BRM Unit 3
BRM Unit 3
CONCEPT OF SAMPLING
Sampling may be defined as the selection of some part of an aggregate or
totality on the basis of which a judgment or inference about the aggregate or
totality is made.
It is the process of obtaining information about the entire population by
examining only a part of it.
Sampling is the process of selecting units from a population of interest so
that by studying the sample we may fairly generalize our results back to the
population from which they were chosen
Sample is the portion of the population which is examined with the view to
estimate the characteristics of the population
LIMITATIONS OF SAMPLING
Misleading Conclusions - the sampling procedures must be correctly
designed and followed otherwise what we call as 'wild sample would crop
up with misleading results. The sampling error may be larger than expected
if the sampling procedure is improperly designed or incorrectly carried out
Need for specialized knowledge - Most of the sampling techniques require
the services of sampling experts or statisticians. In social research one needs
to have adequate experience of dealing with and with collecting social data
accurately
Problematic for Social Sciences - Social scientists are faced with additional
problems in Sampling because of scattered distribution of sample units, non
cooperative nature of respondents, in accessibility to respondents and so on
Limited scope to know certain population characteristics like population
growth rate, population density etc, census of population at regular intervals
is more appropriate than studying by sampling
Changeability of units - If the units to be measured are rare and highly
variable a very large sample is required in order to yield enough cases for
achieving statistically reliable information
Essentials of Sampling
Usefulness and reliability of sampling depends on the following features of
sample:
SAMPLING TECHNIQUES
There are different type of sample designs based on two factors (i.e.)
representation basis and the elements selection technique
On the representation basis, sampling may be probability sampling or non-
probability sampling
Probability Sampling - It is also known as 'random sampling' or
'choice sampling'. Under this sampling design, every item in the
universe have an equal or known chance of inclusion in the sample
(e.g.) lottery method in which individuals are picked up not
deliberately but by some mechanical process
Non- probability Sampling - It is also known as 'deliberate sampling'
or 'purposive sampling. In this type, items for the samples are selected
deliberately by the researcher. His choice concerning the items remain
supreme
On clement selection basis, the sampling may be either unrestricted or
restricted
Unrestricted sampling - When each sample element is drawn
individually from the population at large, then the sample so drawn is
known as unrestricted sample
Restricted sampling - remaining all other forms of sampling are
covered under the term restricted sampling
PROBABILITY SAMPLING
A SIMPLE RANDOM SAMPLING(UNRESTRICTED)
In the unrestricted probability sampling design, more commonly
known as simple random sampling every element in the population
has a known and equal chance of being selected as a subject
If there are 1000 elements in the population and if we need a sample
of 100 pieces of paper were dropped in the hat bearing the names of
all the elements and the 100 samples were selected with eyes closed.
The first piece drawn will have/1000 chance of being drawn and the
second piece will have 1/999 chance of being drawn and so on.
In the current scenario computers are used to generate random
numbers and one does not need to go through the tedious process of
picking out numbers
Types of simple random sampling are lottery method and table of
random numbers
ii) Table of random numbers: In this method, first the size of sample is
determined. Then using random number table, the required number of items is
selected to form the sample. Different Random number tables are available. In
these tables, random numbers are printed in columns and rows Random number
tables are available with two digits, three digits and more number of digits.
Suppose the sample size is 25 out of a total of 200 items. First, a serial number
[two or three or four digits) is assigned to each item. Then using a random number
table, row - wise or column wise, 25 numbers that match with the serial number are
selected. This will constitute the sample. For example, whenever a new car is
introduced in the market, people who are booking for the new car are assigned a
serial number. After the last date for booking is over, the Company will generate
those who would be given the allotment for the month, going by the random
numbers. Once the required numbers are selected, the customers who are assigned
that number are intimated about the allotment.
There are three different methods of restricted random sampling. These are
explained below.
Merits:
1. Stratified random sampling has better representativeness.
2. It also gives more accurate information and there would be a better
coverage of the population
Limitations:
1. This method requires lot of care and pre-planning. Unless the different strata
existing in the population is known with certainty, using this method is not
possible
3. Care should be taken to make selection from each stratum random, which
may not be practically possible. Any bias in selection from each stratum will
affect the accuracy of results.
