Data Collection:

Collection of data is the first and most important statistical investigation. The process of
collection necessary information from the field of inquiry by the investigators themselves
or by their agents is called data collection

Primary Data– Data originally collected in the process of investigation are known as primary data.
This is original form of data which are collected for the first time.It is collected directly from its
source of origin.

Secondary data It refers to collection of data by some agency, which already collected the data and
processed. The data thus collected is called secondary data.

Differences Between Primary and Secondary Data:

BASIS FOR
COMPARISON PRIMARY DATA SECONDARY DATA

Primary data refer to the first Secondary data means data

hand data gathered by the collected by someone else
Meaning researcher himself. earlier.

Data Real time data Past data

Process Very involved Quick and easy

Surveys, observations, Government publications,

experiments, questionnaire, websites, books, journal
Source personal interview, etc. articles, internal records etc.

Cost
effectiveness Expensive Economical

Collection time Long Short

Always specific to the May or may not be specific to

Specific researcher’s needs. the researcher’s need.

Accuracy and
Reliability More Relatively less
Methods of collecting primary data
There are three basic ways of collecting data :
(i) Personal interview OR Direct Personal Investigation
(ii) Mailing (questionnaire surveys)
(iii) Telephone interviews
(iv) Indirect verbal investigation
(v) Information from local sources
(vi) Enumerator method

1. Direct Personal Interview:

In this method, the investigator or his agent collects the data by asking the questions
individually. There is face to face contact between the respondence and the interviewer.
This method is suitable:

a) When the area of inquiry is limited.

b) When the data is to be kept secret.
Merits:
a) The data obtained by this method are pure and original in character.
b) The investigator can clarify any doubts during the interview.
Demerits:
a) This method takes a lot of time so, it is expensive method.
b) There is chance of getting biased information.

2. Indirect Personal Interview:

In this method, the investigator does not directly ask the questions to the persons who are
directly concerned with the problem. The investigator obtains the necessary information
by contacting the third person who is familiar with the problem. The third person is said
witness. This method is suitable:

a) When the direct sources of information are not available.

b) When the informants are unwilling to give the correct information.
Merits:
a) It can cover a wider area very easily.
b) This method is cheaper, requires less time, money and labour
Demerits:
a) The witness may be biased to provide accurate information.
b) The selected informants may not be appropriate people to answer the questions.

3. Information from Corresponds.

In this method, the investigator appoints local agents or correspondence in different parts
or field of enquiry to collect information. These corresponds collected and send to the
central office. This method is suitable for:
a) Newspaper agencies, televisions agency for regular information.
Merits:
a) The information through this method is accurate.
b) This method is cheaper and less time consuming.

Demerits:
a) Information sent by agents may not be reliable.
b) It is difficult to check the accuracy of the data.

4. Mail Questionnaire.
In this method, the investigator prepares a questionnaire (a set of questions). The
questionnaire is sent to the various informants by post of mail. They are requested to
answer the questions and post back to the indicator. This method is suitable:

a) When the field of investigation is large.

b) When the investigator requires quick result at low cost.
Merits:
a) This method is cheaper, saves time and labour.
b) The information can be collected from very far areas.
Demerits:
a) There is high degree of non-response error.
b) Informants may feel fear to provide information.

5. Schedule sent through Enumerator.

In this method, the enumerators go to the informants personally with the schedule (a list of
questions) and fill off the list of questions in their own handwriting by interviewing the
informants. Population census in Nepal is conducted by this method. This method is
suitable:

a) When the informants are illiterate

Merits:
a) The results obtained by this method are reliable and accurate.
b) Some additional information can also be obtained.
Demerits:
a) This method is more expensive.
b) The enumerators should be well trained, skilled and tactful. Otherwise, data may not be
correct.

