MEASUREMENT AND SCALING

In our daily life, we involve ourselves in some kind of measurement though we are not aware of it. For
instance, when we see a movie, we either like or dislike it. Here, we are using some form of measurement
to indicate how that movie was.
In marketing, a firm may be interested to know the response in the market for its product or service. This
means it must measure the response, which can be in terms of sales, peoples attitude etc. thus, we find that
in marketing research, the measurement process involves the use of data (i.e. numbers to represent the
marketing phenomenon. In short, our ob!ective is to evolve some form of measurement scale.
MEASUREMENT
"easurement may be defined as the assignment of numbers to characteristics of ob!ects or events according
to rules. In marketing, we don not measure a buyer or a product, but we measure a certain characteristic of
a buyer or a product. For instance, a buyers preference, income, or attitude can be measured. #ikewise, in
respect of a product, we can measure its si$e, colour or its suitability for a particular purpose.
SCALING
%hile designing a &uestionnaire with a purpose to get peoples opinion, the most challenging task before a
researcher is 'how to measure and &uantify the responses that are to be collected from the respondents(
For that you need a SCALE!!
) scale is a measurement tool that determines &uantitatively sub!ective or abstract concepts. The process of
defining a suitable scale is called scaling. *caling is defined as a procedure for the assignment of numbers
to a property of ob!ects to impart some of the characteristics of numbers to the properties in &uestion.
It may be stated that a scale is a continuum, consisting of the highest point (in terms of some characteristic
e.g., preference, favorableness, etc. and the lowest
point along with several intermediate points between these two e+treme points. These scale,point positions
are so related to each other that when the first point happens to be the highest point, the second point
indicates a higher degree in terms of a given characteristic as compared to the third point and the third point
indicates a higher degree as compared to the fourth and so on. -umbers for measuring the distinctions of
degree in the attitudes.opinions are, thus, assigned to individuals corresponding to their scale,positions.
Hence the term 'scaling is applied to the procedures for attempting to determine &uantitative measures of
sub!ective abstract concepts.
USE OF SCALING IN MARKETING RESFARCH
"arketing segmentation strategy is based on attitudinal segmentation. In measuring the effectiveness of
advertising, attitude measurement plays an important role. *caling refers to the process of measuring of
attitudes.
TYPES OF MEASUREMENT SCALES
There are four types of measurement scales, namely/
0. -ominal scale,
1. 2rdinal scale,
3. Interval scale, and
4age 0
5. 6atio scale.
Nominal scales
-ominal scale is simply a system of assigning numbers that serve as labels to identify persons, ob!ects or
events. The usual e+ample of this is the assignment of numbers of basketball players in order to identify
them. *uch numbers cannot be considered to be associated with an ordered scale for their order is of no
conse&uence7 the numbers are !ust convenient labels for the particular class of events and as such have no
&uantitative value.
%e might further use the nominal scale by counting students with a certain characteristic or attribute such
as those who reside in the university hostels and others. 8onsider the following e+ample/
Number of students
Undergraduates Postgraduates Total
9ay scholars :;; 1;; 0;;;
Hostellers 5;; 0<; <<;
Total 01;; 3<; 0<<;
In this e+ample, students have been identified and counted by two characteristics, namely, whether they are
pursuing an undergraduate or post,graduate course of study and their place of residence. The numbers
given in each of the four cells are mutually e+clusive and the total of these four cells gives the total number
of students. In a nominal scale, the only operation involved is the counting of numbers in each group. It is,
therefore, the simplest of the four scales and also the least powerful. The scale does not show any order or
distance relationship nor does it have any arithmetical origin. In view of these limitations of a nominal
scale, it is unsuitable in determining relationships but is very useful
in preliminary or e+ploratory work, where it is sufficient to know the broad dimensions of a certain
phenomenon.
Ordinal Scales
)n ordinal scale, besides having uni&ue characteristics of a nominal scale, this scale also indicates order.
The ordinal scale places events in order, but there is no attempt to make the intervals of the scale e&ual in
terms of some rule. 6ank orders represent ordinal scales and are fre&uently used in research relating to
&ualitative phenomena. ) student=s rank in his graduation class involves the use of an ordinal scale. 2ne
has to be very careful in making statement about scores based on ordinal scales.
For instance, if 6ams position in his class is 0; and "ohans position is 5;, it cannot be said that 6ams
position is four times as good as that of "ohan. The statement would make no sense at all. 2rdinal scales
only permit the ranking of items from highest to lowest. 2rdinal measures have no absolute values, and the
real differences between ad!acent ranks may not be e&ual. )ll that can be said is that one person is higher or
lower on the scale than another, but more precise comparisons cannot be made. Thus, the use of an ordinal
scale implies a statement of 'greater than or 'less than without our being able to state how much greater or
less. The real difference between ranks 0 and 1 may be more or less than the difference between ranks <
and >.
Interval Scales
The third type of scale is the interval scale that possesses not only the power of the nominal scale and the
ordinal scale but also additional strength, which is the determination of e&uality of differences. The classic
e+ample of an interval scale is the measurement of the temperature. ?oth the Fahrenheit and 8entigrade
scales belong to this type. 2ne can say, on the basis of this scale, that a temperature of 0;;@ is 5;@ warmer
than >;@ and 1;@ cooler than 01;@ but one cannot say that the temperature of 0;;@ is five times as warm as
the temperature
4age 1
of 1;@ because both numbers are dependent on the fact that the $ero on the scale is set arbitrarily at the
temperature of the free$ing point of water. Asing the conversion formula from Fahrenheit to 8entigrade, we
get/
>;@F B 0<.<<@8
5;@F B 5.55@8
1;@F B ,>.>>@8
Therefore, 5;@F C 1;@F B 00.0;@8
)nd >;@F , 1;@F B 11.10@8
The above e+ample shows, that on a particular scale, e&ual differences indicate e&ual differences in value
with regard to that scale only. Interval scales are more powerful than the nominal and ordinal scales for
interval scale also incorporates the concept of e&uality of interval. )lso, they are &uicker to complete and
researchers find them more convenient.
Ratio Scale
6atio scales possess the powers of the preceding three scales as also the concept of absolute $ero or origin.
Thus, they have order, distance and uni&ue origin and are the most superior amongst all the scales.
D+amples of ratio scales are the commonly used physical dimensions such as height, weight, distance, and
money value.
%ith ratio scales involved one can make statements like EFyotisG typing performance was twice as good as
that of E6ituG. The ratio involved does have significance and facilitates a kind of comparison which is not
possible in case of
an interval scale.
Thus, proceeding from the nominal scale (the least precise type of scale to ratio scale (the most precise,
relevant information is obtained increasingly. If the nature of the variables permits, the researcher should
use the scale that provides the most precise description.
CRITERIA FOR A GOOD TEST
There are three important criteria for ascertaining whether the scale developed is good or not. These are/
6eliability
Halidity and
4racticality
Reliability
) measuring instrument is reliable if it provides consistent results. This can be undertaken by using the
same scale or test on the same set of respondents, using the same methods. However, in practice, this
becomes difficult especially as the same set of respondents may not be inclined to answer the same
&uestions again. )nother point to note is that same respondents may remember their previous answers to
the &uestions and give the same answers. This would indicate greater reliability of the scale or test than it
may be in reality. )nother factor that might distort the criterion of reliability is that the respondents may
become e+tremely cautious and careful the second time may avoid giving accurate answers or their attitude
might have changed during the intervening period. In either of these cases, the use of the reliability
criterion will not be very helpful as the test and retest scores will not be rigidly comparable.
If the time period between the test and retest is a short one, the respondents are likely to remember their
earlier answers and as such the memory effect may distort the reliability test though there would be meager
chances of change in the respondents attitude. 2n the contrary, if there is a long interval between the test
and the retest, developments during this interval may lead to a change in the attitudes of the respondents.
4age 3
Thus, one has to choose the timing of the retest in such a way so as to strike a balance between the memory
effect and the effect of
changes in views.
Validity
?y validity is meant the success of the scale in measuring what it sets out to measure. There are four
approaches to validity/
1. Content validity
8ontent validity is the e+tent to which a measuring instrument provides ade&uate coverage of the
topic under study. If the instrument contains a representative sample of the universe, the content
validity is good.
2. Construct validity
2ne of the difficulties arising in attitude measurement is that it is perhaps impossible to measure
attitude directly. It can be measured only indirectly on the basis of answers given by the
respondents. In a situation of this type, the test of construct validity is used. In order to apply
construct validity, the researcher postulates the nature of and e+tent of association between the
attitude and other specified variables. He then e+amines whether these relationships e+ist. If not,
there could be two possible e+planations. First, his scale might be invalid as it does not
satisfactorily measure what it set out to measure. *econd, his theory might be deficient in some
way and it may be difficult for him to identify it. The point to emphasi$e is that the construct
validity is based on theoretical considerations. For e+ample, the status of an individual in a society
may be dependent upon such variables as the level of education, occupation or ownership of a car
and a house. Thus, on the basis of theory, an elite class should have a high degree of association
amongst these factors. The e+istence of a high degree of correlation in this case is supporting
evidence and can be regarded as a test of validity.
