1

.• Collection of Primary and Secondary Data tools are used in studying the economic problems of
.• Organisation of Data a country. Thus, these statistical
tools help in testing, evolving and creating new fields
.• Presentation of Data of different sciences. Collection of
data-Sources and methods, is the first step in
UNIT 2 statistics towards achieving the goal of
conclusion.
COLLECTION AND ORGANISATION OF DATA
Enquiry means a search for truth, knowledge or
---------- Chapter 3 ---------~ information. Statistical enquiry,
therefore, means a search conducted by statistical
COLLECTION OF PRIMARY AND SECONDARY methods. There are different subjects
DATA .-------, on this earth; some are described by the degree of
expression (quality) and some by the
1. What is a Statistical Enquiry? degree of figures or magnitudes (quantity). The
application of a statistical technique is
2. Sources of Data possible when the questions are answerable in
figures (quantity), in other words the first
3. Primary and Secondary Data Collection of Primary and Secondary Data .tld 27

4. Drafting the Questionnaire and the foremost condition for the answer to the
questions in statistical enquiry should
5. Methods of Collecting Primary Data be quantitative, for instance :

6. Census and Sample Surveys Profit of firms measured in rupees;

Income of families measured in rupees;
7. Sample Surveys Weight of students measured in kg;
Age of students measured in years;
8. Methods of Sampling
Intelligence measured in marks obtained by students
9. Random Sampling in a particular test.

10. Non-Random Sampling But, there are questions like-How great was
Jawaharlal Nehru? How brave was
11. Advantages of Sampling Bhagat Singh? etc., which cannot be answered
through statistical methods. Questions that
12. Reliability of Sample Data can be answered in quantity lies within the purview
of statistics, viz., What is the average
13. How Secondary Data is Collected? production of rice per acre in India? What is the total
population of India? How many
14. Some Important Sources of Secondary Data students are there in a class?

15. Census of India Thus, statistical enquiry means statistical

investigation or statistical survey; one who
16. National Sample Survey Organisation (NSSO) conducts this type of enquiry is called an
investigator. The investigator needs the help
Meaning of economics and the importance of of certain persons to collect information, they are
statistics in economic analysis and known as enumerators, and respondents
economic planning have been discussed in Unit 1. are those from whom the statistical information is
We have seen briefly how the statistical collected. Survey is a method of
 2

collecting information from individuals.

2. How do we collect data?

3. By which method of survey is data collected?

There are different sources of collection of data. This

is the first stage in statistics.

Before deciding the source to collect the data one

has to make a proper planning of
investigation and the purpose of enquiry. Following
are the various sources of collection
of data.
Let us observe the following table. 28 ill Statistics for Economics-XI

TABLE 1: Production of Finished Steel in India (in I SOURCES OF DATA I

Million Tons)
I
Year Production
(X) (Y) *
1950-51 1.0
1980-81 6.8 ~
1990-91 13.5
2000-01 30.3 +*
2001-02 31.1
2002-03 34.5 Primary Secondary I
2003-04 36.9
Internal : A large number of individual organisations
Source : Government of India, Economic Survey and Government departments
2004-05. generate the data as their regular function which is
the internal information. They gather
We observe that the production of finished steel in this information for printing regular reports for their
India is different from one year own purpose. Internal data may be
to other. They are not same. They varies from year available in the organisation about sales, production,
to year. They are called Variables in salary, wage, profit, dividends, etc.
statistics which are represented as X, Y or Z Such data are always available in the books of the
variables. The finished steel production in organisation which are compiled and
India varies between 1.0 million tons in 1950-51 to sent to management for future planning.
36.9 million tons in 2003-04 Government departments, viz, Railways, Social
(see Table 1). Here years are represented by X- Welfare, Education, etc., do generate a large number
variable and the production of finished steel of internal data for preparing
Y-variable. The figures of X and Y variables are the financial reports and extraordinary reports.
'data'. We can say that the trend
of production of finished steel in India is increasing. External : Information collected from outside
We need the data to understand the agencies is called external data which
given information on which the conclusion is drawn. can be obtained from primary source or secondary
source. This type of information can
From the following text, we will understand: be collected by census or sample methods by
conducting surveys and investigations.
1. What is the source of data?
 3

Internal

Primary data is original and first-hand information. If

you want to get some statistical
information about payment of wages of workers of a
factory you as an investigator can
visit the workers' colony, get this information by
contacting every worker by enquiring
about the wages paid to him and record it, this first-
hand information is the primary
data. But, you also have the other choice that of
visiting the factory's accounts department,
and record the information from the salary register
or, may gather this information from
the published report of the factory about the
payment of wages. This is secondary source
for an investigator but, for the factory it is a primary
source.

Thus, primary data is collected originally and

secondary data is collected through
other sources. Primary data is first-hand information
for a particular statistical enquiry
while the same data is second-hand information for
an another enquiry. The same data
is primary in one hand and secondary in the other,
e.g., any Government publication is
first-hand (Primary) for Government and second-
hand (Secondary) for a research worker.
Thus, secondary data can be obtained either from
published sources or from any other
source, for example, a website which saves time and
cost.

PRIMARY DATA ~ m.

PUBLISHED

SECONDARY DATA

I
 4

Collection of Primary and Secondary Data Jld 29

How Primary Data is Collected

The most popular and common tool is

questionnaire/interview schedule to collect the
primary data. The questionnaire is managed by the
enumerator, researchers or trained,
investigators. Sometimes the questionnaire is
managed by the respondents also.

Following are the basic principles of drafting

questionnaire:

(1) Covering ·letter : The person conducting the

survey must introduce himself and
make the aims and objectives of the enquiry clear to
the informant. A personal
letter can be enclosed indicating the purposes and
aims of enquiry. The informant
should be taken into confidence. He should be
assured that his answers will be
kept confidential and he will not be solicited after he
fills up the questionnaire.
A self-addressed and stamped envelope should be
enclosed for the convenience
of the informant to return the questionnaire.

(2) Number of questions : The informant should be

made comfortable by asking
minimum number of questions based on the
objectives and scope of enquiry. More
the number of questions, lesser the possibility of
response. Therefore, normally
fifteen to twenty-five questions are sufficient
enough for making the required
enquiry. Lengthy questions should preferably be
divided into parts. Irrelevant
questions should be eliminated. The questionnaire
should not be too long to enable
the respondent to answer quickly, correctly and
clearly.

(3) Personal questions should be avoided: The

informant may not desire to answer
such questions which may disclose his confidential,
private or personal information.
Questions affecting the sentiments of informants
should not be asked.
 5

(4) The questions should be simple and clear: The Are you married?
language of the questions should Are you employed?
be easy to understand. It should not be ambiguous.
It should be simple and clear. Yes/No
The questionaire should start from general Yes/No
questions. 30 .tL.J. Statistics for Economics-XI

(5) The questions should be arranged logically: It A SPECIMEN QUESTIONNAIRE

helps in classification and tabulation
of data. It is not logical to ask a man his income LAKME LIMITED, MUMBAI
before asking him whether he
is employed or not.: There should be a proper Questionnaire
sequence of the questions.
Note : Please tick (-V) mark in the squares which
(6) Instructions to the informations: Definite apply to you.
instructions for filling in the questionnaire
should be given, instructions about when and where I. General. (This information will be kept
to return the questionnaire, confidential).
instructions about units of measurement should be
given. Name................................. Age .

