COLLECTION OF DATA

BY OWNYOUR GURUKUL
STATISTICAL ENQUIRY
An investigation on a topic wherein relevant quantitative information
is collected.
The person who conducts the enquiry is known as investigator.
The investigator needs help of certain persons to collect the
information, known as enumerators.
The persons, from whom statistical information is collected, are
known as respondents.
 SOURCES OF
COLLECTION
OF DATA

EXTERNAL INTERNAL
SOURCES SOURCES

PRIMARY SECONDARY
DATA DATA
SOURCES OF DATA
INTERNAL SOURCE EXTERNAL SOURCE
When the data is collected from Data collected from outside the
within the organization it is called organization is called external
internal source of data. source of data.
It can be collected from primary
as well as secondary sources.
 EXTERNAL SOURCE
PRIMARY DATA
Data collected by investigator for his own purpose, for the first time, from
beginning to end, are called primary data.
It is based on first hand information.
It is collected from primary source of origin.
EXAMPLE – If you want to know the popular singer among the college students,
you have to enquire a large number of students by asking questions to know
their choice to collect the desired information.
EXTERNAL SOURCE
SECONDARY DATA
The data which is not directly collected but rather obtained from the
published or unpublished sources, is known as secondary data.
This is the data which is collected by someone and used by others.
EXAMPLE – After collecting the data about the popularity of the
singer among college students , you publish a report. If somebody
uses the data collected by you for a similar study, it becomes
secondary data.
 DIFFERENCE BETWEEN
PRIMARY AND
SCEONDARY DATA
Basis Primary data Secondary data

Originality Original as collected by the Not original as investigator makes

investigator himself. use of the data collected by other
agencies.
Source Collected from the source of origin. Collected from secondary source.
Time factor Requires longer time for data Requires less time.
collection.
Cost factor Costlier in terms of time, money and Comparatively less costlier.
efforts involved.
 DIFFERENCE BETWEEN
PRIMARY AND
SCEONDARY DATA
Basis Primary data Secondary data

Reliability and More reliable and suitable as Less reliable and suitable as
suitability collected by investigator himself. collected by someone else.
Precautions No great need for precautions. Should be used with great care and
caution.
Adjustment No adjustment needed. Need to be adjusted to suit the
objective of study.
Organization Requires elaborate organization set No need of organizational set up.
factor up.
METHODS OF COLLECTING
PRIMARY DATA
DIRECT ORAL INDIRECT ORAL
INVESTIGATION INVESTIGATION