. Predictive validity
4redictive validity signifies how well the researcher can guess the future performance form his
knowledge. For e+ample, an opinion &uestionnaire which forms the basis for correctly forecasting
the demand for a product has predictive validity. The procedure for predictive validity first measures
the attitude and then predicts the future behavior on the basis of this measurement. This is followed
by the, measurement of the future behavior at an appropriate time. Finally, the obtained scores are
compared with the earlier predicted scores. If the two series of scores are closely associated, the
scale is said to have predictive validity.
!. Concurrent validity
In the case of concurrent validity, an attitude scale on one variable can be used to estimate scores on
another variable. For e+ample, one may decide the social status of respondents on the basis of their
attitude towards savings.
6eliability does contribute to validity, but a reliable instrument need not be a valid instrument. For instance,
a scale that consistently overweighs ob!ects by <kgs is a reliable scale, but it does not give a valid measure
of weight. 2n the other hand, a valid instrument is always reliable.
This is evident from the following figures of a dart/
4age 5
9art 0 9art 1 9art 3
9art 0 shows absence of reliability and validity.
9art 1 shows that the results are reliable but not valid.
9art 3 shows that the results are both reliable and valid.
Practicality
The practicality characteristic of a measuring instrument can be !udged in terms of economy, convenience
and interpretability. From the operational point of view, the measuring instrument ought to be practical i.e.,
it should be economical, convenient and interpretable.
0. Econo"y consideration suggests that some trade,off is needed between the ideal research,pro!ect
and that which the budget can afford. The length of measuring instrument is an important area
where economic pressures are &uickly felt. )lthough more items give greater reliability as stated
earlier, but in the interest of limiting the interview or observation time, we have to take only few
items for our study purpose.
1. Convenience test suggests that the measuring instrument should be easy to administer. For this
purpose one should give due attention to the proper layout of the measuring instrument. For
instance, a &uestionnaire, with clear instructions (illustrated by e+amples, is certainly more
effective and easier to complete than one which lacks these features.
3. #nter$reta%ility consideration is especially important when persons other than the designers of
the test are to interpret the results. The measuring instrument, in order to be interpretable, must
be supplemented by (a detailed instructions for administering the test7 (bscoring keys7 (c
evidence about the reliability and (d guides for using the test and for interpreting results.
DIFFICULTY OF MEASUREMENT
Anlike physical sciences, measurement in social sciences is &uite difficult. The measurement of length,
height, weight is a simple task involving the use of ratio scale. In contrast, measurement in marketing is
more difficult as it involves lower scale of measurement as compared to those used in the physical sciences.
%hy measurement is more difficult in marketing compared to that in physical sciences( This can be
e+plained by taking an e+ample. *uppose we have to conduct a survey on the e+istence of medical facilities
in certain localities in a city. The point to note is that the attitude towards medical facilities would vary
from respondent to respondent. Dven the &uestionnaire design may not be a perfect instrument in eliciting
the desired information. Thus, we see that measurement of attitude poses a ma!or problem.
This would be evident form the following diagram/
4age <
I
I
I
I

I I
I
I
III
III
I
I
I
I
I
I
I
Difficulty of the measuring process
SOURCES OF ERROR IN MEASUREMENT
The following are the possible sources of error in measurement.
0. &es$ondent / )t times the respondent may be reluctant to e+press strong negative feelings or it is
!ust possible that he may have very little knowledge but may not admit his ignorance. )ll this
reluctance is likely to result in an interview of 'guesses. Transient factors like fatigue, boredom,
an+iety, etc. may limit the ability of the respondent to respond accurately and fully. Further,
measurement may get distorted on account of different opinions of the respondent on the given
issue. These differences may arise on account of the status of the respondent, level of education,
social class, etc.
1. Situation / *ituational factors may also come in the way of correct measurement. )ny condition
which places a strain on interview can have serious effects on the interviewer,respondent
rapport. For instance, if someone else is present, he can distort responses by !oining in or merely
by being present. If the respondent feels that anonymity is not assured, he may be reluctant to
e+press certain feelings.
3. 'easurer / The interviewer can distort responses by re,wording or re,ordering &uestions. His
behavior, style and looks may encourage or discourage certain replies from respondents.
8areless mechanical processing may distort the findings. Drrors may also creep in because of
incorrect coding, faulty tabulation and.or statistical calculations, particularly in the data analysis
stage.
5. #nstru"ent( Drror may arise because of the defective measuring instrument. The use of comple+
words, beyond the comprehension of the respondent, ambiguous meanings, poor printing,
inade&uate space for replies, etc. are a few things that make the measuring instrument defective
and may result in measurement errors. )nother type of instrument deficiency is the poor
sampling of the universe of items of concern.
4age >
4hysical *ciences
,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
*ocial *ciences
,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
#ength 4references Happiness
%eight )ttitude 8reativity
Dasy Hery
9ifficult
6atio *cale Interval *cale 2rdinal *cale -ominal *cale
COMPARISON BETWEEN MAKETING RESEARCH AND MARKETING INFORMATION
SYSTEM
Basis Mareting Research Mareting Information System
"eaning "6 is a systematic collection, recording and
analysis of data about problems relating to
marketing of goods and services.
"I* includes a set of procedures and methods for
the continuous analysis and presentation of
information for decision making in the area of
marketing.
*cope It handles problems pertaining to a particular
area. Thus it has a limited scope.
It is concerned with the problems in areas of
marketing such as sales, cost of distribution,
advertising
It operates on a specific marketing problem and
gets over after the research study.
Its a continuous process.
2b!ective Its concerned with finding solution to marketing
problem.
Its cautious against post marketing problem and
thus presents problems. It also helps in solve
problem when they arise.
2rientatio
n
It may be a post mortem of what has already
taken place or may relate to present situation for
a future decision.
Its always directed to future aspects of "arketing
management.
DATA PROCESSING, ANALYSIS AND INTERPRETATION
2nce the data have been collected, the researcher has to process, analy$e and interpret the same. It was
emphasi$ed that the researcher should e+ercise good care to ensure that reliable data are collected. )ll this
effort, however, will be in vain if the collected data are not properly processed and analy$ed. Thus
processing is essential for ensuring that we have all relevant data for making contemplated comparisons
and analysis. Technically speaking, processing implies editing, coding, classification and tabulation of
collected data so that they are amenable to analysis.
PROCESSING OPERATIONS
!" #diting
Dditing of data is a process of e+amining the collected raw data (especially in surveys to detect
errors and omissions and to correct these when possible. )s a matter of fact, editing involves a
careful scrutiny of the completed &uestionnaires and.or schedules. Dditing is done to assure that
the data are accurate, consistent with other facts gathered, uniformly entered, as completed as
possible and have been well arranged to facilitate coding and tabulation. )s it is a very sub!ective
process, it is necessary that persons, who are well,&ualified and trained in the !ob of editing,
should alone be entrusted with this responsibility.
For e+ample/
The respondent has given answers which are inconsistent with each other. In such a case, the
editor has to change one of the answers so as to make it consistent with the others. He has to use
his !udgment to decide which answer is correct so that the other one can be suitably changed.
The respondent has marked two answers instead of one for a particular &uestion. In such a case,
the editor has to carefully e+amine which of the two answers would be more accurate. *ometimes,
when a decision cannot be made categorically, he may prefer to code 'no information for that
&uestion.
4age J
?efore undertaking the coding, tabulating and analysis of responses contained in &uestionnaires,
they should be checked for co"$leteness) accuracy and uni*or"ity. The first point to check is that
&uestionnaires are complete and do not have any omissions or partial responses. *ometimes, the
interviewer might have forgotten to record the answer. In such cases, it may be difficult to fill in
the gap as the interviewer may be unable to recollect the answer given by the respondent. %hen
several &uestions remain unanswered in a &uestionnaire, the whole &uestionnaire may have to be
e+cluded.
The second point to check is that &uestionnaires contain accurate answers. The editor should look
for inconsistent answers. *uch inconsistencies may be a result of the carelessness of the
interviewer who may mark a wrong code or put the mark in such a way that it is not clear which of
the two codes is intended. )t times the respondent may give wrong or misleading answers
deliberately. In all such cases the editor has to go over the answers carefully and try to remove the
inconsistency in the best possible manner.
Finally, one should check whether the interviewers have interpreted &uestions and instructions
uniformly. %hile going through the &uestionnaires filled in by different interviewers (or
respondents in case of a mail &uestionnaire, the editor would be able to make out such
inconsistencies.
Dditing can be undertaken both at the time when the field survey is in progress and when it has
been completed. In the former case, it is known as *ield editing. %hen the interviewer fills in the
information at the time of the interview, he often uses several abbreviations due to the paucity of
time. These need to be spelt out fully later. It is advisable for the interviewer to carefully look into
the &uestionnaire at the earliest possible opportunity after the interview so that he can enter the
proper responses or complete the partial answers.
)nother type of editing is central editing, which is undertaken after the &uestionnaires have been
received at the head&uarters. )s far as possible, a single editor should carry out this task so that
consistency in editing can be ensured. However, in the case of large studies, this may not be
physically possible. %hen two or more editors are entrusted with the task of editing, it is
necessary that they are given uniform guidelines so that ma+imum possible consistency in their
approaches can be attained.