Important points to be kept in mind while drafting the questionnaire

A. Introduction and purpose of investigation
B. Reasonable number questions.
C. Questions should be small & clear.
D. Questions should be arranged logically.
E. Instructions should be clear.
F. Proper space for answer.
G. Questions should be relevant to the investigation.
H. Personal questions should be avoided.
I. Avoid questions of calculations.
J. Cross Verification.

.
Problems in Primary Data Collection:

a. The informants may be uneducated and uncooperative. Thus, the relevant and
accurate information may not be collected.
b. If there is no transportation facility, the investigator may take the information
without interviewing the informants.
c. The real facts of the data are not available due to the different conclusions given by
the different informants about the same topic.
d. The limitation of time and money is the other problem involved in primary data
collection.
e. Only few questionnaires are send back to the investigators by informants while
using mail questionnaire method. So, accurate data can’t be expected.
f. Due to the shame, hesitate, fear of respondence, the investigator may not be
collected reliable information.

Sources of Secondary Data:

Secondary data can be obtained from two sources. They are:

1. Published sources.
Published can be categorized in to following chart:
2. Unpublished sources:
All information may not be published but they may be suitable for the purpose of
investigation. The sources of unpublished data are:

a) Reports of private office.

b) Hospital records.
c) Records of VDCs.
d) Records of schools and campus administration
e) Thesis, field reports etc. of University students.

Problems in Secondary Data Collection:

1. Necessary information may not be published during the period of investigation.

2. The data obtained from secondary sources may not be suitable for the further
research study.
3. The data obtained from secondary source may not be adequate for the purpose of
finding the solution of research problems.
4. The data obtained from secondary sources may not be reliable due to the limitations
of primary data.

Pilot Survey: Before sending the questionnaire to the information. It should be pretested. As a result
of its short comings if any, can be removed. Such pretesting named as pilot survey

A pilot survey is a preliminary survey used to gather information prior to conducting a survey on a
larger scale. Pilot surveys, typically taken by smaller groups, help determine the efficiency of the
future survey while also helping organizations smooth out difficulties before administering the main
survey.

Methods of sampling:

1. Random sampling

a. Simple or unrestricted random sampling

b. Restricted random sampling

i. Stratified

ii. systematic

iii. multistage or cluster sampling

2. Non-Random Sampling

a. Judgment sampling

b. Quota sampling

c. Convenience sampling
Census survey : In this method every element of population is included in the investigation.

Sample survey : In this method a group of units representing all the units of the population is investigated.
BASIS FOR
COMPARISON CENSUS SAMPLING

A systematic method that collects and Sampling refers to a portion of the

records the data about the members of population selected to represent the
Meaning the population is called Census. entire group, in all its characteristics.

Enumeration Complete Partial

Only a handful of units of the

Study of Each and every unit of the population population.

Time required It is a time consuming process. It is a fast process.

Cost Expensive method Economical method

Less reliable and accurate, due to the

Results Reliable and accurate margin of error in the data collected.

Error Not present. Depends on the size of the population

Appropriate for Population of heterogeneous nature. Population of homogeneous nature.

Population or universe
In Statistics, population or universe simply refers to an aggregate of items to be studied for an
investigation.
Sample: A group of items taken from the population for investigation and representative of all
the items.
Sampling Errors: Sampling error is the difference between the result of studying a sample and
the result of the census of the whole population.
Non-Sampling Error: Can occur in any type of survey whether it be a census or sample
survey.
Sampling errors
1. Biased errors
2. Unbiased errors
Non sampling errors
1. Error in data acquisition
2. Non Response error

3. Sampling Bias

Census of India and National Sample Survey Organization:

The census of India provides the complete and continuous demographic record of population.
The NSSO was established by the Govt. of India to conduct nation wide survey on socio-
economic issues like employment, literacy, maternity, child care, utilisation of public distribution
system etc.

The data-collected by NSSO survey are released through reports and its quarterly journal
‘’Sarvekshana’’.

Eg. Size, growth rate, distribution of population, density, population, projections, sex
composition and literacy.