INFORMATION FROM TELEPHONIC

CORRESPONDENTS INTERVIEWS

MAILED QUESTIONNAIRE QUESTIONNAIRE FILLED

METHOD BY ENUMERATORS
DIRECT PERSONAL
INVESTIGATION
MEANING – Data is personally collected by the investigator from
informants. Under this method investigator conducts on spot enquiry.
SUITABILITY –
Detailed information required.
Area under investigation is limited.
Confidentiality of information required.
Maximum degree of accuracy needed.
Data needs to be original.
DIRECT PERSONAL
INVESTIGATION
MERITS DEMERITS
Accuracy Not suitable for wide areas
Originality Expensive and time consuming
Flexibility Personal prejudice
Uniformity Complex
Economical Trained personal
Suitable for all type of questions
INDIRECT ORAL
INVESTIGATION
MEANING – Investigator approaches the third parties, who are in possession
about the subject matter of enquiry.
People from whom information is collected are known as witnesses.
SUITABILITY –
Field of enquiry is large
Direct contact with concerned informants cannot be established.
When the problem of investigation is complex and needs expert’s opinion.
When secret or sensitive information about the informant has to be
gathered.
INDIRECT ORAL
INVESTIGATION
MERITS DEMERITS
Wide coverage Indirect information
Economical Lack of accuracy
Free from Bias Partiality of witness
Expert opinion Lack of interest of informants
Lack of uniformity
INFORMATION FROM
CORRESPONDENTS
MEANING – under this method, local agents or correspondents are
appointed and trained to collect the information from
correspondents.
SUITABILITY –
Regular and continuous information is required.
Area of investigation is very large.
High degree of accuracy not required.
INFORMATION FROM
CORRESPONDENTS
MERITS DEMERITS
Wide coverage Lack of uniformity
Economy Danger of partiality
Suitable for special purpose Lack of accuracy of high degree
Continuity Time consuming
TELEPHONIC INTERVIEWS
MEANING – data is collected through an interview over the
telephone with the interviewer.
SUITABILITY –
Respondents have a telephone connection
Data is to be collected in shorter duration
TELEPHONIC INTERVIEWS
MERITS DEMERITS
Wide coverage Limited use
Economical Visual feedback is not possible
Clarify doubts
MAILED QUESTIONNAIRE
METHOD
MEANING – Questionnaire prepared according to the objective of
study is mailed to the informants, with the request to answer and
return them within a specified time period.
Method is adopted by research workers and other official and non-
official agencies.
SUITABILITY –
Field of investigation is very large.
Informants are literate and are co-operative.
MAILED QUESTIONNAIRE
METHOD
MERITS DEMERITS
Economical Limited scope
Wide coverage Lack of accuracy
Original and reliable Lack of interest
Free from bias Time consuming
Maintains secrecy Inflexible
QUESTIONNAIRE FILLED BY
ENUMERATORS
MAENING – under this method, the enumerator personally visits the
informants along with a questionnaire, asks questions and note down
their replies in the questionnaire in his own language.
This method is used by the government, semi-government
organizations, big business houses research institutes, etc.
Questionnaire filled by the investigator is known as schedule
QUESTIONNAIRE FILLED BY
ENUMERATORS
SUITABILITY –
Adequate finance and trained enumerators.
Area under investigation is large.
Informants are not educated enough.
 QUESTIONNAIRE FILLED BY
ENUMERATORS
MERITS DEMERITS
Wide coverage Costly
Accurate and reliable information Time consuming
Personal contact Biased
Better response Inability of enumerator
Limited chance of partiality Unsuitable for small investigations
Useful in case of illiterate
respondents
CONSTRUC
TION OF
QUESTIONN
AIRE OR
SCHEDULE
PRINCIPLES TO BE FOLLOWED WHILE DRAFTING THE
QUESTIONNAIRE
Covering letter
Decision regarding questions
Number of questions
Simple and short questions
Avoid questions requiring calculations
Avoid questions using double negatives
Avoid leading questions
No personal questions
Questions should be logically arranged
Proper division and sub-division of questions
Instructions to the informants
Cross examination
Questionnaire should look attractive
Pre-testing the questionnaire
CENSUS METHOD
MEANING – when a statistical investigation is conducted wherein,
the data is collected from each and every element of the population,
it is known as census method. Also known as –
Method of complete enumeration
100% Enumeration
SUITABILITY
1. Size of the population is small
2. Extensive study of diverse item is required
3. High degree of accuracy is needed
4. Selection of sample items from universe is not possible.
5. Reliable data is required.
MERITS
1. Accurate and reliable
2. Less chances of biasness
3. Extensive study of diverse items
4. Complex investigations
5. Indirect investigations
DEMERITS
1. Expensive
2. Large manpower requirement
3. Unsuitable for large case studies
4. Not universally operative.
SAMPLE METHOD
MEANING – When only some items of a
population are selected and data collected
from these items are used for the analysis,
the method is known as sampling method.
SAMPLE METHOD
FACTORS AFFECTING THE SIZE OF SAMPLE –
Size of population
Nature of study
Respondent’s Nature
Degree of accuracy desired
Homogeneity or heterogeneity of population
SAMPLE METHOD
REQUISITES OF A GOOD SAMPLE
1. Representative
2. Adequacy
3. Homogeneity
4. Independence of selected sampling units
5. Matches with objective of investigation
SAMPLE METHOD
MERITS
1) Reduced cost
2) Greater speed
3) Greater accuracy
4) Greater scope
SAMPLE METHOD
MERITS
5) Administrative convenience
6) Scientific approach
7) Detailed enquiry
8) Exceptional applicability
SAMPLE METHOD
DEMERITS
1) Difficult to achieve cent percent accuracy
2) Lack of representativeness of the sample
3) Bias in the selection of sample
4) Lack of specific or specialized knowledge
5) Not possible in case of heterogeneous population
CENSUS METHOD VS
SAMPLE METHOD
BASIS CENSUS METHOD SAMPLE METHOD
Nature of Extensive enquiry Limited enquiry
enquiry
Economy It requires large amount Relatively less money, time
of money, time and labor. and labor is required.
Suitability Population is Population is
heterogeneous. homogeneous.
CENSUS METHOD VS
SAMPLE METHOD
BASIS CENSUS METHOD SAMPLE METHOD
Reliability Results are quite reliable Results are less reliable and
and accuracy and accurate. accurate.
Nature of Error of bias. Error of sampling.
error
Organization It is very difficult to It is comparatively easy to
and organize and supervise organize and supervise
supervision census method.
 TYPES OF
SAMPLING