Merits:
1. It is very convenient and simple to adopt
2. The time and cost involved are relatively less
3. With a large population, this method of sampling is easy to use.
4. Random selection of items is ensured, once the sampling interval is
determined
5. Sampling interval is determined scientifically depending upon the size of
Sample desired
Limitations :
1. It is less representative, as once the first item is selected at
random, subsequent items are all lying at uniform interval. So the selected
items may lack complete representativeness.
2. This method requires correct understanding of the
methodology as otherwise, the sample selected will not be correct.
3. The first item selected should be strictly at random. If there is
any bias in this first stage, this will influence the items selected at
subsequent stages.
AREA SAMPLING
The area sampling design constitutes geographic clusters
When the research pertains to populations within identical
geographical areas such as done countries, city blocks or particular
boundaries within a locality area sampling can be done.
Area sampling is therefore a form of cluster sampling within an area
Location plans for retail stores, advertisements specially focused on
local population could use an area sampling design
DOUBLE SAMPLING
This method is resorted when further information is needed from a
subset of the group from which some information has already been
collected for the same study
A sample design where initially a sample is used in a study to collect
some preliminary information of interest and later a subsample of this
primary sample is used to examine the matter in a more detailed
manner it is called double sampling
A structured interview might reveal that a subgroup of the respondents
have more insight into the problems of the organization and these
respondents are interviewed again by asking some additional
questions
a) Judgment Sampling:
In this method, the sample selection is purely based on the judgment of the
investigator or the researcher. This is because, the researcher may lack information
regarding the population from which he has to collect the sample. Population
characteristics or qualities may not be known, but sample has to be selected. In
such instances, judgment sampling method is used. The investigator on the field is
allowed to select any item of his choice to constitute the sample. Once the sample
Size is determined, the investigator is free to select any item on the field. For
example, suppose 100 boys are to be selected from a college with 1000 boys. If
nothing is known about the students in this college, then the investigator may visit
the college and choose the first 100 boys he meets. Or he may select 100 boys all
belonging to III Year. Or he might select 25 boys from Commerce course, 25 from
Science courses, 25 boys from Arts courses and 25 from Fine arts courses. Hence,
when only the sample size is known, the investigator uses his discretion and selects
the sample.
The strength of this method is that the sample elements are selected by the
investigator on the spot using his understanding of the population. So the samples
may fairly represent the population. But if the investigator may be biased, this type
of sampling may not be fully representing the population. Further, there is a need
to train investigator to do this job correctly.
c) Quota Sampling :
In this method, the sample size is determined first and then quota is fixed for
various categories of population, which is followed while selecting the sample.
Suppose we want to select 100 students, then we might say that the sample should
be according to the quota given below : Boys 50%, Girls 50% Then among the
boys, 20% college students, 40% plus two students, 30% high school students and
10% elementary school students. A different or the same quota may be fixed for
the girls.
In this method the quota has to be determined in advance and intimated to
the investigator. The quota for each segment of the population may be fixed at
random or with a specific basis. Normally such a sampling method does ensure
representativeness of the population.
D) SNOWBALL SAMPLING
Snowball sampling uses recommendations to find people with the
specific range of skills that has been determined as being useful.
An individual or a group receives information from different places
through a mutual intermediary
This is referred to metaphorically as Snowball sampling house as
more relationships are built through mutual association more
connections can be made through the new relationships and a plethora
of information can be shared and collected, much like a snowball that
rolls and increase in size as it collects more snow.
Snowball sampling is a useful tool for building networks and
increasing the number of participants. However, the success of this
technique depends greatly on the initial contacts and connections
made.
SAMPLING ERRORS
While using sampling, errors are committed. These errors are broadly
classified a sampling errors and non-sampling errors.