(7) The questions may be divided and subdivided Address.... Sex : Male 0 Female 0
under different heads and subheads
and should be properly numbered for the Marital Status :
convenience of the informant and the
in vestiga tor. Married 0 Unmarried 0 Any other 0

(8) Multiple choice questions: Questions should be Monthly income No. of members in the family
capable of objective answers. They
should be framed in such a way that the answers are Rs. 0-10,000 0 1-3 0
factual or objective. For this
the informant should be able to give the answers Rs. 10,000-20,000 0 4-6 0
simply by using a tick-mark in
the blank space, e.g., Rs. 20,000-30,000 0 Above 6 0

Which of the following languages you use most for Rs. Above 30,000 0
writing? (Put a cross)
II. Which of our cosmetics do you like most?
(i) English D (ii) Hindi D
Lipstick 0 Cream 0 Powder 0 Nail Polish 0
(iii) Punjabi D (iv) Urdu D ill. Do you like our cosmetics because:

(v) Any other D (i) They are reasonably priced? Yes 0 No 0

(9) Simple alternative questions: (Yes/No) As far as (ii) They are easily available? Yes 0 No 0
possible the questions should be
framed in such a way that they are answerable in (iii) They are liked and used by your friends? Yes 0
'Yes' or 'No' or 'Right' or No 0
'Wrong', ««: (iv) You have a fancy for them? Yes 0 No 0
 6

IV. Do you prefer Lakme mainly due to : (12) The questions should be directly related to the
point under enquiry for which the
(i) Perfume Yes 0 No 0 (ii) Material Yes 0 No 0 data is being collected. The question should not use
negatives as starting with
(iii) Colour Yes 0 No 0 (iv) Packing Yes 0 No 0 'wouldn't', 'Don't you'.

V. Do you feel that our cosmetics add to your grace (13) Avoidance of leading questions: As far as
and beauty? Yes 0 No 0 possible leading questions should be
VI. Do you find our product costly? Yes 0 No 0 avoided. Why do you like 'Broke Bond Tea'? Instead
of such simple question, two
VII. Do your friends often ask you about the questions can be framed for enquiry, namely:
cosmetics you have used in parties,
functions, marriages and get-togethers? Yes 0 No 0 (a) Which brand of tea do you take?

VIII. Do you visit our showrooms in various parts of (b) Why do you prefer it?
the city? Yes 0 No 0
(14) Attractive layout: The questionnaire should be
IX. Do you think our company should manufacture a made to look as attractive as
new cosmetics? Yes 0 No 0 possible. Keeping in view the possible answer to the
questions of schedule, sufficient
x. Which product do you think needs improvement : space should be provided.

Lipstick 0 Cream 0 Powder 0 Nail Polish 0 (15) Avoidance of questions of calculations: As far as
possible no question should be
(10) Specific information question : The asked which requires mathematical calculations like
questionnaire should start from general percentage, ratio, etc. It gives
questions to more specific ones. We get specific strain to the informant and he may avoid sending
answers to such type of questions. the questionnaire back.
These questions are simple and direct.
(16) Avoidance of certain questions: Correct answers
In which class do you read? to certain questions cannot be
obtained. Such debatable questions should never be
In which subject you are more interested? included in the
questionnaire.
(11) Open question: It makes the informant free to
give any reply he chooses. Such Example:
questions are difficult to tabulate and increase
labour in statistical work. Such (a) Do you tell a lie?
questions should be minimum in number in the
questionnaire. (b) Do you love your children?
Collection of Primary and Secondary Data .ild 31 (c) Do you beat your wife?

Example .: (17) Cross checks : Some questions should provide

the means of checking inaccuracies
(a) Suggest the measures to solve the problems of in the answers. For example, question on age and
poor students in University of Delhi. date of birth is a cross check.
It helps to decide whether the informant is
(b) How will you solve the wage problem in your answering the questions correctly and
industry? consistently.
 7

(18) Questions on familiar topics: Questions which

require strain should be avoided.

Too much reliance on memories of distant past may

elicit wrong answers.
Informants should be able to answer from their own
memory and knowledge.

(19) Pre-testing of questionnaire : Before taking the

enquiry on a large scale the
questionnaire drafted should be pre-tested with a
small number of a group of
persons.

The following are the methods of pnmary data

collection which are In common
use:

(I) Direct personal interview: According to this

method, data are collected .by the
investigator personally from persons who are the
subject to enquiry. He interviews
personally everyone who is in a position to supply
information he requires. We can use
this method of collection of data when area of
enquiry is limited or when a maximum I COLLECTION OF DATA I
degree of accuracy is needed. Thus, if a teacher
wants .to know about the personality of ____ I ~
students in a class he may interview his students
personally, asking a number of questions, ,---'----, SECONDARY I
uiz., about marks obtained in previous examination,
interest in sports, music, dramatics, ~I1
etc., and collect the desired information. In the same
way one can think of personal Published
enquiry in a case of collection of information Sources
regarding family budget and living conditions
in a group area. The investigator must be skilled, Unpublished
tactful, accurate, pleasing and should Sources

. not be biased. Direct Personal Interview

32 .tld Statistics for Economics-XI Indirect Personal Interview
Telephone Interview
Information from
Correspondents

Mailed Questionnaires
Questionnaires Filled by
Enumerators
 8

Government Publications
Publications of Internal Bodies 3. Personal bias can give wrong results.
Semi-Official Publications
4. Investigators need to be trained and supervised
Report of Committees and Commissions for the job, otherwise results
Private Publications obtained may not be reliable.

(a) Journals and Newspapers 5. This method is lengthy and complex.

(b) Research Institutions
(IT) Indirect personal interview: Direct personal
(c) Professional Trade Bodies. enquiry cannot be used in the case
of the refusal and reluctance of the persons who are
(d) Annual Reports of Joint Stock Companies to be interviewed. Then an indirect
(e) Articles, Market Reviews and Reports personal enquiry can be conducted to get necessary
information from an indirect source.
Merits Persons who have the knowledge of relevant
material or event are interviewed and asked
1. Original data are collected by this method. questions for collecting the data. For example, an
enquiry of addiction to alcoholic drinks,
2. There is uniformity in collection of data. Collection of Primary and Secondary Data 1Lt. 33

3. The required information can be properly direct personal enquiry can be a failure as a person
obtained. regularly taking liquor obviously
may not give answers properly or accurately to
4. There is flexibility in the enquiry as the questions. In this kind of a situation, we
investigator is personally present. can have a choice for indirect personal enquiry, by
interviewing friends, neighbours,
5. Information can be obtained easily from the employers, relatives and liquor sellers, ete. In the
informants by a personal same way, in the case of an accident
interview. on the road, we can find the cause of the accident by
enquiring the persons present at
6. Since the enquiry is intensive and in person, the or near the place of the accident.
results obtained are normally
reliable and accurate. This method is used by committees and commissions
appointed by a government.
(7. Informants' reactions to questions can be
properly studied . They gather facts about different problems by asking
and cross examining different types
. 8. Investigators can use the language of of people who appear as witnesses. Thus the views
communication according to the educational of different persons are obtained for
standard and attitude of the informant. finding out the real position. In this method a
precaution is necessary in selecting the
Limitations informants. An informant should be a person (i) who
is not biased or prejudiced,
1. This method can be used if the field of enquiry is (ii) must know the facts of problem, (iii) must be
small. It cannot be used when capable of answering correctly for
giving true information, and (iv) is not motivated to
field of enquiry is wide. give colour to the fact. The collected
data are required to be given due allowance for the
2. It is costly method and consume more time. conscious and unconscious bias of
 9

the person giving the information and answering (ill) Telephone interview : The investigator asks
questions. To get success of collection questions over landline telephone,
of data by this method, it is necessary that the mobile telephone and even through website. Various
evidence of one person alone is not to researchers, newspapers, television
be relied upon; the opinions of various persons channels, mobile service providers, banks, ete., use
should be obtained to find out the real telephone service to get information
and true picture of the situation. from different people, e.g., exit poll, political or
economical opinions, music or dance
Merits performance opinion, etc. Even sometimes websites
or internet are used for obtaining
. 1. This method covers a wide area of investigation. statistical data. These days online surveys through
Whenever the informant in direct Short Message Service, i.e., SMS has
investigation is reluctant to give information, or become popular.
cannot be contacted, this method
is a good alternative. Merits

2. As the information is obtained from the third 1. Telephone interviews are cheaper than personal
party, it is more or less free from interviews.
biased or prejudiced approach of the investigator
and the informant. 2. It can be conducted in a shorter period of time.