NON
RANDOM RANDOM
SAMPLING SAMPLING

SIMPLE RESTRICTED
JUDGEMENT QUOTA CONVENIENCE
RANDOM RANDOM
SAMPLING SAMPLING SAMPLING
SAMPLING SAMPLING

STRATIFIED SYSTEMATIC CLUSTER

RANDOM SAMPLING SAMPLING
SAMPLING
RANDOM SAMPLING
It is a method in which every item in the universe has a known
chance (probability) of being chosen for the sample.
Selection of sample is independent of person making the study.
There are 2 methods of random sampling
1. Simple random sampling
2. Restricted random sampling
SIMPLE RANDOM
SAMPLING
A simple random sample is one in which every item of population
has an equal chance of being selected.
It is known as unrestricted random sampling method as method of
selection shall not favor one item or another.
Under this method sample is selected by two methods:
1. Lottery method
2. Tables of random numbers
 RANDOM
NUMBER TABLE

SIMPLE
RANDOM
SAMPLING

LOTTERY
METHOD
1. LOTTERY METHOD
Under this method chits or paper slips are
made of each item in population.
Then chits are shuffled in the container
without any biasness, some slips are drawn
out.
The chits then become sample of universe.
Used in lottery draws for allotment of
plots, houses or selecting topics for group
discussions or in extempore.
2. TABLE OF RANDOM NUMBERS
2952 6641 3992 9792 7969 5911
3170 624 4167 9524 1545 1396

7203 5356 1300 2693 2370 7483

3408 2762 3563 1089 6913 7691
MERITS
1. Free from bias.
2. Equality
3. Representative
4. Simple
5. Accuracy can be assessed
DEMERITS
1. Unsuitable for small sampling
2. Time consuming
3. No proportionate presentation
4. Ignores important items
 RESTICTED RANDOM
SAMPLING
Simple random sample is suitable when items of the universe are
homogeneous.
In case where different items of population are heterogeneous, samples
are selected under some restrictions. Methods are:-
1. Stratified random sampling
2. Systematic Sampling
3. Cluster Sampling
4. Multistage Sampling
STRATIFIED RANDOM
SAMPLING
The entire population is divided into a number of groups or strata
and the certain number of items are taken from each group at
random.
It ensures that all characteristics of a heterogeneous population are
adequately represented in the sample.
 Proportionate Disproportionate
Optimum manner
manner manner
• Sample size of • Sample sizes are • Size of the
different strata is fixed arbitrarily samples of the
fixed at certain without any regard different strata is
percentage or to the sizes of the fixed on the
ratio. different strata. principle of
obtaining
maximum benefits
at the minimum
costs.
MERITS DEMERITS
 covers diverse characteristics Limited scope
Representative Difficult to determine the size of
Accuracy strata
More expensive
Administrative convenience
Difficult if size of the population
is small
SYSTEMATIC SAMPLING
Under this method out of complete list of available population, the
sample is selected by taking every nth item from the list.
Example : if 10 items are to be selected from 100 items, then every
10th item from the population would be selected randomly.
If the first selected item is 9th item, then the subsequent numerical
are 19th, 29th, 39th, 49th, 59th, 69th, 79th, 89th, 99th .
SYSTEMATIC SAMPLING
MERITS DEMERITS
 simple and convenient Misleading results
Less time consuming Possible only when complete
Unbiased list is available
Unsuitable in case of large
Satisfactory conclusions
population
CLUSTER SAMPLING
This method is used when the size of population is very large.
This universe is divided into some clusters from which sampling is
carried out in a number of stages.
Example : if we need to conduct a survey in a country, we will first
divide a country into regions or states (stage I), then into cities or
towns (stage II) then further into localities (stage III) etc.
MERIT DEMERIT
Covers large area Expensive
Flexible Less Accurate
NON RANDOM SAMPLING
The selection of sample depends on the
judgement of the investigator rather than on
chance.
If the investigator is unbiased and intelligent, the
selected sample may be more representative of
the population rather than random sample.
NON RANDOM SAMPLING
Not every item of population has a known
chance of being selected.
It is not a scientific method. Hence law of
statistics cannot be used to analyze it.
JUDGEMENT SAMPLING
Method in which investigator himself chooses the
samples from the universe, which in his opinion are
the best representative of the population.
The method is used when some items are of more
importance then the others and need to be
selected as a sample item.
JUDGEMENT SAMPLING
Example, An investigator conducting an
opinion poll to ascertain the view of
students in regards to introduction of
online classes. The investigator may select
100 students out of 1000 students who
according to him is representative of
population.
JUDGEMENT SAMPLING
MERITS DEMERITS
simple and easy Personal prejudice
Proper presentation Error of judgment
Avoid irrelevant items Inaccurate results
QUOTA SAMPLING
Methods in which population is divided into
different groups or classes according to different
characteristics of the population, is called quota
sampling.
But within the given quota the selection of sample
units depends upon the personal judgement of the
investigator.
QUOTA SAMPLING
Quota can be fixed based upon characteristics like,
gender, age, annual income, etc.
Quota sampling is mixture of stratified sampling and
judgement sampling
QUOTA SAMPLING
MERITS DEMERITS
Reliable results Personal bias
Economical Impossible to detect sample
error.
CONVENIENCE SAMPLING
Under this method, while selecting the sample units,
the investigation gives special attention to his
convenience.
This method of selecting sample is called is also called
chunk.
CONVENIENCE SAMPLING
MERITS DEMERITS
Economical  lack of representation
Easy command Personal prejudice
Suitable for pilot surveys Unsatisfactory results
PRINCIPLES OF SAMPLING
LAW OF INERTIA OF LARGE
LAW OF STATISTICAL REGULARITY
NUMBERS
If a random sample of adequate The aggregates or averages
size is selected from a large obtained from a large group are
population it tends to possess more stable than the aggregate
the same characteristics as those or averages obtained from small
of population . group.
STATISTICAL ERRORS
The difference
between the Selection of wrong Incorrect information
collected data samples. given by respondents.
and actual value
of fact is termed
as statistical CAUSES
errors.
Collection of data by Personal prejudice of
estimates investigators
SOURCES STATISTICAL
ERRORS