Sampling errors arise due to draw inferences about the population on the
basis of few observations. That is when the sample size is not a true
representative of the population, any conclusion drawn on the basis of
sample will not be applicable to the population
Such errors may be due to several factors. It may be due to biased errors and
non-biased errors
UNBIASED ERRORS
These errors are caused due to chances not strictly due to any bias on
the part of the investigator
Sometimes the findings from the selected sample may be different
from other elements not included in the sample
NON-SAMPLING ERRORS
Non-sampling errors are those errors which are not due to any
sampling process
Investigators may collect data without using the complete schedules
or proper measurement
Faulty method of interview or observation may also contribute to non
sampling errors
Use of untrained or unskilled investigators only causes more
confusion rather than carrying out the work properly. Such
investigators would report incomplete or irrelevant data
There are instances when the organizer of the survey does not plan for
close supervision and control over investigators in the field. As a
result coverage of respondents will not be complete
Mistakes committed in data processing is also a source of non-
sampling errors
Interpretation of data requires enormous experience and skill. Wrong
interpretation of results will result in wrong interpretation of outcome
The total sampling error is a combination of both sampling error and
non-sampling error. To minimize the sampling error if we increase the
sample size non-sampling error is bound to increase. Effective steps
should be taken to minimize the non-sampling errors.
Sample Size
One of the most important issues to be settled while using sampling method is to
determine the size of sample. More often than not, researchers commit mistakes in
deciding the sample size. There is a thumb rule, which says that to satisfy
statistical requirement, the size of the sample should be a minimum of 30 and
above. In statistics any sample with 30 and more elements is considered large
sample. Less than 30 elements in a sample constitute only small sample. By
definition this might appear to be an easy way out to determine the sample size.
But it should be noted that there are several other considerations or factors which
influence the sample size. These are discussed hereunder.
1. The size of the population : Depending upon the size of population, the size of
sample has to be decided. If the population is very small (say for example only 50],
then the sample size could also be small. But, it should be remembered that larger
the size of population, larger should be the sample size to achieve
representativeness and accuracy.
2. The resources available: What amount of time and financial resources are
available to the researcher will determine the size of sample. With sufficient time
and large volume of funds available, the sample size could be large, otherwise it
should be small.
3. The extent of accuracy desired: One of the requirements of data analysis is the
accuracy level. There is no choice about the level of accuracy that one wants to
achieve or maintain. But while dealing with large population, due to constraints on
resources, sampling has to be adopted. In that case, the level of accuracy that is
desired must be spelt out. Suppose, the choice is for achieving cent per cent
accuracy, then sampling itself cannot be thought of. Alternatively, if the accuracy
desired is 95% then sample size should be large. Any level of accuracy less than
this [95%) may require only small sample size.
PRIMARY DATA - Primary Data are those which are collected afresh and
for the very first time and thus happens to be original in character.
SECONDARY DATA - Secondary Data are those which are already
collected by someone else and which have been passed through the
statistical process.
The methods of collecting primary and secondary data differ since primary
data are to be originally collected, while in case of secondary data the nature
of data collection work is merely that of compilation
PANELS
Panels like focus groups are another source of primary information for
research purposes
Focus groups meet for a one time group sessions, panels meet more than
once
When the effects of certain interventions or changes are to be studied over a
period of time ,panel studies are very useful
Individuals are randomly chosen to serve as panel members
For e.g., if the effects of a proposed ad for a certain brand of coffee is to be
assessed quickly, the panel members can be exposed to that ad and their
intentions of purchasing that brand can be assessed. A few months later, the
product manager might think about introducing a change in the flavor of the
same effect and explore its effects on the panel.
Thus, a continuing set of experts serves as the sample base. Such expert
members composed the panel, and research that uses them is called panel
study.
Nielsen TV index is based on the television viewing patterns of the pencil
SURVEYS
Survey research is one of the most important areas of measurement in
applied social research. The dictionary meaning of survey is: "Survey is a
technique for gathering information from a large number of users" |Brehob,
20011.
Surveys are usually conducted in case of descriptive research. Survey type
research usually have larger samples.
Surveys are concerned with describing, recording, analyzing and
interpreting conditions that either exist or existed.
A “survey" can be anything from a short paper-and-pencil feedback form to
an intensive interview.
A survey is a data collection tool used to gather information about
individuals. Surveys are commonly used in psychology research to collect
self-report data from study participants.
A survey can be administered in a couple of different ways. In one method
known as a structured interview, the researcher asks each participant the
questions. In the other method known as a questionnaire, the participant fills
out the survey on his or her own.