3. It saves labour, time and money. 3. The investigator can assist the respondent by
clarifying the questions.
4. As the information covers a wide range, different 34 .Ild Statistics for Economics-XI
aspects of problems can properly
be studied. 4. Sometimes respondents are reluctant to answer
some questions in personal
Limitations interviews. Telephone interviews are better in such
cases.
1. As the information is obtained from the third-
party and not by the person directly Limitations
concerned, there exists as possibility of not getting
true information. 1. Information cannot be obtained from people who
do not have their own
2. Various evidences obtained may be sometimes telephones.
exaggerated according to the interest
of the person answering the question or supplying 2. Reactions of respondents on certain issues cannot
the information. be judged; but it sometimes
becomes helpful in obtaining information from
Thus, we find that both the above methods-direct respondents.
and indirect personal interviews
have certain plus and minus points. For this reason (IV) Information from correspondents : In this
the choice of the method depends method, local agents or correspondents
on the nature of enquiry and sometimes we balance are appointed in different parts of the investigation
the demerits of one method by using area. These agents regularly supply
the other method also for the same investigation. the information to the central office or investigator.
This way we can counter check the They collect the information according
data collected by one method with the other. to their own judgements and own methods. Radio
and newspaper agencies generally
 10

obtain information about strikes, thefts, accidents, informants are spread over a wide geographical
etc. by this method. It is adopted by area.
Government departments to get estimates of
agricultural crops and the wholesale price 2. This is not an expensive method. It is cheap as
index number. It is suitable when the information is mailing cost. is less than the cost
to be obtained from a wide area of personal visits.
and where a high degree of accuracy is not required.
3. We can obtain original data by this method.
Merits
4. It is free from the bias of the investigator as the
1. This method is comparatively cheap. information IS grven by the
informants themselves.
2. It gives results easily and promptly.
Limitations
3. It can cover a' wide area under investigation.
1. It is difficult to presume the cooperation on the
Limitations part of the informants. They
may fail to send back the schedules or may
1. In this method original data is not obtained. misinterpret or may not understand
some questions.
2. It gives approximate and rough results. Collection of Primary and Secondary Data JL.t 35

3. As the correspondent uses his own judgement, his 2. Schedules sent back by the informants may be
personal bias may affect the incomplete or inaccurate and it
accuracy of the information sent. may be difficult to verify the accuracy.

4. Different attitudes of different correspondents 3. There may be delays in getting replies to the
and agents may increase errors. questionnaires.

(V) Mailed questionnaires : A list of questions 4. This method can be used only when the
relating to inquiry, which is called informants are educated or literate, so
schedule or questionnaire, is prepared. This list of that they return the questionnaires duly read,
questions provides a space for each understood and answered.
answer. Schedules are sent to informants by post,
with a request to answer and return 5. There is a possibility of getting wrong results due
it ~ithin a specified time. Such schedules generally to partial responses, and, those
have prepaid postage stamp affixed who do answer may not include certain type of
to them. If necessary, an assurance is given that the persons from whom the specific
answers will be kept confidential. It information is required.
is the most popular method used these days by
individuals. The success of this method 6. There may be loss of questionnaires in mail.
depends on the cooperation that the informant is
prepared to extend and the manner in This method is suitable for the following situations :
which the questionnaire is drafted.
(a) When it is compulsory by law to fill the
Merits questionnaire, e.g., Government agencies
compel bank and companies, etc., to supply
1. Large field can be studied by this method. We can information regularly to the
use this method in cases where Government in a prescribed form.
 11

(b) This method can be successful when the

informants are educated. 1. It can cover a wide area.

Following are some suggestions for making this 2. The results are not affected by personal bias.
method more effective and
successful. 3. True and reliable answer to difficult questions can
be obtained through
(a) Questions should be simple and easy so that the establishment of personal contact between the
informants may not find it a enumerator and the informant.
burden to answer them.
4. As the information is collected by-trained and
(b) Informants should not be required to spend for experienced enumerators, it is
posting the questionnaires back, reasonably accurate and reliable.
therefore, prepaid postage stamp should be affixed.
5. This method can be adopted in those cases also
(c) This method should be used in a large sample or where the informants are
wide universe. illiterate. '

(d) This method is preferred in such enquiries where 6. Personal presence of investigator assured
it is compulsory by la_w to fill complete response and respondents can
the schedule. Thus, there is little risk of non- be persuaded to give the answers to the
response. questionnaire.
36 ILt Statistics for Economics-XI
(e) The language of the schedule should be polite
and should not hurt the sentiments Limitations
of the informants.
1. It is an expensive method as compared to other
(VI) Questionnaire filled by enumerators : Mailed methods of primary collection
questionnaire method poses a of data, as the enumerators are required to be paid.
number of difficulties in collection of data. Generally,
these filled questionnaires received 2. This method is time consuming since the
are incomplete, inadequate and unrepresentative. enumerator is required to visit people
spread out over a wide area.
The second alternative approach is to send trained
investigators or enumerators to 3. This method needs the supervision of
informants with standardised questionnaires which investigators and enumerators.
are to be filled in by the investigators.
The investigator helps the informants in recording 4. Enumerators need to be trained. Without good
their answers. The investigators should interview and proper trammg,
be honest, tactful and painstaking. This is the most most of the collected information is vague and may
common method used by research lead to wrong conclusions.
organisations. They train investigators properly
specifically for the purpose of an enquiry 5. It needs a good battery of investigators to cover
and also train them in dealing with different persons the wide area of universe and,
tactfully, to get proper answers to therefore, it can be used by bigger organisations.
the questions put to them. The statistical
information collected under this method is Pilot Surveyor Pre-Test : From practical point of view
highly reliable. it is found useful to conduct
a pre-test or a guiding survey known as Pilot Survey,
Merits on a small scale before starting
 12

the main survey. This is done to tryout the weight will be obtained by dividing the total weight
questionnaire and the field methods for of the girls by the number of girls.
obtaining the general information about the The population is the weight of the entire group of
population to be sampled. The information 500 girls. However, we can save the
supplied by the pilot survey helps in :
GOVERNMENT SENIOR SECONDARY SCHOOL
(i) Estimating the cost of the main sample survey and
also the time needed for the UNIVERSE-SOO GIRLS
availability of the results.
SAMPLE-50 GIRLS
(ii) Improving the organisation of the field work by Collection of Primary and Secondary Data JhJ 37
removing the defects or faults
observed in, the pilot survey.

(iii) Formulating effective methods of asking

questions and also in the improvement
of the questionnaire.