$" %lassification
"ost research studies result in a large volume of raw data which must be reduced into
homogeneous groups if we are to get meaningful relationships. This fact necessitates classification
of data which happens to be the process of arranging data in groups or classes on the basis of
common characteristics. 9ata having a common characteristic are placed in one class and in this
way the entire data get divided into a number of groups or classes.
9ata are classified on the basis of common characteristics which can either be descriptive (such as
literacy, se+, honesty, etc. or numerical (such as weight, height, income, etc..
9escriptive characteristics refer to &ualitative phenomenon which cannot be measured
&uantitatively7 only their presence or absence in an individual item can be noticed. 9ata obtained
this way on the basis of certain attributes are known as statistics o* attri%utes.
Anlike descriptive characteristics, the numerical characteristics refer to &uantitative phenomenon
which can be measured through some statistical units. 9ata relating to income, production, age,
weight, etc. come under this category. *uch data are known as statistics o* varia%les and are
classified on the basis of class intervals. For instance7 persons whose incomes, say, are within 6s.
1;0 to 6s. 5;; can form one group, those whose incomes are within 6s. 5;0 to 6s. >;; can form
another group and so on.
4age :
&" %oding
8oding refers to the process of assigning numerals or other symbols to answers so that responses
can be put into a limited number of categories or classes. 8oding is necessary to carry out the
subse&uent operations of tabulating and analy$ing data. If coding is not done7 it will not be
possible to reduce a large number of heterogeneous responses into meaningful categories. They
must also possess the characteristic of e+haustiveness (i.e., there must be a class for every data
item and also that of mutual e+clusively which means that a specific answer can be placed in one
and only one cell in a given category set.
8oding involves two steps. The first step is to specify the different categories or classes into which
the responses are to be classified. The second step is to allocate individual answers to different
categories.
4roblem in coding arises, especially in respect of an open,ended &uestion. The response to such a
&uestion is in a descriptive form, in the words of the respondent himself. In such cases, coding
needs e+tra care in framing the possible categories in which various responses can be classified.
*ometimes the interviewer himself decides the category in which a particular response to an open,
ended &uestion is to be included. He may first take down the entire response and then decide the
category in which it should be included.
'" (abulation
%hen a mass of data has been assembled, it becomes necessary for the researcher to arrange the
same in some kind of concise and logical order. This procedure is referred to as tabulation. Thus,
tabulation is the process of summari$ing raw data and displaying the same in compact form (i.e.,
in the form of statistical tables for further analysis. In a broader sense, tabulation is an orderly
arrangement of data in columns and rows.
Tabulation is essential because of the following reasons.
a It conserves space and reduces e+planatory and descriptive statement to a minimum.
b It facilitates the process of comparison.
c It facilitates the summation of items and the detection of errors and omissions.
d It provides a basis for various statistical computations.
The tabulation may be done by hand or by machine or some part by hand and the other by
machine. The number of tabulations will depend on the number of variables, while the number of
responses to a &uestion will depend on the sample si$e. If the number of variables in a survey is
not large and the sample si$e is small, then tabulation by hand is preferable. 2n the other hand,
when the number of variables involved is large as a result of which there may be several
relationships and the sample si$e is large, it will be more appropriate to opt for machine
tabulation.
The simplest way to tabulate is to count the number of responses to one &uestion. This is also
called univariate ta%ulation or one+,ay ta%ulation. %here two or more variables are involved in
tabulation, it is called %ivariate or "ultivariate ta%ulation or cross ta%ulation. In constructing
cross classification tables, one has to first determine which data should be given primary emphasis
and which should be given secondary emphasis. 9ata with primary emphasis are normally given
in columns while those with secondary emphasis are shown in rows. This order is repeated for
higher order table, i.e., those having three or more dimensions.
No. of movies see
!e" #ee$
No. of
"es!o%e&s
4age K
; K
0 0<
1 03
3 0;
5 <
< :
(otal '(
E-a"$le o* univariate ta%ulation
No. of movies see
!e" #ee$
Ge%e" No. of
"es!o%e&s
M)*e Fem)*e
; < 5 K
0 0; < 0<
1 K 5 03
3 3 J 0;
5 0 5 <
< > 1 :
(otal +, +' '(
E-a"$le o* %ivariate ta%ulation
%hile tabulating the data collected through &uestionnaire, the researcher has to be careful in
respect of some types of responses. *ometimes, the respondent has 'no answer to a particular
&uestion or 'does not know the answer. *uch answers are even encountered in respect of
dichotomous and multiple choice &uestions. *uch responses cannot be ignored. The researcher
should include them in the tabulation. ?ut before that, it should be ensured that these responses are
coded in the normal manner.
PROBLEMS IN PROCESSING
!" (he problem concerning )Don*t no+* responses
The researcher can handle '9ont know responses in three different ways/
a '9ont know may be shown as a separate category in the table.
b The researcher may allocate the '9ont knows among the other categories,
proportionately.
c The researcher may look into other information in the &uestionnaire where the '9ont
know response has figured. )t times, it may be possible for the researcher to infer from
other information readily available with him or her.
$" (he problem concerning multiple responses
*ometimes, multiple,choice &uestions allow respondents to give more than one response. %hen
there is more than one response to a &uestion, the percentage of responses will not add to 0;; per
cent. *uch responses can be handled in more than one way. The researcher has to look into this
aspect carefully and come to a decision which seems to be most appropriate.
&" Percentages
4ercentages are often used in data presentation for they simplify numbers, reducing all of them to
a ;,0;; range. Through the use of percentages, the data are reduced in the standard form with base
e&ual to 0;; which fact facilitates relative comparisons.
However, percentages have the following limitations that should be kept in view by researchers/
a Two or more percentages must not be averaged unless each is weighted by the group si$e
from which it has been derived.
4age 0;
b Ase of too large percentages should be avoided, since a large percentage is difficult to
understand and tends to confuse, defeating the very purpose for which percentages are
used.
c 4ercentages hide the base from which they have been computed. If this is not kept in
view, the real differences may not be correctly read. In other words, changes in
percentages cannot be interpreted.
ANALYSIS OF DATA
?y analysis we mean the computation of certain indices or measures along with searching for patterns of
relationship that e+ist among the data groups. )nalysis may, therefore, be categori$ed as descriptive
analysis and inferential analysis.
.escri$tive analysis is largely the study of distribution of one variable. This study provides us with profiles
of companies, work groups, persons and other sub!ects. This sort of analysis may be in respect of one
variable (described as unidimensial analysis, or in respect of two variables (described as bivariate analysis
or in respect of more than two variables (described as multivariate analysis.
#n*erential analysis (described as statistical analysis is concerned with the various tests of significance for
testing hypotheses in order to determine with what validity data can be said to indicate some conclusion(s.
It is mainly on the basis of inferential analysis that the task of interpretation (i.e., the task of drawing
inferences and conclusions is performed.
%e may as well talk of correlation analysis and causal analysis. Correlation analysis studies !oint variation
of two or more variables for determining the amount of correlation between the variables. Causal analysis
is concerned with the study of how one or more variables affect changes in another variable. This analysis
can be termed as regression analysis.
-owadays, with the availability of computer facilities, there has been a rapid development of "ultivariate
analysis which may be defined as Eall statistical methods which simultaneously analy$e more than two
variables.G Asually the following analyses are used for the purpose/
"ultiple regression analysis
"ultiple discriminant analysis
"ultivariate analysis of variance ( or multi, )-2H)
INTERPRETATION
)fter collecting and analy$ing the data, the researcher has to accomplish the task of drawing inferences
followed by report writing. This has to be done very carefully, otherwise misleading conclusions may be
drawn and the whole purpose of doing research may get vitiated. It is only through interpretation that the
researcher can e+pose relations and processes that underlie his findings. In case of hypotheses testing
studies, if hypotheses are tested and upheld several times, the researcher may arrive at generali$ations. ?ut
in case the researcher had no hypothesis to start with, he would try to e+plain his findings on the basis of
some theory.
Interpretation means drawing inferences from analysis of data. The task of interpretation has two ma!or
aspects/
the effort to establish continuity in research through linking the results of a given study with those
of another, and
the establishment of some e+planatory concepts.
4age 00
,hy interpretation-
Interpretation is essential for the simple reason that the usefulness and utility of research findings lie in
proper interpretation. It is being considered a basic component of research process because of the following
reasons/
0. It is through interpretation that the researcher can well understand the abstract principle that works
beneath his findings. Through this he can link up his findings with those of other studies, having
the same abstract principle, and thereby can predict about the concrete world of events. This way
the continuity in research can be maintained.
1. Interpretation leads to the establishment of e+planatory concepts that can serve as a guide for
future research studies7 it opens new avenues of intellectual adventure and stimulates the &uest for
more knowledge.
3. 6esearcher can better appreciate only through interpretation why his findings are, what they are
and can make others to understand the real significance of his research findings.
5. The interpretation of the findings of e+ploratory research study often results into hypotheses for
e+perimental research and as such interpretation is involved in the transition from e+ploratory to
e+perimental research.
(echni.ue of Interpretation
The task of interpretation is not an easy !ob7 rather it re&uires a great skill and de+terity on the part of
researcher. In order to do so, the researcher should possess a high degree of skill and e+ercise utmost care
and ob!ectivity, for errors of interpretation, if committed, can nullify even the best research.