Surveys are generally standardized to ensure that they have reliability and
validity. Standardization is also important so that the results can be
generalized to the larger population.
Types of Interview:
Interview may be broadly classified as
1. Direct interviews and 2. Indirect interview.
In both these types different approaches may be followed like: structure interview,
unstructured interview, focused interview, non-directive interview and telephonic
interview. These types of interview are discussed hereunder.
Direct Interview:
In this type of interview, the interviewer and the interviewee meet personally either
with prior appointment or not. This type of interview ma further be classified as
structured interview, unstructured interview, focused interview, clinical interview
and non-directive interview. Usually when this technique is adopted, the
interviewer may brief the respondent about the purpose of interview and its scope
in advance. This enables the respondents to be read with necessary details or data.
For example, suppose the Chairman of ISRO is t be interviewed. Then an
appointment must be obtained and the subject matter interview or the purpose of
the interview should be stated in advance. This will enable the Chairman to collect
the necessary data in advance so as to give authentic information at the time of
interview .But there are occasions when the respondent is contacted and his
opinion collected. In such cases there is n possibility for briefing him about the
interview. For example, television crew like SUN TV] meet the people coming out
a theatre screening a new film an ask them about the film. The response is simply
the opinion of the audience an there is no advance intimation given.
QUESTIONNAIRE METHOD
QUESTIONNAIRES
A questionnaire is a research instrument consisting of a series of questions to
be asked from the respondents and thereby gathering the desired information
from respondents. It also contains a suitable space where the answers can be
recorded.
The term questionnaire usually refers to a self-administered process where
the respondent himself reads the questions and records the answers without
the assistance of the interviewer.
A questionnaire is a method of obtaining specific information about a
defined problem so that the data after analysis and interpretation results in a
specific solution to the problem.
The success of collecting data through the questionnaire method largely
depends on the proper design of the questionnaire.
This is a specialized job and requires high degree of skill, experience,
thorough knowledge about the research topic, ability to frame questions and
a high degree of patience.
Questionnaires have advantages over some other types of surveys in that
they are cheap, do not require as much effort from the researcher as verbal or
telephone surveys, and often have standardized answers that make it simple
to compile data.
Questionnaires are also sharply limited by the fact that respondents must be
able to read the questions and respond to them.
TYPES OF QUESTIONNAIRE
Structured Questionnaire
Structured questionnaires are those questionnaires in which there
definite, concrete and predetermined questions. The questions are
presented with exactly the same wording and in the same order to all
respondents.
Such questionnaires may also have fixed alternative questions in
which the responses of the informants are limited to the stated
alternatives. The respondents own words are held to the minimum.
Structured questionnaires are simple to administer and relatively
inexpensive to analyse .
Unstructured Questionnaire
The interviewer is provided with the general guide on the type of
information to be obtained, but the exact question formulation is
largely his own responsibility and the replies are to be taken in the
respondent's own words as possible.
In some situations tape recorders may be used.
PRINCIPLES OF WORDING
CONTENT & PURPOSE OF THE QUESTIONS- The nature of the
variables tapped - subjective feelings or objective facts - will determine what
kind of questions will be asked.
If the variables tapped are of subjective in nature (satisfaction,
involvement) where respondent's beliefs, attitudes and perceptions are
to be measured, the questions should tap the elements and dimensions
of the concept.
DOUBLE-BARRELED QUESTIONS:
A question that lends itself to different possible responses to its sub
parts is called a double- barreled question.
Such questions should be avoided and two or more separate questions
should be asked instead. (e.g.) Do you think that there is good market
for the product and that it will sell well? could bring yes response to
the first part and no response to the second part.
Therefore it is better to ask two questions like (1) Do you think that
there is a good market for the product? (2) Do you think the product
will sell well?
AMBIGUOUS QUESTIONS :
Even questions that are not double barreled might be ambiguously
worded and the respondent may not be sure what exactly they mean
(E.g.) To what extent you say that you are happy? The respondents
may be confused whether the question refers to their workplace or at
home or in general. The respondents may not provide a correct
answer to the question.
RECALL-DEPENDENT QUESTIONS:
Some questions might respondents to recall experiences from the past
that are hazy in their memory. Answers to such questions may have
bias
(E.g.) employee with 30 years of experience if he is asked to state
when he first started working in a particular department and for how
long will not be able to recollect and may give biased answers. The
best source is therefore personal records.