(iv) Training of field staff.

time and labour by taking only 50 girls out of 500
(v) Disclosing certain problems or troubles that may
and obtain the average of this part
otherwise be of a serious nature
of the total population. The average of 50 girls
in a large scale main survey.
reasonably be representative of average
weight of 500 girls. In this case weight of 50 girls is
Every individual who is using the statistical technique
the sample.
is required to collect information.
Census Surveys
There are two methods of survey to collect data :
(a) Census method/Census Survey, and
The objective of a census method or complete
enumeration is to collect information
(b) Sample method/Sample Survey.
for each and every unit of the population/universe.
In this method every element of
For discussing the above methods, it is necessary to
population is included in the investigation. Thus,
have a clear understanding of
when we make a complete enumeration
terms "population" and "sample".
of all items in population, it is known as 'Census
Method' or 'Method of Complete
Population and Sample
Enumeration'. In above example, collecting weights
of all the 500 girls in Senior Secondary
In statistics the term 'Population' and 'Universe'
School is census method of collection where no
means the inclusion of all the items
student is left over, as each student is a
in the field of statistical enquiry and sample means
unit.
selection of few items as representatives
of all the items. A part of the whole population is
Following are few examples of census:
called sample and the process of
selection is termed as sampling. Suppose, there are
1. The population census is carried out once in every
500 girls in Government Senior
ten years in India. Most
Secondary School. If we want to know the average
recently population census in India was carried out
weight of those girls, we will weigh
in February, 2001 by house
each girl and wjll get the information about all the
five hundred girls. Then the average
 13

to-house enquiry to cover all households in India. 1951 6.24 29.87 36.11 17.3 82.7
Now, population census is in 1961 7.89 36.02 43.91 18.0 82.0
progress in 201l. 1971 10.89 43.93 54.82 19.9 80.1
1981 16.22 52.11 68.33 23.7 76.3
2. Demographic data obtained by census method on 1991 21.76 62.87 84.63 25.7 74.3
death rates and birth rates, 2001 28.50 74.2 102.7 27.8 72.2
literacy, work force, life expectancy and composition
of population, etc. are
published by Registrar General of India.

3. The data relating to estimation of the total area

under principal crops in India
are obtained by using village records maintained
regularly by Patwari.

Let us review the following census data in the

following Table 2 regarding relative
growth of Urban and Rural Population in India
obtained from Reports and Economic
Survey 2002-2003.

Rural areas in India account for more than 72 per

cent of India's population. In
2001, 74.2 crore persons, out of about 102.7 crore Source: Census Reports and Economic Survey 2002-
total population lived in around 2003.
5.5 lakhs villages while the remaining 28.5 crore Quota Sampling
persons lived in towns and cities or
urban areas. In 1901, .about 2.6 crore persons out of 38 Statistics for Economics-XI
a total population of around
24 crores lived in urban areas. The urban population The a aye table shows the relative growth of rural
formed about 11 per cent and rural and urban population in India since
population 89 per cent of total population. The ratio 1901.
of urban population had gone up
to around 28 per cent in 2001 while still over 72 per The net addition to rural population between 1991-
cent people lived in rural areas. 2001 was 11.33 crore, while
urban population increased by 6.74 crore persons.
TABLE 2: Relative Growth of Urban and Rural The decadal growth rate for rural and
Population in India urban population was 17.9 and 31.2 respectively.
There was a net increase of 2.1 per
Year Urban Rural Total As Percentage of Total cent in the growth rate of urban population in the
Populatio.n decade ending 2001 over the decade
Population Population Population Urban Rural ending 1991.
(in crores) (in crores) (in crores) Population
Population We may study a sample drawn from the large
- population and if that sample is
1901 2.58 21.25 23.83 10.9 89.1 adequaterepresentative of the population, we
1911 2.59 22.62 25.21 10.3 89.7 should be able to arrive at valid conclusions.
1921 2.80 22.32 25.12 11.2 88.8 The results obtained by sample study can be applied
1931 3.35 24.54 27.89 12.1 87.9 to the whole universe or population.
1941 4.41 27.44 31.85 13.8 86.2
 14

Thus, when we make: sample enumeration of some

units of population is called 'Sample Non-Random Sampling
Method' of collecting of data. In above example,
collecting the weights of 50 girls out (b)
of 500 girls in Senior Secondary School is sample
method of collection. In this method Simple or Unrestricted (a)
few students as sample considered for our study.
Random Sampling
Following are a few common examples of sampling :
Restricted Random Sampling (b)
(a) We look at a handful of grain to evaluate the
quality of wheat, rice or pulses, (I) Stratified Sampling

etc. (il) Systematic Sampling (c)

(b) A few items' are tested to find out the lifespan of or Quasi-Random Sampl!ng .
electric bulbs out of each lot.
(c) A drop of blood is tested for diseases like malaria (HI) Cluster Sampling or Multi-stage Sampling
or typhoid, etc.
Judgement Sampling
(d) A few nuts or bolts are tested from the complete
lot of production for final Convenience Sampling
judgement of the quality.

(e) The television network provides election

coverage by exit polls and prediction is
made.

Thus, in our statistical terminology population or

universe does not mean the total
number of people in an area; it means the total
number of observations or items in
particular enquiry. And sample means a small
population of items selected from a universe
for statistical studies.

Broadly speaking, vanous methods of sampling can

be grouped under two main
heads:

(a) Random Sampling, and (b) Non-Random

Sampling.

Let us discuss now the various sampling methods Collection of Primary and Secondary Data .tl.d 39
which are popularly used m
practice. Random Sampling is one where the individual units
(samples) are selected at random.
I METHODS OF SAMPLING I
It is called as probability sampling.
(a)
 15

Random sampling does not mean unsystematic or metal pieces on which numbers 0 to 9 are written.
selection of units. It means the chances The drum is rotated by a mechanical
of each item of the universe being included in the device and each time one piece comes out. The
sample is equal. The term 'Random process is repeated to get the full number
Sampling' here is not used to describe the data in of digits, for instance, if the number is in thousands,
the sample but it refers to the process the drum is to be rotated four time.
used for selecting the sample. Following are the This method is used in drawing pr:,izes for lotteries,
methods of random sampling. prize bonds, etc.

Simple or Unrestricted Random Sampling (b) Table of random numbers: Ifthe population size is
very large the above procedures
This method is also known as simple random will become quite unwieldy and time consuming.
sampling. In this method the selection Furthermore, it may introduce biases,
of item is not determined by the investigator but the if the disks, balls or slips are not thoroughly mixed.
process used to select the terms of Therefore, in recent years, there has
the sample decides the chances of selection. Each been a marked tendency to use tables of random
item of the universe has an equal chance digits for the purpose of drawing such
of being included in the sample. It is free from samples. A table of random digits is simply a table of
discrimination and human judgement. digits which have been generated
Random sampling is the scientific procedure of by a random process. The following tables of random
obtaining a sample from the given digits are available :
population. It depends on the law of probability
which decides the inclusion of items in (a) Tippett's Random Sampling Numbers. There are
a sample. To ensure randomness, mechanical 10400 numbers arranged 4 digits
devices are used. There are two methods a time.
of obtaining the simple random sample. They are :
(b) M.G. Kendall and Babington Smith's Random
(a) Lottery Method, and Sampling Numbers, having 1 lakh

(b) Table of Random Numbers . digits.

., (c) Rand Corporation's a million random digits.

(d) Snedecor's 10000 random numbers.
(a) Lottery Method: A random sample can generally
be selected by this simple and (e) Fisher and Yates Table having 15000 digits.

popular method. All the items of the universe are Tippett's table of random numbers is most popular
numbered and these numbers are which can be used in taking out
written on identical pieces of paper (slip). They are sample. The first thirty sets of numbers out of 10400
mixed in a bowl and then there starts are given below:
the selection by draw one by one by shaking the
bowl before every draw. The numbers
are picked out blind folded. All slips must be
identical in size, shape and colour to avoid
the biased selection.

A special kind of rotating drum is used for finding

random numbers. It is called the
Electronic Random Numbers Indicator Equipment.
The drum contains round wooden
 16

5. Sampling error can be measured.

Demerits

1. This requires complete list of population but up-

to-date lists are not available in
many enqumes.

2. If the size of the sample is small, then it will not be

a representative of a
40 1M Statistics for Economics-XI population.

Tippett Numbers 3. When the distribution between items is very large,

this method cannot be used.
2952 6641 3992 9792 7969 5911
3170 5624 4167 9524 1545 1396 4. The numbering of units and the preparation of the
7203 5356 1300 2693 2370 7483 slips is quite time consuming
3408 2762 3563 1069 5913 7691 and not economical particularly if the population is
0560 5246 1112 9025 6608 8126 large.