The techni&ue of interpretation often involves the following steps/
0. 6esearcher must give reasonable e+planations of the relations which he has found and he must
interpret the lines of relationship in terms of the underlying processes and must try to find out the
thread of uniformity that lies under the surface layer of his diversified research findings. In fact,
this is the techni&ue of how generali$ation should be done and concepts be formulated.
1. D+traneous information, if collected during the study, must be considered while interpreting the
final results of research study, for it may prove to be a key factor in understanding the problem
under consideration.
3. It is advisable, before embarking upon final interpretation, to consult someone having insight into
the study and who is frank and honest and will not hesitate to point out omissions and errors in
logical argumentation. *uch a consultation will result in correct interpretation and, thus, will
enhance the utility of research results.
5. 6esearcher must accomplish the task of interpretation only after considering all relevant factors
affecting the problem to avoid false generali$ation. He must be in no hurry while interpreting
results, for &uite often the conclusions, which appear to be all right at the beginning, may not at all
be accurate.
Precautions in Interpretation
6esearcher must pay attention to the following points for correct interpretation/
0. )t the outset, researcher must invariably satisfy himself that/
the data are appropriate, trustworthy and ade&uate for drawing inferences7
the data reflect good homogeneity7 and that
proper analysis has been done through statistical methods.
4age 01
1. The researcher must remain cautious about the errors that can possibly arise in the process of
interpreting results. Drrors can arise due to false generali$ation and.or due to wrong interpretation
of statistical measures. He should be well e&uipped with and must know the correct use of
statistical measures for drawing inferences concerning his study.
3. He must always keep in view that the task of interpretation is very much intertwined with analysis
and cannot be distinctly separated. )s such he must take the task of interpretation as a special
aspect of analysis and accordingly must take all those precautions that one usually observes while
going through the process of analysis vi$., precautions concerning the reliability of data,
computational checks, validation and comparison of results.
5. He must never lose sight of the fact that his task is not only to make sensitive observations of
relevant occurrences, but also to identify and disengage the factors that are initially hidden to the
eye. This will enable him to do his !ob of interpretation on proper lines.
HYPOTHESIS
2rdinarily, when one talks about hypothesis, one simply means a mere assumption or some supposition to
be proved or disproved. ?ut for a researcher hypothesis is a formal &uestion that he intends to resolve.
Luite often a research hypothesis is a predictive statement, capable of being tested by scientific methods,
that relates an independent variable to some dependent variable. For e+ample, consider statements like the
following ones/
E*tudents who receive counseling will show a greater increase in creativity than students not receiving
counselingG 26
E)utomobile A is performing as well as automobile /.0
These are hypotheses capable of being ob!ectively verified and tested. It may be mentioned that while a
hypothesis is useful, it is not always necessary. "any a time, the researcher is interested in collecting and
analy$ing the data, indicating the main characteristics without a hypothesis. However, in a problem
oriented research, it is necessary to formulate a hypothesis or hypotheses in as clear terms as possible.
CHARACTERISTICS OF HYPOTHESIS
0. Hypothesis should be clear and precise. If the hypothesis is not clear and precise, the inferences
drawn on its basis cannot be taken as reliable.
1. Hypothesis should be capable of being tested.
3. Hypothesis should state relationship between variables, if it happens to be a relational hypothesis.
5. Hypothesis should be limited in scope and must be specific. ) researcher must remember that
narrower hypotheses are generally more testable and he should develop such hypotheses.
<. Hypothesis should be stated as far as possible in most simple terms so that the same is easily
understandable by all concerned.
>. Hypothesis should be consistent with most known facts i.e. it must be consistent with a substantial
body of established facts. In other words, it should be one which !udges accept as being the most
likely.
J. Hypothesis should be amenable to testing within a reasonable time for one cannot spend a life,
time collecting data to test it.
:. Hypothesis must e+plain the facts that gave rise to the need for e+planation. This means that by
using the hypothesis plus other known and accepted generali$ations, one should be able to deduce
the original problem condition.
PROCESS OF HYPOTHESIS TESTING
4age 03
there%y run the ris1 o* there%y run the ris1 o*
co""itting Ty$e # error co""itting Ty$e ## error
Step !/ Set Null hypothesis and 0lternative 1ypothesis
The conventional approach to hypothesis testing is to set up two hypotheses instead of one in such a way
that if one of them is true, the other is false. If we are to compare method ) with method ? about its
superiority and if we proceed on the assumption that both methods are e&ually good, then this assumption
is termed as the null hypothesis. )s against this, we may think that the method A is
superior or the method / is inferior. %e are then stating what is termed as alternative hypothesis.
The null hypothesis is generally symboli$ed as H; and the alternative hypothesis as Ha. *uppose we want to
test the hypothesis that the population mean (2 is e&ual to the hypothesi$ed mean (23 B 0;;.
Then we would say that the null hypothesis is that the population mean is e&ual to the hypothesi$ed mean
0;; and symbolically we can e+press as/
H;/ 2 B 2 B 0;;
4age 05
*tate H
;
as well as H
a
*pecify the level of significance
9ecide the correct sampling distribution
8hoose a test criterion
8alculate the probability
Is this probability e&ual to or smaller than M
value in case of one,tailed test and half M in
case of two,tailed test
6e!ect H
;
if
yes
)ccept H
;
if no
If our sample results do not support this null hypothesis, we should conclude that something else is true.
%hat we conclude re!ecting the null hypothesis is known as alternative hypothesis. In other words, the set
of alternatives to the null hypothesis is referred to as the alternative hypothesis. If we accept Ha then we are
re!ecting H; and if we re!ect H;, then we are accepting Ha.
For H;/ 2 B 2 B 0;;, we may consider three possible alternative hypotheses as follows=/
Alternative
4y$othesis
To %e read as *ollo,s
Ha/ 2 N 2 The alternative hypothesis is that the population mean is not e&ual to 0;; i.e. it may
be more or less than 0;;
Ha/ 2 O 2 The alternative hypothesis is that the population mean is greater than 0;;
Ha/ 2 P 2 The alternative hypothesis is that the population mean is less
than 0;;
The null hypothesis and the alternative hypothesis are chosen before the sample is drawn (the researcher
must avoid the error of deriving hypotheses from the data that he collects and then testing the hypotheses
from the same data.
In the choice of null hypothesis, the following considerations are usually kept in view/
0. )lternative hypothesis is usually the one which one wishes to prove and the null hypothesis is the
one which one wishes to disprove. Thus, a null hypothesis represents the hypothesis we are trying
to re!ect, and alternative hypothesis represents all other possibilities.
1. If the re!ection of a certain hypothesis when it is actually true involves great risk. It is taken as null
hypothesis because then the probability of re!ecting it when it is true is M (the level of significance
which is chosen very small.
3. -ull hypothesis should always be specific hypothesis i.e., it should not state about or
appro+imately a certain value.
Qenerally, in hypothesis testing we proceed on the basis of null hypothesis, keeping the alternative
hypothesis in view. %hy so( The answer is that on the assumption that null hypothesis is true, one can
assign the probabilities to different possible sample results, but this cannot be done if we proceed with the
alternative hypothesis. Hence the use of null hypothesis (at times also known as statistical hypothesis is
&uite fre&uent.
Step $/ Set the level of significance
The ne+t step is to test its validity at a certain level of significance. The confidence with which a null
hypothesis is re!ected or accepted depends upon the significance level. In case we take the significance
level at < per cent, then this implies that in the long run, the risk of making a wrong decision (a true H;
re!ected or a false H; accepted is < out of 0;; occasions. *imilarly, a significance level of say 0 per cent
implies that the researcher is running the risk of being wrong in 0 out of 0;; occasions.
The factors that affect the level of significance are/
the magnitude of the difference between sample means
the si$e of the samples
the variability of measurements within samples
whether the hypothesis is directional or non,directional () directional hypothesis is one which
predicts the direction of the difference between, say, means
Step &/ Decide the correct sampling distribution
The ne+t step in hypothesis testing is to determine the appropriate sampling distribution. The choice
generally remains between normal distribution and the t,distribution. The rules for selecting the correct
distribution are similar to those which we have stated earlier in the conte+t of estimation. .
4age 0<
Step '/ %hoose the test criterion
-e+t comes the selection of an appropriate statistical techni&ue as a test criterion. Fro e+ample, when the
hypothesis pertains to a large sample (3; or more, 5,test implying normal distribution is used. The t,test is
more suitable when the sample is small (less than 3;. The test criterions which are used fre&uently in
hypothesis testing are 5) t) F and 6
2
.
The researcher has to be careful in selecting a proper test. His choice should depend on a number of
considerations such as number of variables involved, the type of data collected, the si$e of sample, and
whether the samples are independent or related.
Step 2/ %omputations
)fter having selected the statistical techni&ue to verify the hypothesis, the ne+t step is the performance of
various computations, necessary for application of that particular test.
Step 3/ Decision
9rawing statistical decision involves acceptance or re!ection of null hypothesis. Qiven a null hypothesis
(H; and an alternative hypothesis (Ha, the researcher makes a decision rule according to which we accept
H; (i.e. re!ect Ha or re!ect H;
(i.e. accept Ha. This will depend on whether the computed value of the test criterion falls in the region of
acceptance or in the region of re!ection at a given level of significance.