LEADING QUESTIONS:
Questions should not be phrased in such a way that they lead the
respondents to give the answers that the researcher would like or
want them to give.
(E.g.) Do you think that in these days of escalating cost of living
employees should be given good pay rises?. By asking such a
question the respondents are pressured to say “yes”.
To what extent do you think that employees should be given higher
pay rises?
LOADED QUESTIONS :
Questions become loaded when they are phrased in an emotionally
charged manner. An example of such a loaded question is asking
employees " To what extent do you think management is likely to be
vindictive if the Union decided to go on strike?. The words “strike”
and “vindictive” are emotionally charged terms, polarizing
management and unions.
SOCIAL DESIRABILITY :
Questions should not be worded in such a way that they elicit socially
desirable responses. For e.g., a question such as Do you think that
older people should be laid off?" would elicit a response of “no"
mainly because society would frown on a person who says that
elderly people should be fired even if they are capable of performing
the jobs satisfactorily .
LENGTH OF QUESTIONS:
Short questions are preferred to long one. As a rule of thumb a
question or a statement in the questionnaire should not exceed 20
words
SEQUENCING OF QUESTIONS:
The sequencing of questions should be such that the respondents are
led from the questions of a general nature to those that are more
specific and from questions that are relatively easy to answer to those
that are progressively more difficult. This funnel approach facilitates
the easy and smooth progress of the respondent.
VALIDATION OF QUESTIONNAIRE
Validation" is also the process by which any data collection instrument,
including questionnaires, is assessed for its dependability.
Validating questionnaires is somewhat challenging as they usually evaluate
subjective measures, which means they can be influenced by a range of
factors that are hard to control.
When validating a questionnaire it is important that a sample is used that is
representative of the population in which the instrument is to be used.
There are no general criteria for the required sample size in a validation
study. A sample size of at least 50-100 participants is generally
recommended. However, certain methods require larger numbers of
participants.
The time interval between assessments is important for determining test
retest reliability. This interval should not be too brief in order to prevent
participants remembering what they previously answered, and should not be
too long either, as participants' scores may have actually changed.
It is important to establish a priori hypothesis when determining validity and
responsiveness, for instance, expected correlations with other questionnaires
or anticipated differences in scores between certain subgroups. These
hypotheses should be as specific as possible.
Validation involves establishing that the instrument produces data that are
reliable and true. There are a number of ways to define this, some of which
outlined below.
VALIDITY: the degree to which a questionnaire reflects reality. There are a
number of different facets to validity.
INTERNAL VALIDITY: the degree in which questions within an
instrument agree with each other. i.e. that a subject will respond to similar
questions in similar way.
STATISTICAL VALIDITY: this is related to internal validity, and assesses
whether the differences in the questionnaire results between patient groups
can appropriately be subjected to statistical tests of significance.
CONTENT VALIDATION: It often refers to face validity. Face validity is
determined by comparing the questionnaire with other similar questionnaire
surveys. Face validity was also determined through pre testing such as the
initial pilot study.
SAMPLING VALIDITY: It is another component of validation. A large
sample size can ensure low sampling errors and high sampling validity.
EXTERNAL VALIDITY: the ability to make generalizations about a
population beyond that of the sample tested.
LONGITUDINAL VALIDITY: whether a questionnaire returns the same
results in a given population over time, assuming all else remains equal.
LINGUISTIC VALIDITY: whether the wording of the questionnaire is
understood in the same way by everyone who completes it.
EMPIRICAL VALIDITY: It examines the survey results by comparison
with other studies. The aim is to check consistency with previous results.
Empirical validation of the questionnaire reliability often involves two
techniques
Test - Retest Techniques : It determines stability of measured
indicators
Construct Validity : It is a scoc to determine internal consistency-
reliability, measured by the Cronbach alpha.
SCHEDULE
Schedule is a device in social research, which is most frequently used in
collecting field data especially where the survey method is employed. It is
used in indirect interview.
It contains questions and blank tables which are to be filled in by the
investigators themselves after getting information from the respondents.