Suppose, we want to decide the sample of 15 Restricted Random Sampling

students out of 2000 students in a They are as follows :
college. We will first number all 2000 students from
1 to 2000. After numbering the . (i) Stratified random sampling: In this method the
students, now we will consult a page of Tippett universe is divided into strata or
Table. We can get sample by taking any homogeneous groups and an equal sample is drawn
15 successive number either horizontally or from each stratum or layer at random.
vertically. This method is, therefore, useful when the
population of the universe is not fully
Merits homogeneous. For example, suppose we want to
know how much pocket money an
1. It is more scientific method of taking out samples average university student gets every month will be
from a universe. There are less taken equal sample from various
chances of individual bias. Every item in the universe strata, namely, B.A. students, M.A. students and
has equal chance of being Ph.D. students, etc. Stratified random
selected. sampling is widely used in market research and
opinion polls, it is fairly easy to classify
2. It is more representative. When size of sample people into occupational, economic, social, religious
increases, it is more representative and other strata.
of the population as the Law of Inertia of large
numbers and the Law of Statistical There are different types of stratified sampling:
Regularity begin to operate.
(a) Proportional stratified sampling is one in which
3. This method-is economical as it saves time, money the items are taken from each
and labour in investigating a stratum in the proportion of the units of the stratum
population. to the total population.
(b) Disproportionate stratified sampling is one in
4. The theory of probability is applicable, if the which units in equal numbers are
sample is random. taken from each stratum irrespective of its size.
 17

by preparing this list in some random order, for

(c) Stratified weighted sampling is one where units example, alphabetical order.
are taken in equal number from
each stratum, but weights are given to different The method consists of selecting every nth item
strata on the basis of their from the list, n stands for any number.
'size.
Suppose we have a universe of 10,000 items and we
Collection of Primary and Secondary Data JhJ 41 want a sample of 1000, then we
take n = 10. The method of selecting the first item
Merits from the list is to decide at random
from the first sampling interval, i.e., between one
1. The sample taken under this method is more and ten. Suppose we pick up the 5th
representative of the universe as it item. Then the other items will be 15th, 25th, 35th,
has been taken from different groups of universe. and so on until we have got our
full sample.
2. It ensures greater accuracy as each group
(stratum) is so formed that it consists In order to get reliable results it should be made sure
that the list of the universe is
of uniform or homogeneous items. fully random and that there are no inherent
periodicities in the list.
3. It is easy to administer as universe is sub-divided.
Merits
4. Greater geographical concentration reduces the
time and expenses. 1. It is systematic, very simple, convenient and
checking can also be done quickly.
5. For non-homogeneous population, it is more
reliable. 2. In this method time and work is reduced much.

6. When original population is not normal (skewed), 3. The results are also found to be generally
this method is appropriate. satisfactory.

Demerits Demerits

1. Stratified sampling is not possible unless some 1. Systematic selection mayor may not approach
information concerning the chance or random selection as
population and its strata is available. random will not be a determining factor in the
selection of a sample.
2. If proper stratification is not done the sample will
have an effect of bias. If 2. It is feasible only if the units are systematically
different strata of population overlap such a sample managed.
will not be a representative
one. 3. If the universe is arranged in wrong manner, the
results will be misleading.
(ii) Systematic sampling or quasi-random sampling:
Systematic sampling is a simpler (iii) Cluster sampling or multi-stage sampling: In this
and a more convenient method. This is used when a method sampling is carried
complete list of the population is out in a number of stages. This is done when we
available. This is called a quasi-random method know that for getting reliable results
because a kind of randomness is achieved we have to divide and sub-divide a universe
according to its characteristics. Thus, if a
 18

survey is to be conducted in a country it -will first be guarantee accuracy and ' it makes detecting of
divided into zones or states or sampling errors difficult. However, this
regions, then into smaller units cities, towns and methods is useful in solving a number of kinds of
villages and then into localities and problems in universe and economics.
households. At each stage sampling is done by a
suitable method, say simple random The purposive or judgement sampling is suitable in
sampling. This method of sampling is very helpful in the following conditions :
many large scale surveys where the
preparation of the list of all units in the population is (a) The number of items in the universe is small to
difficult, time consuming and which some items of important
expensive.
characteristics are likely to be left out.
42 1hl Statistics for Economics-XI
(b) When small-sized sample is to be drawn.
Non-random sampling is done on the basis of
convenience and judgement of the (c) When some known characteristics of the universe
investigator and not on the basis of probability, as are to be intensively studied.
against random sampling. Thus, non (d) It is also appropriate for pilot survey.
random sampling is not free; it is determined by
certain considerations. The following Quota Sampling
are some important methods of non-random
sampling: It is a method of sampling that saves time and cost
and is commonly used in surveys
(a) Judgement or purposive sampling of political, religious and social opinion. Interviewers
(b) Quota sampling are allotted definite quotas of the
universe and they are required to interview a certain
(c) Convenience sampling number from their quota. Quotas
are decided on the basis of the proportion of
Judgement Sampling persons in various categories. In other
words, the investigator is given instructions about
This is also called purposive or deliberate sampling. how many interviews should be taken
In this method individual items say in a given locality and what proportion should be
of sampling are selected by the investigator from say upper, middle and lower
consciously using his judgement. Therefore, income groups, as by some other classification which
it requires that the investigator should have a good is predetermined. For example, for
knowledge of the universe and some a study of truancy (running away) from school in
experience in the field of investigation. Obviously, Delhi the investigators are allotted
the choice of samples will vary from quotas of say 10 schools each out of which two
one investigator to another. For example, from a should be public schools (Boys), one
universe of 10,000 ladies who use a public school (Girls), three Boys' Senior Secondary
particular brand of hairdye, the investigator will Schools, two Girls' Senior Secondary
select a sample of say, 1,000. His choice Schools, two Co-education Schools and from each
of this sample will be such that it is irrespective of school he is asked to interview
the universe. For this an exercise of 50 students, taking 10 students each from Classes
judgement is required. VIII, IX, X, XI and XII. The interviewer
can select any 10 students according to his own
In order for the judgement sampling to be reliable, it judgement.
should be free from individual
lies or prejudice. Since the choice of sample is not It is a kind of judgement sampling and provides
based on probability it does not satisfactory results only when
 19

interviewers are carefully trained and personal 4. Minimum errors: In this method it is possible to
prejudice is kept out of the process of check the extent of errors. While
selection. in census a large number of errors may creep in it, is
almost impossible to detect
Convenience Sampling them. The total error, therefore, is much lower in
sampling method.
This method is different from the other two above-
mentioned methods since the 5. Follow up. Because of the smaller size of the data
portion, or 'chunk', or a universe is selected purely and the employment of trained
on the basis of convenience. For investigators in sampling it is possible to cross check
Collection of Primary and Secondary Data lhl 43 the reliability of the data
easily.
example, for the study of truancy (running away)
from school the investigator may select 6. Information in depth : The economy of time and
a school or schools in the neighbourhood because it money allows a sampling method
is convenient for him to go to the to collect more detailed information. For instance, if
only 10 out of 100 persons
chools. This method is used when the universe is not are to be interviewed more questions can be asked
well defined or the sample unit is than if all the 100 are to be
not clear or complete source list is not available. interviewed.
Thus, samples may be obtained from
easily available lists, such as, telephone directories, 7. Indispensability. In certain circumstances the
automobile registrations, etc. Obviously, sample method is the only possible
results obtained by this method are not truly or practicable means of investigation, for instance, in
representatives of the universe and are the case of quality control
unsa tisfactory. of cigarettes or bolts, manufactured at the rate of 1
lakh a day.
Having discussed the various methods of sampling
we will now enumerate the 8. Checking census results. The sample method is
advantages of sampling method in general. generally used to check the accuracy
of data collected by census method. This is possible
1. Economy of time: Time is saved not only in the due to the scientific nature of
collection of data but because sampling method.
of the smaller size, also in analysing the data.
Therefore, sampling is very useful However, certain cautions are required in using the
in getting quick results. sampling method. Careful planning,
proper training of investigators, expert analysis and
2. Economy of cost : Because of less effort involved it above all applications of the method
is much more economical to an appropriate field is necessary for the success of
method than that of census. Since, it is cheaper and sampling method.
possible to conduct more
investigations within the same budget of an The mam purpose of sampling is to collect maximum
organisation or a country. information with rrumrnum
expenditure of money, time and labour and yet
3. Reliability: Since it is a scientific method and achieve' a high degree of accuracy and
trained investigators are employed reliability. For ensuring reliability certain principles
for the collection of data sampling is in some ways must be followed. In sampling method
more reliable than census it is presumed that whatever conclusions are drawn
method. from a sample are also true for the
 20