STATISTICAL ERRORS IN HYPOTHESIS TESTING
There are four possibilities in hypothesis testing/
The null hypothesis is true but the test leads to its re!ection.
The null hypothesis is false but the test leads to its acceptance.
The null hypothesis is true but the test leads to its acceptance.
The null hypothesis is false but the test leads to its re!ection.
2f these four possibilities, the first two lead to erroneous decisions. %hen the researcher re!ects a true null
hypothesis, he makes Type I error (sy"%olically 789 or a *alse $ositive and when he accept a false null
hypothesis, he makes Type II error (sy"%olically 7:9 or a *alse negative.
In tabular form the said two errors can be represented as follows/
De-isio
Acce$t 4; &e<ect 4;
4; is true 8orrect decision Type I error (M
4; is
*alse
Type II error
(R
8orrect decision
The Type I error is also called the level of significance. The probability of Type I error is usually
determined in advance. If type I error is fi+ed at < per cent, it means that there are about < chances in 0;;
that we will re!ect H; when H; is true. %e can control Type I error !ust by fi+ing it at a lower level. For
instance, if we fi+ it at 0 per cent, we will say that the ma+imum probability of committing Type I error
would only be ;.;0.
?ut with a fi+ed sample si$e, say n, when we try to reduce Type I error, the probability of committing Type
II error increases. ?oth types of errors cannot be reduced simultaneously. There is a trade,off between two
types of errors which means that the probability of making one type of error can only be reduced if we are
willing to increase the probability of making the other type of error. To deal with this trade,off in business
4age 0>
situations, decision,makers decide the appropriate level of Type I error by e+amining the costs or penalties
attached to both types of errors.
If Type I error involves the time and trouble of reworking a batch of chemicals that should have been
accepted, whereas Type II error means taking a chance that an entire group of users of this chemical
compound will be poisoned, then in such a situation one should prefer a Type I error to a Type II error. )s
a result one must set very high level for Type I error in ones testing techni&ue of a given hypothesis.
Hence, in the testing of hypothesis, one must make all possible effort to strike an ade&uate balance between
Type I and Type II errors.
%omparison bet+een (ype I error and (ype II error
Ty$e # error Ty$e ## error
0. Its the error of first kind. Its the error of second kind.
1. Its the error of re!ecting a null hypothesis
when it is actually true.
Its the error of failing to re!ect a null hypothesis
when the alternative hypothesis is the true state of
nature.
3. Its a false positive. Its a false negative.
5. %hen we reduce Type I error, Type II
error increases.
%hen we reduce Type II error, Type I error
increases.
<. Its also known as 8 error. Its also known as : error.
For e+ample, re!ecting a good
friend
For e+ample, accepting a bad friend
TWO TAILED AND ONE TAILED TESTS
In the conte+t of hypothesis testing, these two terms are &uite important and must be clearly understood. )
two,tailed test re!ects the null hypothesis if, say, the sample mean is significantly higher or lower than the
hypothesi$ed value of the mean of the population. *uch a test is appropriate when the null hypothesis is
some specified value and the alternative hypothesis is a value not e&ual to the specified value of the null
hypothesis.
*ymbolically, the two tailed test is appropriate when we have H;/ 2 B 2 and Ha/ 2 N 2 which may mean 2
O 2 or 2 P 2.
Thus, in a two,tailed test, there are two re!ection regions, one on each tail of the curve which can be
illustrated as under/
"athematically we can state/
Acce$tance &egion )/ STSU 0.K>
&e<ection &egion 6/ STSV 0.K>
If the significance level is < per cent and the two,tailed test is to be applied, the probability of the re!ection
area will be ;.;< (e&ually split on both tails of the curve as ;.;1< and that of the acceptance region will be
;.K< as shown in the above curve. If we take 2 B 0;; and if our sample mean deviates significantly from
4age 0J
0;; in either direction, then we shall re!ect the null hypothesis7 but if the sample mean does not deviate
significantly from 2) in that case we shall accept the null hypothesis.
?ut there are situations when only one,tailed test is considered appropriate. ) one,tailed test would be used
when we are to test, say, whether the population mean is either lower than or higher than some
hypothesi$ed value. For instance, if our H;/ 2 B 2 and Ha/ 2 P 2, then we are interested in what is known
as left,tailed test (wherein there is one re!ection region only on the left tail which can be illustrated as
below/
"athematically we can state/
Acce$tance &egion )/ STSO ,0.>5<
&e<ection &egion 6/ STSU ,0.>5<
If our 2 B 0;; and if our sample mean deviates significantly from 0;; in the lower direction, we shall re!ect
H;, otherwise we shall accept H; at a certain level of significance. If the significance level in the given case
is kept at <W, then the re!ection region will be e&ual to ;.;< of area in the left tail as has been shown in the
above curve.
In case our H;/ 2 B 2 and Ha/ 2 O 2 we are then interested in what is known as one tailed test (right tail
and the re!ection region will be on the right tail of the curve as shown below/
4age 0:
"athematically we can state/
Acce$tance &egion )/ STSU 0.>5<
&e<ection &egion 6/ STSO 0.>5<
If our 2 B 0;; and if our sample mean deviates significantly from 0;; in the upward direction, we shall
re!ect H;, otherwise we shall accept the same. If in the given case the significance level is kept at <W, then
the re!ection region will be e&ual to ;.;< of area in the right,tail as has been shown in the above curve.
It should always be remembered that accepting H; on the basis of sample information does not constitute
the proof that H; is true. %e only mean that there is no statistical evidence to re!ect it, but we are certainly
not saying that H; is true (although we behave as if H; is true.
TESTS OF HYPOTHESES

Hypothesis testing helps to decide on the basis of a sample data, whether a hypothesis about the population
is likely to be true or false. *tatisticians have developed several tests of hypotheses (also known as the tests
of significance for the purpose of testing of hypotheses which can be classified as/
(a 4arametric tests or standard tests of hypotheses7 and
(b -on,parametric tests or distribution,free tests of hypotheses.
4arametric tests usually assume certain properties of the parent population from which we draw samples.
)ssumptions like observations come from a normal population, sample si$e is large, assumptions about the
population parameters like mean, variance, etc., must hold good before parametric tests can be used. ?ut
there are situations when the researcher cannot or does not want to make such assumptions. In such
situations we use statistical methods for testing hypotheses which are called non,parametric tests because
such tests do not depend on any assumption about the parameters of the parent population. ?esides, most
non,parametric tests assume only nominal or ordinal data, whereas parametric tests re&uire measurement
e&uivalent to at least an interval scale. )s a result, non,parametric tests need more observations than
parametric tests to achieve the same si$e of Type I and Type II errors.
The important parametric tests are 5,test, t,test, 6
2
test and F,test. )ll these tests are based on the
assumption of normality i.e., the source of data is considered to be normally distributed.
LIMITATIONS OF TESTS OF HYPOTHESES
0. The testing is not decision making itself. The tests are only useful aids for decision making. Hence
proper interpretation of statistical evidence is important to intelligent decisions.
1. Tests do not e+plain the reasons as to why do the difference e+ist, say between the means of two
samples. They simply indicate whether the difference is due to fluctuations of sampling or because
of other reasons.
3. 6esults of significance tests are based on probabilities and as such cannot be e+pressed with full
certainty.
METHODS OF DATA COLLECTION
%hile deciding about the method of data collection to be used for the study, the researcher should keep in
mind two types of data vi$., primary and secondary. The primary data are those which are collected afresh
and for the first time, and thus happen to be original in character. The secondary data are that which have
already been collected by someone else and which already has been passed through statistical processes.
The researcher would have to decide which sort of data he would be using (thus collecting for his study
and accordingly he will have to select one or the other method of data collection.
4age 0K
%omparison bet+een primary data and secondary data
46I")6X 9)T) *D82-9)6X 9)T)
0. These are first,hand original
data collected by research through various
methods.
These are second,hand ready,made data
collected by some other agency but not
by the researcher.
1. 8ollection is time and cost consuming. It is
complicated too.
Dasy and &uick collection at less cost.
3. *ources of primary data are the sampling units
chosen.
*ources could be internal or e+ternal
records.
5. 4rimary data are reliable and accurate when
properly collected.
*upplementary to primary data.
<. 4rimary data affect the research directly. They provide secondary information.
>. 4rimary data are latest useful for current
problems.
*econdary data are historical.
COLLECTION OF PRIMARY DATA
.. OBSER/ATION
2bservation is one of the methods of collecting data. It is used to get both past and current information. For
e+ample, instead of asking respondents about their current behaviour, we may observe it and record our
observations. )lthough it is not possible to observe past behaviour, we may observe the results of such
behaviour. In a way, secondary data reflect the results of the past behaviour of people as also of past
occurrences.
In marketing research, the observational method is not used fre&uently. )t the outset, let us see an e+ample
indicating how observation may be used in marketing research. 2ne of the factors influencing the sale of a
branded product is how readily it is kept in stock. )n interested manufacturer may send some observers to a
sample of stores to find out how fre&uently the product is out of stock. #ikewise, sales are also influenced
by its display position in a store. ) prominently displayed product will receive greater attention of the
people visiting the store as against another product displayed in an obscure corner.
Advantages o* o%servation
0. *ub!ective bias is eliminated, if observation is done accurately.