Outwardly schedule and questionnaire appear to be the same but there is
difference between the two. Schedule is used in direct interview and direct
observation and is filled in by the research work himself .
Schedule is the name usually applied to a set of questions which are asked
and filled in by an interviewer in a face-to-face situation with another
person Goode and Hatt
Use of Schedules:
Whenever a researcher uses schedule to collect data, the following steps should be
taken.
1. The researcher should prepare the list of respondents to be contacted or the area
where they can be contacted.
2. Depending on requirement, investigators should be selected.
3. The investigators should be explained the purpose of the survey, their job,
responsibilities, accuracy desired, importance of answers recorded, the need to be
clear in recording details, etc.,
4. The investigators should be trained with the schedule.
5. Necessary instructions to the investigators to deal with different types of
problems [like non- availability of respondent, non-cooperative respondent, etc) on
field should be given.
6. Sufficient arrangement for field control, monitoring and supervision should be
made.
7. Method of follow up of non-response should be determined in advance.
8. The investigators selected should be honest, sincere, hard working, patient and
take all efforts to collect the details.
9. Time frame for completing the data collection should be clearly spelt out.
10. Periodical review of the filled up schedules would help to reframe the
questions or change the sequence or initiate corrective action.
OBSERVATION METHOD
It is the most common method used in studies related to behavioral sciences.
Observation becomes a scientific tool and the method of data collection for
the researcher when it serves a formulated research purpose, is
systematically planned and recorded and is subjected to checks on validity
and reliability.
Under this method, the information is sought by investigator's own direct
observation without asking from the respondent.
Observation as a method of data collection is used very frequently,
whenever collection of data through other methods is difficult For example,
it is not always possible to conduct interviews with every person to collect
required information. Similarly, using questionnaire or schedule for data
collection is not free from limitations, especially when the response to
questionnaire is very poor. In such occasions observation is more useful for
collecting data. There are occasions when no other method can be adopted
for data collection. For instance, suppose a scholar wants to study the life
style of a hill tribe. It is certainly not possible to use a questionnaire or
schedule or conduct interview. Only alternative available is observation as
the respondents would not reply any question orally or in written.
. Observation may be defined as, "sensible application of sense organs in
understanding less explained or unexplained phenomena."
Types of Observation:
1. Structured observation : When observation takes place strictly in accordance
with a plan or a design prepared in advance, it is called structured observation. In
such a type, the observer decides what to observe, what to focus on, what type of
information or activity should be given importance who are all to be observed,
what conditions are to be fulfilled to carry out observation, etc. in advance. He
sticks to these decisions when he observes.
Advantages:
a) filtering what is relevant and focusing on it,
b) avoiding waste of time
c) studying only the target group
d) Minimizing the bias of the observer and
e) Eliminating any scope for vagueness.
Disadvantages:
a) Unless a person is well trained he cannot be effective in undertaking
structured observation,
b) It is possible that vital aspects are missed, as they were not conceived at
the time of designing the observation.
c) By limiting the independence of the observer, crucial things might be left
out from being observed.
2. Unstructured observation: In this type of observation, there is no advance
designing of what, how, when, who, etc., of observation. The observer is given the
freedom to decide on the spot, to observe everything that is relevant. For example,
while observing the life style of a hill tribe in their place, the observer may be able
to observe several interesting things like how they discuss their plans, how they
entertain themselves, how they train the youngsters, how they settle quarrels, who
takes decision, etc. Some these may not be observed if structured observation is
adopted. The main strength of the unstructured observation is that at the time of
observation everything is taken note of and the researcher then segregates the
related and relevant details. Of course, one important quality required of the
observer is that he should be well trained and experienced.
3. As there is no need to get any reply or details from the respondents, observation
does not require any cooperation of the respondents.
Limitations of observation :
1. This is a relatively costly method of data collection, because the observer
has to be present on the spot to observe whatever is happening . Further training an
observer is the additional cost to be incurred.
2. The scope for the bias of the observer interfering in what is observed and
understood is high.
3. It could be noticed that what is observed may bring out only part of the facts,
while data collected through questionnaire or interview ensure better coverage.
4. Sometimes, it may not be possible to observe what is targeted. This may be due
to the difficulty in reaching such people or spot.