whole population. This presumption is based mainly over a short period, but the male-female ratio of a
on the following two laws : country as a whole will remain almost
for the period. To take another example, if a coin is
(a) The Law of Statistical Regularity, and tossed 6 times we may get heads,
(b) The Law of Inertia of Large Numbers. four times and tails two times. But if a coin is tossed
44 JhJ Statistics for Economics-XI 5000 times, there is a high probability
of getting heads and tails 2,500 times each. This
(a) Law of statistical regularity: The law of statistical happens due to compensation in the
regularity is derived from the operation of this law. That is, when one part of large
mathematical theory of probability. It says that a group is changing in one direction
comparatively small group of the other moves in the opposite direction.
items chosen at random from a very large group will,
on the average, represent Thus, reliability of sampling depends mainly on
the characteristics of the large group. Basically, it randomness of selection of data and
applied to random selection. the large size of universe, expressed by the above
Thus, so in the process of sampling each unit of the two laws.
universe has an equal chance
of being selected. Therefore, the selected items can Statistical Errors
be said to be representative
of the universe. Although the law is not as accurate There is a great difference in the meaning of mistake
as a scientific law is, it does and error in statistics. Mistake
insure a reasonable degree of accuracy. Since there means a wrong calculation or use of inappropriate
is a certain regularity in natural method in the collection or analysis
phenomena, we assume a certain uniformity in of data. Error means "the difference between the
nature. A random sampling is true value and the estimated value."
said to follow the law of statistical regularity because In other words, the difference between the
of this basic uniformity in approximated (estimated value) and the actual
a universe. value (true value) is called statistical error in a
technical sense. For example, we make
(b) Law of inertia of large numbers: This law is also an estimation that in a particular meeting, 1,000
called the law of stability of persons are there. But we count persons,
mass data. It is based on the law of statistical it may be wrongly counted as 1,030. There is a
regularity. Basically, it states that difference of 30 between the estimated
if the numbers involved are very large, the change in value and counted value. This difference is called
a sample is likely to be 'error' in statistics. But when we make
very small, in other words, the individual units of a wrong calculation, following wrong method, draw
universe very continually but wrong conclusions, etc. They are
the total universe changes slowly. That is, large known as 'mistake'. For example, there is a meeting,
aggregates are most stable than we sent a person to count the
small ones. Because of the slow change in the nature audience, he counts the number of persons as 600,
of total universe this law but actually, there are 590 persons.
is called the law of inertia (laziness) of large This is called 'mistake' in counting.
numbers.
Sources of Errors
For example, sugar production of factory will vary
significantly from year to year Following errors are likely to occur in collection of
but the sugar production of a country as a whole will data
remain comparatively stable. Or
a great change may take place in the male-female 1. Errors of origin: Errors arise on account of
ratio of family may appreciably change inappropriate definiti·ons of statistical
 21

unit, scale, or defective questionnaire, etc. For U'-U

example, wrong scale to measure U
height of students, measured height to nearest of
inch or approximated. The Ue = Absolute error
differences may also occur due to differences in
measuring tapes due to e = Relative error
manufacturing defect. In Physics or Chemistry such V' = Actual value
errors of measurements will
occur while taking readings on various instruments. V = Approximate value
Collection of Primary and Secondary Data thL 4S
lllustration. Sales of commodity approximated Rs.
2. Errors of inadequacy : Errors arise on account of 497 and actual sale Rs. 500.
incomplete data, inadequate :. Absolute error (Ve) = 500 - 497 = 3
number of items in the sample, non-response of
respondent, incomplete answers e=
in questionnaire, misinterpretation of questions in
questionnaire, careless or Here,
unqualified investigators, etc.
. ' 500 - 497 3
3. Errors of calculation: Such errors arise because of
wrong arithmetic calculations and Relative error (e) = = -- = 0.006
due to clerical errors, arithmetic slips, etc. by
omitting some figure considering 500 500
wrong value, making wrong totals, etc. by
respondent or investigator. Relative error (e) can also be represented in
percentage
4. Errors of interpretation: Such errors are
committed by statisticians in misinterpreting = ~ x 100 = 0.6%.
the data.
500
Types of Errors

(a) Absolute and relative 'errors: Absolute error is

the difference between the actual
true value and estimated approximate value while
relative error is the ratio of absolute
error to the approximated value.

Absolute error = Actual value - Estimated value

Symbolically,

Ue = V' - V

Actual value - Estimated value

Relative error =

Estimated value

Symbolically,
 22

of data, are called non-sampling errors. They are

common both in census enumeration
and sample surveys. ,
46 ill Statistics for Economics-XI

To avoid these errors, the statistician must take

proper precaution and care in using
the correct measuring instrument. He must see that
the enumerators are also not biased.
Unbiased errors can be removed with proper
planning of statistical investigations.
Statisticians should have none of these errors.

Secondary data are those which are collected by

some other agency and are used for
further studies. It is not necessary to conduct special
surveys and investigations. We can
obtain the required statistical information from
other institutions, or reports which are
already published by them as a part of their routine
work. It saves cost and time which
Relative error is generally used in statistical are involved in collection of primary data. Secondary
calculations because absolute error gives data may be either (a) published
wrong or misleading calculations. or (b) unpublished.
(b) Biased and unbiased errors: Biased errors arise Published Sources
due to some prejudice or bias in
the mind of investigator or the informant or any The various sources of published data are as under :
measurement instrument. Suppose the
enumerator used the deliberate sampling method in (i) Government publications : Different ministries
place of simple random sampling and departments of Central and
method; then it is called biased error. These errors State Governments publish regularly current
are cumulative in nature and increase information along with statistical
when the sample size also increases. Biased errors data on a number of subjects. This information is
arise due to faulty process of selection, quite reliable for related studies.
faulty work during the collection of information and The examples of such publications are : Annual
faulty method of analysis. Survey of Industries, Labour
Gazette, Agriculture Statistics of India, Indian Trade
Unbiased errors are not the result of any prejudice Journal, etc.
or bias. They are those which
arise acccidently just on account of chance in the (ii) Publications ofinternational organisations : We
normal course of investigation. Unbiased can obtain valuable international
errors are generally compensating. statistics from official publication of different
international organisations, like,
(c) Sampling and non-sampling errors: The errors the United Nations Organisation (UNO),
arising on account of drawing International Labour Organisation (ILO),
inferences about the population on the basis of few International Monetary Fund (IMF), World Bank, etc.
observations (sampling) are called
sampling errors. The errors mainly arising at the (iii) Semi-official publications: Local bodies such as
stages of ascertainment and processing Municipal Corporations, District
 23