1. It enables a researcher to record behaviour as it occurs. The information obtained is not
complicated by either the past behaviour or future intentions or attitudes.
3. It can be used regardless of whether the respondent is willing to report or not. In a field survey, if
an enumerator comes across an unwilling and hostile respondent, he cannot collect the desired
information. ?ut, this problem does not arise at all in the case of direct observation.
4age 1;
5. It can be used even when it pertains to those who are unable to respond such as infants and
animals.
Li"itations o* o%servation
0. The information provided by this method is very limited.
1. The current behaviour of a person or a group of persons can only be observed. 2ne is not able to
observe the past behaviour nor can one observe a persons future behaviour because the act of
observation takes place in the present.
3. *ometimes unforeseen factors may interfere with the observational task. For e+ample,
inaccessibility of people to be observed.
5. The method is very slow and e+pensive. )s such, when a large number of persons are to be
contacted, it becomes unsuitable.
Methods of Observation
Structured+unstructured =%servation
*tructured observation is used when the research problem has been formulated precisely and the observers
have been told specifically what is to be observed. They may be given a simple form to record their
observations. They are considered appropriate in descriptive studies.
Anstructured observation implies that observers are free to observe whatever they think is relevant and
important. %hile structured observations are free from sub!ective bias, unstructured observations are
sub!ect to this limitation. The e+tent of the bias may vary to the e+tent an observation is unstructured.
Partici$ant+non $artici$ant o%servation
%hen the observer observes the group by becoming a part of the group so that he can e+perience what the
members of the group e+perience, the observation is called participant observation.
In non,participant observation, the observer observes as a detached emissary without any attempt on his
part to e+perience through participation what others feel.
There are several merits of the participant type of observation/
The researcher is enabled to record the natural behaviour of the group.
The researcher can even gather information which could not easily be obtained if he observes in a
disinterested fashion.
The researcher can even verify the truth of statements made by informants in the conte+t of a
&uestionnaire or a schedule.
?ut there are also certain demerits of this type of observation/
The observer may lose the ob!ectivity to the e+tent he participates emotionally7 the problem of
observation,control is not solved7 and it may narrow,down the researchers range of e+perience.
.isguised+undisguised =%servation
In the case of disguised observation, the sub!ects do not know that they are being observed. In some cases,
disguised observation may be made by the observer by posing as one of the shoppers who are being
observed. This type of observation is often preferred because it is feared that people may behave differently
when they know they are being observed.
Controlled+uncontrolled =%servation
2bservations in field studies are in their natural setting and are, therefore, called controlled observation.
The ma!or aim of this type of observation is to get a spontaneous picture of life and persons.
Ancontrolled observations are observations in a laboratory setting, on the other hand, enables the observer
to control e+traneous variables which influence the behaviour of people. 2bservational studies in
4age 10
laboratory settings enable the collection of data promptly and economically and in addition, permit the use
of more ob!ective measurements. *uch observation has a tendency to supply formali$ed data upon which
generali$ations can be built with some degree of assurance. The main pitfall of non,controlled observation
is that of sub!ective interpretation. Qenerally, controlled observation takes place in various e+periments that
are carried out in a laboratory or under controlled conditions, whereas uncontrolled observation is resorted
to in case of e+ploratory researches.
.irect+#ndirect =%servation
In the case of direct observation, the event or the behaviour of a person is observed as it occurs. In contrast,
indirect observation implies that some record of past behaviour is observed. In other words, the behaviour
itself is not observed, rather its effects are observed. )n observer engaged in indirect observation generally
looks for physical traces of behaviour or occurrence of an event. *uppose, he is interested in knowing about
the li&uor consumption of a household, he would like for empty li&uor bottles in the garbage. *imilarly, the
observer may seek the permission of the housewife to see the pantry. He may carry out a pantry audit to
ascertain the consumption of certain types of products. It may be noted that the success of an indirect
observation largely depends on how best the observer is able to identify physical traces of the problem
under study.
4u"an+"echanical =%servation
)nother way of classifying observations is whether they are made manually or by machines. "ost of the
studies in marketing research are based on human observation wherein trained observers are re&uired to
observe and faithfully record their observations.
In some cases, mechanical devices such as eye cameras and audiometers are used for observation. 2ne of
the ma!or advantages of electrical. mechanical devices is that their recordings are free from sub!ective bias.
)s against this, such observations may be less valid than human observations because the observers power
of integration can lead to a more valid evaluation of the observation.
+. INTER/IEW METHOD
The interview method of collecting data involves presentation of oral,verbal stimuli and reply in terms of
oral,verbal responses. This method can be used through personal interviews (like depth interviews, focus
group interviews and, if possible, through telephone interviews.
Personal intervie+s
4ersonal interview method re&uires a person known as the interviewer asking &uestions generally in a face,
to,face contact to the other person or persons.
The method of collecting information through personal interviews is usually carried out in a structured
way. )s such we call the interviews as structured intervie,s. *uch interviews involve the use of a set of
predetermined &uestions and of highly standardi$ed techni&ues of recording. Thus, the interviewer in a
structured interview follows a rigid procedure laid down, asking &uestions in a
form and order prescribed.
)gainst it, the unstructured intervie,s are characteri$ed by a fle+ibility of approach to &uestioning.
Anstructured interviews do not follow a system of pre,determined &uestions and standardi$ed techni&ues of
recording information. In a non,structured interview, the interviewer is allowed much greater freedom to
ask, in case of need, supplementary &uestions or at times he may omit certain &uestions if the situation so
re&uires. He may even change the se&uence of &uestions. ?ut this sort of fle+ibility results in lack of
comparability of one interview with another and the analysis of unstructured responses becomes much
more difficult and time consuming.
Depth intervie+s
4age 11
%hen a researcher is interested in in,depth investigation of perceptions, attitudes or motivations of the
respondents, an unstructured interview called 9epth Interview is conducted. 2n account of their fle+ibility,
such interviews help a researcher in asking probing &uestions to secure as much information as possible. )
procedure similar to that used by a psychiatrist is followed. The researcher asks such &uestions as are
appropriate without any formal &uestionnaire with him. For e+ample/
o %hat make of scooter did you buy(
o 9id you compare different scooters before buying a ?a!a!(
o %hat has been your e+perience with it(
o Is it easy to maintain(
o Is it reliable on busy roads(
o )re you satisfied with its performance(
Thus, the researcher will continue asking &uestions till he is satisfied.
The main advantages of depth interview are/
0. The researcher is able to access the information that he might not have otherwise got. *uch
interview offers fle+ibility to an interviewer because it is like an informal discussion between him
and the interviewee on the latters e+periences.
1. 9epth interviews may lead to motivational pattern wit respect to a particular brand or other action
under investigation.
3. It proves a strong stimulus to the insight of the interviewer.
9isadvantages of dept interview are/
0. 9epth interviews lack a systematically defined way of analy$ing the information obtained.
1. The information obtained is non,&uantifiable and is based on human !udgment that increases the
sub!ectivity.
3. It needs far more vigilance and training on the part of the interviewer. In the Indian marketing
research conte+t depth interviews lack popularity because of non,availability of professional
personnel.
5. This method also re&uires considerable amount of time in interviewing.
4ocus56roup intervie+s
In this method, the interviewer collects a small number (say about : people of representative consumers
for discussion on a particular sub!ect. ) group smaller than this proves to be insufficient for a focus group.
) group, say 0;,01 people, is regarded as too large as the people have to wait for long for their turn to
come.

'a<or advantages o* $ersonal intervie,s are as *ollo,s(
0 "ore information and that too in greater depth can be obtained.
1 Interviewer by his own skill can overcome the resistance, if any, of the respondents.
3 There is greater fle+ibility under this method as the opportunity to restructure &uestions is always
there, especially in case of unstructured interviews.
5 2bservation method can as well be applied to recording verbal answers to various &uestions.
< 4ersonal information can as well be obtained easily under this method.
> *amples can be controlled more effectively as there arises no difficulty of the missing returns7
non,response generally remains very low.
J The interviewer can usually control which person(s will answer the &uestions.
: The interviewer may catch the informant off,guard and thus may secure the most spontaneous
reactions than would be the case if mailed &uestionnaire is used.
K The language of the interview can be adapted to the ability or educational level of the person
interviewed and as such misinterpretations concerning &uestions can be avoided.
0; The interviewer can collect supplementary information about the respondents personal
characteristics and environment which is often of great value in interpreting results.
>ea1nesses o* $ersonal intervie, "ethod are as *ollo,s(
4age 13
0 It is a very e+pensive method, especially when large and widely spread geographical sample is
taken.
1 There remains the possibility of the bias of interviewer as well as that of the respondent7 there also
remains the headache of supervision and control of interviewers.
3 8ertain types of respondents such as important officials or e+ecutives or people in high income
groups may not be easily approachable under this method and to that e+tent the data may prove
inade&uate.
5 This method is relatively more,time,consuming, especially when the sample is large.
< The presence of the interviewer on the spot may over,stimulate the respondent, sometimes even to
the e+tent that he may give imaginary information !ust to make the interview interesting.
> Ander the interview method the organi$ation re&uired for selecting, training and supervising the
field,staff is more comple+ with formidable problems.