6. A very important limitation is that only the observer can effectively establish
link among whatever he has observed to give a meaningful interpretation. Based on
his report nobody else can give satisfactory interpretation. Therefore, the ability to
relate things and interpret them intelligently is a quality expected of an observer.
3. The person should avoid any of his personal likes and dislikes intervene in the
process of observation.
4. Unless the observer is alert and intelligent he might ignore important things.
5. The observer should be able to convince his target audience the need and
importance of what is observed.
6. Unless an observer is able to relate various things seen or heard or felt, he may
not be able to give a satisfactory explanation for what is observed. So he should be
a person of analytical and critical nature.
Before concluding, it should be noted that observation alone may not be useful in
collecting data. If it is combined with other methods of data collection like
questionnaire / schedule method or interview method, it would help to make data
collection comprehensive.
a) Book : This refers to the published documents like books. The data
contained in standard books published in the past give very valuable
information. But there are occasions when data published by different books
differ, causing confusion among the people compiling data.
b) Periodicals or journals: This is one of the very valuable source of
information. While books may deal with historical information, periodicals
contain articles and abstracts which provide the latest information. The
current thinking on any aspect is brought out through such publications.
c) Research thesis and dissertations: This is another useful source of
secondary data . But while using this source, the scholar should be careful to
study the objectives of the thesis being referred. It should also be
remembered that thesis submitted in the distant past may be used only for
understanding the relationship but not for replication.
d) Bibliographies: These are alphabetically arranged list of useful publications
with details about the year of publication, author, title, etc. This may serve as
a source of information to locate and access the information relevant to the
work being done.
e) Foot notes: At the end of each page of any published material, foot notes are
given. These notes would throw light on the source of information. By
tracing this source, original information and additional facts may be
available.
f) Encyclopedia: This is a publication where every useful information is
presented with authority, supported by diagrammes and facts. This will give
an authentic base for any fact being used in the research study. Several
standard encyclopedia are available and the scholar may refer them for
reliable information.
g) Statistical data source : Several organization publish at regular intervals
compiled statistical data on various aspects. For example, Statistical Hand
Book, United Nations Statistical year book, Demographic year book, Census
publications, etc., compile, classify and present codified statistical
information on various aspects. Use of authentic source of statistical data
would add to the quality of the research work.
h) Directories and Year Books: This is another source of secondary data in
which specific and focused information is available. For example, in
Kothari's Directory of Industries give reliable data industry-wise in India
every year. Year books are published by various agencies giving details
about the current status of various things. Examples : Manorama Year Book,
India - an Year Book, etc
2. Suitability of the data: Sometimes the data may be available, but it may not be
suit the purpose of the study on hand. For example, RBI bulletin published data
regarding Foreign Investment. Before 1997, no break up of the Foreign investment
used to be given. But after 1997, the portfolio component and the Direct
investment component of the Foreign investment is given. A scholar working on
the Foreign Direct Investment will feel handicapped in the absence of specific data.
Though data are available, they are not suitable for the study.
4. Continuity of data: This another problem using secondary data. For example, the
format used for providing some data may not be consistently followed by the
source. Depending upon the page constraints or availability of material, the data
presentation in a format is decided. So frequent change in the format would only
add to confusion to the people using the data from such sources.
3. Availability of Time: The time frame fixed for the research work will
directly influence the choice between primary and secondary data. If
sufficient time is available, the researcher could probe into the past with
secondary data and capture the present using primary data.
Other methods
d) Feedback form: In the case of some of the consumer goods, the supplier or the
manufacturer send the product along with a pre-paid reply cover in which
questions on the product and its usage are raised and the customer is requested to
fill it up and send. Based on this the first hand information about the product from
the consuming public is obtained.
f) Panty audit : In this method, the investigators stay at the consuming point and
observe the purchasing behavior or the people. From this very valuable
information like the price response, quality consciousness, response to prize
incentive schemes, who takes buying decision, credit facility required, etc;All these
will influence the marketing strategy of the organization concerned.
j) Content analysis: In this method, the investigator would listen through various
audio visual records to obtain the required information. Of course, in this process,
what is already recorded by somebody for some purpose is used to cull out useful
information.