Errors of Error of Error of

origin manipulation inadequacy
TYPES OF STATITICAL
ERRORS
Sampling errors

Non – sampling errors

Biased errors

Unbiased errors
SAMPLING ERRORS

Difference between the sample estimate and actual value of a

characteristic of the population.

The magnitude of sampling error can be reduced by taking a larger

sample.
NON-SAMPLING ERRORS

Errors that occurs in acquiring, recording or tabulating statistical data.

Non sampling errors are more serious than sampling errors because a
sampling error can be minimized by taking a larger sample.
 ERRORS DUE TO
MEASUREMENT

ERRORS DUE TO
ERRORS DUE TO PERSONAL BIAS
NON RESPONSE OF THE
INVESTIGATOR

CAUSES
OF NON
SAMPLING
ERRORS
EXAMPLES OF NON
SAMPLING ERROR

ERRORS IN DATA ACQUISITION: This type of error arises from

recording of incorrect responses or in recording data.

SAMPLING BIAS: It occurs when the sampling plan is such that some
members of the target population could not possibly be included in
the sample.
EXAMPLES OF NON
SAMPLING ERROR

NON-RESPONSE ERRORS: Non response occurs if an interviewer is

unable to contact a person listed in the sample or a person from
the sample refuses to respond.
In this case, the sample may not be representative.
 • Arises on account of some bias on the part of
Biased the investigator, informants or instruments of
counting, measurement or experiment is
error called biased error.

• An error which does not take place

on account of any bias with anybody
Unbiased but for a chance or a principle is
error called unbiased error.
• Such errors arise automatically
without any motive.
MEASURES OF STATISTICAL
ERRORS
• Difference between the
Absolute Error true value and its
estimated value.

• Ratio of the absolute error

Relative Error to the actual value.
PUBLISHED UNPUBLISHED
SOURCES SOURCES

SOURCES OF
SECONDARY
DATA
PUBLISHED SOURCES
Official publications of central and state governments
Semi-government publications
Reports of committees and commissions
Publications of the research institute
International publications
Private publications
Publications of trade associations
Newspapers and magazines
UNPUBLISHED SOURCES
When data collected by someone is not
published and is taken by the other
persons for the investigation, then such
data is known as ‘unpublished secondary
data’.
LIMITATIONS OF SECONDARY
DATA
No proper procedure is adopted to collet the
data.
Secondary data may be influenced by the
prejudices of the investigator.
Secondary data sometimes lacks standard of
accuracy.
Secondary data may not cover the full period
of investigation.
 SUITABILITY CENSUS AND
FOR THE SAMPLING
PURPOSE METHOD

ADEQUACY
RELIABILITY OF
PRECAUTIONS AND
THE DATA
IN THE USE ACCURACY
OF
SECONDARY
DATA