Boards, etc. publish periodical reports which give

factual information about health, (d) Annual reports of joint stock companies are also
sanitation, births, deaths, etc. useful for obtaining statistical
information. These are published by companies
(iv) Reports of committees and commissions : every year.
Various Committees and Commissions
are appointed by the Central and State Governments (e) Articles, market review and reports also provide
for some special study and valuable data for research
recommendations. The reports of such committees study.
and commissions contain
valuable data. Some of the reports are : Report of Unpublished Data
National Agricultural
Commission, Report of the Tariff Commission, the Research institutions, trade associations,
Patel Committee Report, universities, labour bureaus, research workers
etc. and scholars do collect data but they normally do
not publish it. Apart from the above
(v) Private publications sources we can get the information from records
and files of government and private
(a) Journal and newspapers. Journals like Eastern offices.
Economists, Journal of Industry
and Trade, Monthly Statistics of Trade; and Limitations of Secondary Data
newspapers, like Financial Express,
Economic Times, collect and regularly publish the One should use the secondary data with care and full
data on different fields of precaution and should not
economics, commerce and trade. accept them at their face value as they may be
suffering from the following
(b) Research institutions. There are a number of limitations:
institutions doing research on
allied subjects. This is the most important source of 1. They may not have been collected by proper
obtaining secondary data. procedure.
The National Council of Applied Economic Research
and Foundation of 2. They may not be suitable for a required purpose.
Scientific and Economic Research are such The information which was
institutions. Research scholars at collected on a particular base may not be suitable
the university level also contribute significantly to and relevant to an enquiry.
the availabilities of secondary
data. 3. They may have been influenced by the biased
investigation or personal
(c) Professional trade bodies. Chambers of
Commerce and Trade. Associations prejudices.
publish statistics relating to trade and commerce.
Federation of Indian 4. They may be out of date and not suitable to the
Chamber of Commerce, Institute of Chartered present period.
Accountants, Sugar Mills
Collection of Primary and Secondary Data ill 47 5. They may not satisfy a reasonable standard of
accuracy.
Association, Bombay Mill Owners Association, Stock
Exchanges, Bank and 6. They may not cover the full period of
Cooperative Societies, Trade Unions, etc. publish investigation.
statistical data.
 24

Precautions in the Use of Secondary Data (E) Price Statistics (F) Industrial Statistics

The investigator should consider the following points (G) Trade Statistics (H) Financial Statistics
before usmg the secondary
(I) Labour Statistics
data:
There are some agencies both at the national and
(a) Are the data reliable? state level, which collect, process
and tabulate statistical data. Some important major
(b) Are the data suitable for the purpose of agencies at the national level are
investigation? Census of India, National Sample Survey
(c) Are the data adequate? Organisation (NSSO), Labour Bureau, Central
Statistical Organisation (CSO), Registrar General of
(d) Are. the data collected by proper method? India (RGI), Director General of
(e) From which source were the data collected? Commercial Intelligence and Statistics (DGCIS), etc.
(f) Who has collected the data?
India had the unique experience of undertaking the
(g) Are the data biased? biggest census in the world in
1981, and has also an unbroken record of more than
Thus, the secondary data should not be used at its hundred years of decadal censuses.
face value. It is risky to use such The Indian census is. universally acknowledged as
statistics collected by others unless they have been most authentic and comprehensive
properly scrutinised and found reLiabLe, source of information about our land and people. In
suitabLe and adequate. 1869 Hunter was appointed Director
General of Statistical Surveys. He not only elaborated
There are various sources and organisations through the statistical system but also
which statistical data are being assisted the statistical surveys of districts and
compiled in India. Since India achieved provinces. That later followed into famous
Independence, great and rapid strides have been Gazetteers. He advised in conducting of census of
made in the field of collection of data. In the context India which undertook explanatory
of economic planning, importance surveys from 1869 to 1872 and thereafter matured
of statistics (data) in the country has become great. into a decennial census which ever
Statistics are necessary for framing since continued without interruption. After 1872 the
48 LhL Statistics for Economics-XI next census was taken in 1881 and
since then it has become a regular feature of holding
and judging the progress of economic planning. The census every ten years uninterruptedly.
study of Indian statistics IS made The Census of India provides the most complete and
under following heads : continuous demographic record of
population. The first census after Independecne was
I. Statistical Organisation of India (CSO) held in 1951 and latest one completed
in 2001. The study of population is important for
II. Indian Statistical Material. several reasons in overall study of
economic development. Information of demographic
This can be studied under following sections characteristics include birth and
death, fertility, sex ratio, age-composition, migration
(A) Agriculture Statistics (B) ational Income and and literacy, etc. The economic
Social Accounting characteristics of population are manifested through
workers' participation in economic
(C) Population Statistics (D) National Sample Survey activities, its distribution and classification of
workers in various occupations, employment
 25

and unemployment. and is headed by an eminent economist/statistician

and the -mernber-secretary of the
The data generated by the Census of India 2001 council is Director General and Chief Executive
provide benchmark statistics on the Officer of NSSO. The Governing Council
people of India at the beginning of the next is empowered to take all technical decisions in
millennium. This is a mirror of a fair respect of survey work, from planning of
representation of the socio-economic and survey to release of survey results. The NSSO headed
demographic condition of our people which by a Director General and Chief
constitute about one-sixth of the human population Executive Officer, has four divisions, namely, Survey
on this planet. The census statistics Design and Research Division (SDRD),
is useful for assessing the impact of the Field Operation Division (FOD), Data Processing
developmental programmes and identify new Division (DPD) and Coordination
thrust areas for focussing the efforts on improving Publication Division (CPD). A Deputy Director
the quality of life in our country. General heads each division except FOD.
Basic population data from Primary Census Abstract. An Additional Director General heads FOD.
Census of India 2001 gives
Functions of NSSO
. information of population in India as
The functions of National Sample Survey
TABLE 3 Organisation are:

Persons Males Females Sex Ratio (i) Collection of data on socio-economic conditions,
1,028,610,328 532,156,772 496,453,536 933 production of small scale house
hold enterprises consumption, etc. on continuous
basis in a comprehensive manner
for whole country. A major objective of NSS has
been to provide data required
to fill up the gaps in information needed for
estimation of national income.
Collection of Primary and Secondary Data Iht. 49
(ii) Collection of data relating to the organised
The National Sample Survey (NSS), initiated in the industrial sector of the country.
year 1950, is a nationwide, large
scale continuous survey operation conducted in the (iii) Supervision of surveys conducted by states in
form of successive rounds. It was agricultural sector through their
established on the basis of a proposal from Prof. P.c. own agencies and also giving guidance to them for
Mahalanobis to fill up data gap analysing and coordinating
for socio-economic planning and policy making the results of these surveys.
through sample surveys. On March 1970,
the NSS was recognised and all aspects of its work The NSSO took a forward view of the data
were brought under a single requirements to planners, research workers
Government organisation, namely, the National and other users and draw up a long-term
Sample Survey Organisation (NSSO) programme. The programme conducts periodical
under the overall direction of a Governing Council to surveys on :
impart objectivity and autonomy
in the matter of collection, processing and (a) Demography, health and family planning;
publication of the NSS data. (b) Assets, debt and investment;

The Governing Council consists of 18 experts from (c) Land holdings and livestock enterprises;
within and outside Government
 26

(d) Employment and unemployment, rural labour • Importance source of secondary data-(a) Census of
and consumer expenditure; and India, (b) National Sample Survey
(e) Self-employment in non-agricultural enterprises. Organisation (NSSO).