J Dffective interview presupposes proper rapport with respondents that would facilitate free and
frank responses. This is often a very difficult re&uirement.
For successful implementation of a personal interview/
0 the interviewers should be carefully selected, trained and briefed. They should be honest, sincere,
hardworking, and impartial and must possess the technical competence and necessary practical
e+perience.
1 occasional field checks should be made to ensure that interviewers are neither cheating, nor
deviating from instructions given to them for performing their !ob efficiently.
3 some provision should also be made in advance so that appropriate action may be taken if some of
the selected respondents refuse to cooperate or are not available when an interviewer calls upon
them.
5 effort should be made to create friendly atmosphere of trust and confidence, so that respondents
may feel at ease while talking to and discussing with the interviewer.
< the interviewer must ask &uestions properly and intelligently and must record the responses
accurately and completely. )t the same time, the interviewer
> must answer legitimate &uestion(s, if any, asked by the respondent and must clear any doubt that
the latter has. The interviewers approach must be friendly, courteous, conversational and unbiased.
J the interviewer must keep the direction of interview in his own hand, discouraging irrelevant
conversation and must make all possible effort to keep the respondent on the track.
(elephone intervie+s
This method of collecting information consists in contacting respondents on telephone itself. It is not a very
widely used method, but plays important part in industrial surveys, particularly in developed regions.
The chie* "erits o* such a syste" are(
0 It is more fle+ible in comparison to mailing method.
1 It is faster than other methods i.e., a &uick way of obtaining information.
3 It is cheaper than personal interviewing method7 here the cost per response is relatively low.
5 6ecall is easy7 callbacks are simple and economical.
< There is a higher rate of response than what we have in mailing method7 the non,response is
generally very low.
> 6eplies can be recorded without causing embarrassment to respondents.
J )t times, access can be gained to respondents who otherwise cannot be contacted for one reason
or the other.
: -o field staff is re&uired.
K 6epresentative and wider distribution of sample is possible.
.e"erits o* tele$honic intervie,s are as *ollo,s(
0 #ittle time is given to respondents for considered answers7 interview period is not likely to e+ceed
five minutes in most cases.
1 *urveys are restricted to respondents who have telephone facilities.
3 D+tensive geographical coverage may get restricted by cost considerations.
4age 15
5 It is not suitable for intensive surveys where comprehensive answers are re&uired to various
&uestions.
< Luestions have to be short and to the point7 probes are difficult to handle.
0. 1UESTIONNAIRE
This method of data collection is &uite popular, particularly in case of big en&uiries. It is being adopted by
private individuals, research workers, private and public organi$ations and even by governments. The
method of collecting data by mailing the &uestionnaires to respondents is most e+tensively employed in
various economic and business surveys.
(ypes of .uestionnaires
The &uestionnaires can be classified into two main types/
a3 Structured ?uestionnaire
) structured &uestionnaire is a formal list of &uestions framed so as to get the facts. The
interviewer asks the &uestions strictly in accordance with a pre,arranged order. If, for e+ample, the
marketing researcher is interested in knowing the amount of e+penditure incurred on different
types of clothing, i.e. cotton, woolen or synthetic, by different households classified according to
their income, he may frame a set of &uestions seeking this factual information. If the marketing
researcher appoints some interviewers to collect information on his behalf, the interviewers are
e+pected to adhere to the same order in asking &uestions as contained in the &uestionnaire.
) structured &uestionnaire can be of two types, namely, disguised and non+disguised. This
classification is based on whether the ob!ect or purpose of the survey is revealed or undisclosed to
the respondent. Thus, a structured,non,disguised &uestionnaire is one where the listing of
&uestions is in a prearranged order and where the ob!ect of en&uiry is revealed to the respondent.
"ost marketing research studies use this type of &uestionnaire. In the case of a structured,
disguised &uestionnaire, the researcher does not disclose the ob!ect of the survey. He feels that if
the respondent comes to know the ob!ect of the survey, he may not be ob!ective in giving the
necessary information and, as such, its purpose may be defeated. Luestionnaires of this type are
known as structured and non,disguised &uestionnaires.
)dvantages of structured &uestionnaire/
It facilitates the collection of information in a systematic and orderly manner as the
&uestions have been formulated in advance.
*ince the &uestions asked by each interviewer happen to be identical and are asked in the
same order, the information is generally not vitiated on account of the varying
characteristics of the different interviewers.
) structured &uestionnaire calls for a straightforward and simple approach on the part of
interviewers. )s such, even less &ualified interviewers can be deployed in canvassing
such a &uestionnaire.
*uch a &uestionnaire makes it far easier to edit, tabulate and interpret the data it contains.
) structured &uestionnaire can be conveniently pre,tested so that suitable modifications
can be made in the phraseology of &uestions or in their se&uence or both.
#imitations of a structured &uestionnaire/
This arises when the respondent is asked &uestions concerning personal or motivational
aspects.
%3 @on+structured ?uestionnaire
4age 1<
) non,structured &uestionnaire is one in which the &uestions are not structured and the order in
which they are to be asked from the respondent is left entirely to the researcher. He asks the
&uestions in the manner in which he deems fit in a particular situation. In fact, he may only have
certain main points on which he may develop the &uestions at the time of the actual interview. )s
it will be seen, a non,structured &uestionnaire is devoid of rigidity and allows considerable
freedom to the researcher is choosing the order as well as the specific content of &uestions.
-ormally, unstructured &uestionnaires are used in e+ploratory research when some respondents
are contacted. It is only subse&uently, on the basis of answers received, that a well,structured
&uestionnaire is developed. This is because the interviewer has a better understanding of the
problem on hand only after the e+ploratory research.
Luestionnaires of this type can be split into two sub,types. %here the purpose of the en&uiry is
disclosed to the respondents, the &uestionnaires are known as non+structured and non+disguised
while in other cases, the &uestionnaires are classified as non+structured and disguised
?uestionnaires.
)dvantages of unstructured &uestionnaire/
The interviewer is free to ask probing &uestions to get at the key motivational factor or
factors.
*uch &uestionnaires facilitate the conducting of interviews in an informal manner.
They also lend fle+ibility to the whole process of interviewing.
#imitations of unstructured &uestionnaire/
) point worth emphasi$ing is that in the case of unstructured &uestionnaires, the role of
the interviewer becomes far more important as compared to the one when a structured
&uestionnaire is used. In view of this, more capable interviewers are re&uired to handle
them, which raise the cost of the survey.
)lso, the interviewer needs more time per interview if it is unstructured. This also
enhances the overall cost of the field survey.
The researcher faces problems while editing and tabulating an unstructured &uestionnaire.
Designing a .uestionnaire
Luite often &uestionnaire is considered as the heart of a survey operation. Hence it should be very carefully
constructed. If it is not properly set up, then the survey is bound to fail. ) &uestionnaire should be designed
in such a manner that responses can be easily tabulated and translated into findings and recommendations
that satisfy the re&uired ob!ectives of the research pro!ect.
%hile designing a &uestionnaire, one must keep the following considerations in mind/
FI6*T, does it provide the necessary information that is re&uired for the research pro!ect(
*D82-9, is it designed for the convenience of the respondents(
THI69, does it meet coding, editing and data processing re&uirements of the research pro!ect(
Ste$s involved in develo$"ent o* a ?uestionnaire
S&e!.2 S!e-if3 #4)& ifo"m)&io #i** 5e so674&.
"ust be clear about the information you need to answer the issues to be addressed in the
research pro!ect.
9o not try to e+tract that information which respondents hesitate in providing either it is
too confidential or too private and personal to be shared with another person.
Try to ask only and the only information that are relevant for your research7 for that one
should have absolute clarity about the research problem.
4age 1>
S&e! +2 De&e"mie &3!e of 86es&io)i"e )% me&4o% of )%miis&")&io.
*tructured or unstructured(
2btaining information through well designed written &uestions or through showing
pictures and telling stories about these pictures, or collecting information verbally or in
writing.
Luestionnaire will be administered by self, by hired personnel.agency7 through mail7 or
telephonic interviews or through internet.email.
%hether interviews will be conducted in phases.
S&e! 02 De&e"mie &4e -o&e&s of i%ivi%6)* 86es&ios
How many &uestions need to be asked and what all to be asked in each &uestions.
%hether related &uestions should be put under one &uestion with various sub,&uestions or
should they be asked as main &uestions without any sub,&uestions.
-ot try to ask so many things !ust in one &uestionY
To determine the contents of &uestions, a researcher should address the following issues/
o Is the &uestion necessary(
o )re several &uestions needed instead of one(
o 9o the respondents have necessary information to provide(
o %ill the respondents like to provide information about the &uestions asked(
S&e! ,2 De&e"mie fo"m of "es!oses &o e)-4 86es&io
*hould the &uestions asked be in the form of open,ended &uestions or close,ended
&uestions( )n open,ended &uestion gives the respondent complete freedom to decide the
form, length and detail of the answer. 2pen &uestions are preferred when the researcher is
interested in knowing what is uppermost in the mind of the respondent.
*hould close,ended &uestions be in the form of dichotomous &uestions or multiple,choice
&uestions(
o )n e+ample of a dichotomous &uestion is/
9o you use tobacco in any way(
Aes @o
o )n e+ample of a multiple choice &uestion is/
%hich of the following brands do you use for washing clothes(
Sur* E-cel Ariel &in Tide >heel =thers
*caled 6esponse Luestions/ These &uestions are designed to capture the intensity of the
respondents answers.
