is the only possible or alternative means of investigation.

SAMPLING AND NON-SAMPLING ERROR

The errors arising on account of drawing conclusions about the population on the basis of few observations are called sampling error.

The errors mainly arising at the stage of assertainment and processing of data are called non-sampling error. The data obtained in the census by the complete enumeration is free from sampling error but it is subject to non-sampling error where as results of the sample survey would be subject to the sampling and non-sampling error. Following are the reasons of non-sampling error.

(i)         Lack of trained and experienced investigator.

(ii)        In appropriate structural unit.

(iii)       Errors due to non-response.

(iv)       Errors in the data processing operations such as coding. Verification and tabulation etc.

Census Survey		Sample Survey
(1)	Every item of the universe is taken under study	(1)	A part of the population is taken under study
(2)	Many difficulties are faced when universe is large	(2)	Lesser difficulties are faced because fewer units of the universe are observed
(3)	Large number of investigator need to be appointed to collect the data	(3)	A small number of trained investigator can collect the complete data.
(4)	It is an expansive method, takes more time and efforts to collect the data	(4)	Less time and money is required to collect a sample
(5)	Results are drawn from the data of complete universe	(5)	Results are drawn from the sample of the universe
(6)	It is not a scientific method	(6)	It is a scientific method
(7)	Higher accuracy can be obtained	(7)	If sample are not drawn properly, results may be inaccurate, false or misleading.

 

NATIONAL SAMPLE SURVEY ORGANISATION

The NSS was started in the year 1950. In March 1870, NSS was recognized and it was bought under single government organization namely, NSSO.

The NSSO has four divisions, namely

• Survey design and research division
• Field operation division.
• Data processing division
• Co-ordination publication division

FUNCTIONS OF N.S.S.O.

(i)         Collection of data on socio-economic concern.

(ii)        Collection of data relating to the recognized industrial sector.

(iii)       Supervision of survey conducted in the agriculture sector and giving guidance for anlysing and co-ordinating the results of the survey.