The data collected by NSSO surveys on different I. Long Answer Questions

socio-economic subjects are released
through reports and its quarterly journal Q.1. What do you understand by Statistical Enquiry?
'Sarvekshana'. The data comprises diffe"rent Explain.
socio-economic subjects like employment,
unemployment literacy, maternity child care, Q.2. Distinguish between primary and secondary
utilisation of public distribution system, utilisation of data. Illustrate your answer with
educational of services, etc. The examples.
NSS 59th round survey (Jan.-Dec. 2003) was on land
and livestock holdings, debts and Q.3. Enumerate the various methods of collecting
investment. The NSS 60th round survey (Jan-June primary data. Discuss the merits and
2004) was on morbidity and heath
care. Apart from collection of rural and urban retail demerits of anyone of them. .
prices for compilation of consumer
price index numbers NSSO also undertakes QA. Discuss the comparative merits of various
'fieldwork of Annual Survey of Industries methods of collecting primary data.
an~ conducts cr~p estimation surveys. Q.5. Distinguish between 'direct personal
HINT FOR REVISION investigation' and 'indirect oral investigations'.

50 J1t Statistics for Economics-XI Give suitable examples.

• Source of data. Internal source and External Q.6. What are the similarities and dissimilarities
source. External source can be primary or secondary between the two methods-question
source. naires to be filled in by informants and schedules to
be filled in by enumerators?
• Primary and secondary data. Primary data is Explain with examples.
original and first-hand information which can be
obtained by direct or indirect interviews or Q.7. What is a questionnaire? Give a specimen of a
telephone interviews, or by questionnaires and questionnaire.
shedules. Secondary data is second-hand
information collected through other sources, like Q.8. Describe the questionnaire method of collecting
published sources or unpublished sources, viz., primary data. What precautions
Government departments and companies must be taken while preparing questionnaire?
records, newspapers, journals, etc.
Q.9. Write short notes on :
• Census and sample. Census means the inclusion of (a) Census of India
all the items in the field of statistical
enquiry and sample means selection of few items as (b) National Sample Survey Organisation (NSSO)
representatives of all the items.
Q.l0. What is secondary data? Discuss the various
• Methods of Random Sampling-(a) Simple or sources of collecting secondary
Unrestricted, (b) Restricted Random Sampling, data.
viz., stratified sampling, systematic or Quasi-random
sampling and cluster sampling or multi Q.l1. What precaution should be taken before using
stage sampling. seco-ndary data? Explain.
 27

Q.12. Point out the major problems in constructing .20. 'Does the lottery method always give you
interview schedules and question random sample? Explain .
naires.
.11. Do samples provide better results than surveys?
Q.13. Frame at least four appropriate multiple Give reasons for your answer .
choice options for following questions: . 22. Examine the important types of sampling
(i) ~ow often do you use computers? methods .

(ii) What is the monthly income of your family? .13. Distinguish between random sampling and
(iii) Rise in petrol price is justified. systematic sampling. Give suitable
examples .
----------- -- - -------. ---- ---------,
.14. Discuss briefly the following :
(iv) Which of the newspaper do you read regularly?
(a) Law of inertia and large numbers
(v) Which of the following most important when you (b) Law of statistical regularity
buy a new dress?
.25. What do you understand by 'Census'
Q.14. Frame two-way questions (with 'Yes' or 'No'). investigation? Explain its suitability with
Q.15. Following statements are true or false. illustrations .

(i) Data collected by investigator is called secondary . 26. What do you mean by 'Sample' investigation?
data. Explain its suitability with
(ii) There are many sources of data.
illustrations .
I
. 27. Explain briefly the different methods of
(iii) Telephone survey is the most suitable method of sampling. Give illustrations .
collection of
. 28. Give a comparative study of stratified sampling
data when the population is literate and spread over and multi-stage sampling .
a large area . . 29. Write a critical note of random sampling
. 16. Distinguish between population and sample . method .

. 17. Distinguish between census and sample .30. How would you distinguish convenience
surveys. List four important types of sampling with judgement (deliberate)
sampling methods; Explain the reasons for preferring sampling? Explain .
sample surveys in the .31. What do you mean by statistical errors?
collection of data . .32. Discuss briefly the following :

. 18. Name the methods of selecting a sample. (a) Biased and unbiased errors
Describe the method of stratified sampling
with merits and demerits . (b) Absolute errors and relative errors
(c) Sampling and non-sampling errors
. 19. The Education Ministry is interested in
determining the level of education of .33. Give two examples each of sample, population
unmarried girls in the country. How would you and variable .
organise a survey for this .34. Which of the following methods gives better
purpose? result and why?
(a) Census
 28

Q.6. Name three published sources of secondary

(b) Sample data.
Ans. (a) Government Publications,
D
D (b) Private Publications,
D
(c) Reports of committees and commissions.

II. Short Answer Questions Q. 7. Distinguish between population and sample.

Q.1. What is statistical enquiry? Ans. Population or Universe means the inclusion of
all the items in the field of statistical
Ans. Statistical enquiry means st-atistical equiry and sample means selection of few items as
investigation or statistical survey, a method of representatives of all the
collecting information from individuals. items.

Q.2. How primary data is collected? Q.8. Name two methods of obtaining the simple
random sample.
Ans. The most popular and common method is Ans. (a) Lottery method,
questionnaire/interview schedule to
collect the primary data. The questionnaire is (b) Tables of random numbers like Tippett's random
managed by the enumerator, numbers, Fisher and Yates
researchers, or trained investigators. tables.

Q.3. What is basic difference between primary and Q.9. What is stratified random sampling?
secondary data.
Ans. In this method the universe is divided into
Ans. Primary data is original and first-hand strata or homogeneous groups and
information collected originally while equal sample is drawn from each stratum or layer at
secondary data is collected through other sources random.
like published reports, website,
Government or company departments, etc. Q.I0. What is systematic or quast-random sampling?

QA. Write any five principles of drafting Ans. This is used when a complete list of. the
questionnaire. population is available in order, e.g.,
alphabetical order. The method consists of selecting
Ans. Principles: (a) covering letter, (b) question every nth item from the list,
should be simple and clear, (c) multiple viz., 15th, 25th or 35th and so on.
choice questions, (d) minimum number of questions,
(e) avoidance of questions Q.11. What is judgement sampling?
of calculations.
Ans. It is non-random sampling which is done on the
Q.5. Name any three methods of primary collection basis of convenience and judge
of data. ment of the investigator. It is also called purposive or
Ans. (a) Direct personal interview, deliberate sampling where
individual items of sampling are selected by the
(b) Indirect personal interview, investigator consciously using his
judgement. It requires that the investigator should
(c) Information from correspondents. have a good knowledge of the
 29

universe and some experience in the field of

investigation. (c) SaI?pling and non-sampling errors .

Q.12. What do you mean by quota sampling? . 16. Which are the important sources of secondary
data in India?
Ans. In this method, interviewers are allotted
definite quotas of the universe and they Ans. (a) Census of India,
are required to interview a certain number from
their quota. Quotas are decided· '(b) National Sample Survey Organisation (NSSO).

- e basis of the proportion of persons in various

categories .• viz -» llppe..r; rniddle
and {ower income groups .

. What is convenience sampling?

. In this method the portion, or chunk, or a universe

is selected purely on the basis
of convenience. It is used when universe is not well
defined or sample unit is not
clear or complete list is not available. The samples
may be drawn from easily
available lists, such as, telephone directories,
automobiles registrations, etc. The
results obtained by this method are not truly
representatives of the universe and
are unsatisfactory .

. 14. What is the difference in the meaning of

mistake and error in statistics? What are
the sources of errors?

Ans. Mistake means a wrong calculation or use of

inappropriate method in the collection
or analysis of data .. Error means the difference
between the true value and the
estimated value.

Source of errors: Errors in origin, errors of

inadequacy, errors of calculation and
errors of interpretation .

. 15. Give names of types of errors.

Ans. (a) Absolute and relative errors,

(b) Biased and unbiased errors,