S&e! 92 De&e"mie #o"%i7: !4")si7 of e)-4 86es&io
Ase simple words. )void technical !argons.
)void any kind of ambiguous words like 'often, 'generally, 'reasonably etc.
)void ambiguous &uestions.
#engthy &uestions should be avoided.
)void leading &uestions/ ) leading &uestion is one which attempts to guide the
respondents answer. For e+ample/
o %ould you like to vote for "r. 6ana *ingh, a man who has criminal background(
Xes.-o
o )re you in favour of a balanced budget( Xes.-o
4age 1J
)void generali$ations and estimates.
)void double barreled &uestions.
2ne should avoid combining two &uestions into one. For e+ample/
o %hich of the following modes of transport is cheaper and more convenient( (i Train
(ii ?us
It is &uite likely that according to the respondent both the attributes, namely cheapness
and convenience may not be applicable to either of the two modes of transport. 2ne may
be cheaper while the other may be more convenient.
S&e! '2 De&e"mie &4e se86e-e of 86es&ios
The &uestion,se&uence must be clear and smoothly,moving, meaning thereby that the
relation of one &uestion to another should be readily apparent to the respondent
Xou may start with simple and interesting &uestions.
*tart with broad &uestions and narrow down to more specific &uestions , it is called
Funnel A$$roach.
?eginning and ending of a &uestionnaire may have simple and less demanding &uestions
and in mid, have more demanding and serious &uestions.
8are should be taken to sustain the interest of the respondent until the last so that he is
able to answer speciali$ed &uestions in a normal manner without fatigue and indifference.
Dnsure proper placing of branching &uestions so as to ensure proper skip pattern.
4lace difficult and sensitive information at the end of the &uestionnaire.
4ut personal information first so as to gain the familiarity with the respondents.
S&e! ;2 De&e"mie &4e *)3o6& of 86es&io)i"e
4hysical appearance of a &uestionnaire may affect the cooperation and thus, the &uality of
responses one can obtain.
Dnsure proper si$e and layout of a &uestionnaire.
) &uestionnaire should not give a feeling of cluttered or crowded appearance.
It should be properly stapled.bounded and handy.
S&e! <2 Ree=)mie s&e!s .>; )% "evise if e-ess)"3
S&e! ?2 P"e&es& 86es&io)i"e )% "evise if e-ess)"3
4re,testing implies that the &uestionnaire is tried out on a few respondents and their
reaction to the &uestionnaire is observed. It helps the researcher decide whether any
changes in the &uestion content or wording are re&uired. If so, specific changes that are
desirable are ascertained and incorporated in the &uestionnaire.
Mail 7uestionnaire
?roadly, there are three types of survey namely, personal, mail and telephone. )s far as the telephone
survey is concerned, it is not commonly used in India. )s such, the personal interview and mail survey are
the only two methods. *ince a mail survey needs a &uestionnaire which should have some additional
characteristics, it is necessary to look into this aspect in some detail.
The merits of collecting information by mailing &uestionnaires are as follows/
0. It is much easier to approach a large number of respondents spread all over the country through
post.
1. It is free from the bias of the interviewer7 answers are in respondents own words.
4age 1:
3. 6espondents have ade&uate time to give well thought answers. The &uality of answers is,
therefore, likely to be superior to that obtained in the personal interview.
5. 6espondents, who are not easily approachable, can also be reached conveniently.
<. *ince a large number of respondents can be approached all over the country through mail in a
short period, a mail &uestionnaire saves a lot of time in collecting the desired information.
>. There is a good deal of saving in the total cost of a mail survey as cost of travelling, boarding and
lodging of interviewers is not to be incurred. This enables the organi$ers of the survey to complete
the investigation within a limited budget. .
J. "ail &uestionnaires also avoid the bias arising from any inhibitions in answering &uestions. In
particular, when &uestions are of a personal nature, the respondents may hesitate to answer them in
the presence of an interviewer. This type of inhibition will not be there if the mail survey is
undertaken.
:. Finally, mail &uestionnaires will not have the problem of non,contacts in the strict sense as might
be the case in personal interviews when the interviewer finds that the respondent, being away from
home, is not available.
The main demerits of this system can also be listed here/
0. It can be used only when respondents are educated and cooperating.
1. The control over &uestionnaire may be lost once it is sent.
3. There is also the possibility of ambiguous replies or omission of replies altogether to certain
&uestions7 interpretation of omissions is difficult.
5. This method is likely to be the slowest of all.
<. It is not suitable when &uestions are difficult and complicated. In such a case, the help of
interviewers is re&uired to offer some introductory e+planation to the respondent.
>. %hen the researcher is interested in the spontaneous answers of the respondent or his own answers
uninfluenced by others who may influence his thinking, the mail &uestionnaire is inappropriate.
J. In case of mail &uestionnaires, it is not possible to verify whether the respondent himself or herself
has filled in the &uestionnaire.
:. The researcher has to accept the answers as they are provided in the mail &uestionnaire. In case
there is any inconsistency or ambiguity in the answers, it will be difficult for the researcher to
make use of such a &uestionnaire. He cannot further probe into the same to get some additional
information or to remove the inconsistency or ambiguity.
K. ) mail &uestionnaire does not allow the researcher to supplement the information by his personal
observations. That will be possible only when the &uestionnaire is canvassed by him personally.
0;. This method is likely to be the slowest of all.
%omparison bet+een personal intervie+s and mail .uestionnaires
4D6*2-)# I-TD6HID%* ")I# LAD*TI2--)I6D*
0. ) limited number of people can be approached at a
time.
"uch easier to approach a large number of
respondents spread all over the country through
post.
1. They are likely to be affected by the personal
preferences or dislikes of the individual
interviewer.
Free from any interviewers bias and errors.
3. Luestions are to be answered on the spot. The &uality of answers is likely to be superior as
respondents can think over the &uestion leisurely
at home before giving his answer.
5. Its a time consuming process. *ince a large number of respondents can be
approached all over the country through mail in
a short period, a mail &uestionnaire saves a lot of
time in collecting the desired information.
<. 8ostlier than mail &uestionnaires. There is a good deal of saving in the total cost of
a mail survey as cost of travelling, boarding and
4age 1K
lodging of interviewers is not to be incurred.
>. 6espondents may hesitate to answer certain
&uestions in the presence of an interviewer.
They avoid the bias arising from any inhibitions
in answering &uestions.
J. Faces the problem of non,contact when the
interviewer finds that the respondent, being away
from home, is not available.
"ail &uestionnaires will not have the problem of
non,contacts.
,. SCHEDULES
This method of data collection is very much like the collection of data through &uestionnaire, with little
difference which lies in the fact that schedules (proforma containing a set of &uestions are being filled in
by the enumerators who are specially appointed for the purpose. These enumerators along with schedules
go to respondents, put to them the &uestions from the proforma in the order the &uestions are listed and
record the replies in the space meant for the same in the proforma.
Dnumerators e+plain the aims and ob!ects of the investigation and also remove the difficulties which any
respondent may feel in understanding the implications of a particular &uestion or the definition or concept
of difficult terms. The enumerators should be trained to perform their !ob well. Dnumerators should be
intelligent and must possess the capacity of cross e+amination in order to find out the truth. )bove all, they
should be honest, sincere, and hardworking and should have patience and perseverance.
%omparison bet+een .uestionnaires and schedules
LAD*TI2--)I6D* *8HD9A#D*
0. The &uestionnaire is generally sent through
mail to the informants.
They are generally filled out by the research worker
or the enumerator.
1. To collect data through &uestionnaire is
relatively cheap and economical.
To collect data through schedules is relatively more
e+pensive since considerable amount of money has
to be spent in appointing enumerators and in
imparting training to them.
3. -on,response is usually high in case of
&uestionnaire as many people do not respond
and many return the &uestionnaire without
answering all &uestions.
-on,response is generally very low in case of
schedules because these are filled by enumerators
who are able to get answers to all &uestions.
5. ?ias due to non,response often remains
indeterminate.
9anger of interviewer bias and cheating is always
there.
<. Identity of respondent is unknown. Identity of respondent is known.
>. Its likely to be very slow since many
respondents do not return
the &uestionnaire in time despite several
reminders.
Information is collected well in time as they are
filled in by enumerators.
J. 4ersonal contact with respondents is generally
not possible.
9irect personal contact is established with
respondents.
:. 8an be used only when respondents are literate
and cooperative.
Information can be gathered even when the
respondents happen to be illiterate.
K. %ider and more representative distribution of
sample is possible.
9ifficulty in sending enumerators over a relatively
wider area.
0;
.
6isk of collecting incomplete and wrong
information is relatively more.
Information collected is generally complete and
accurate as enumerators
can remove the difficulties, if any, faced by
respondents in correctly understanding the
&uestions.
00
.
The success of &uestionnaire method lies more
on the &uality of the &uestionnaire itself.
The success of schedules depends upon the honesty
and competence of enumerators.
4age 3;
01
.
2bservation is not possible in this case. )long with schedules, observation method can also
be used.
CHOICE OF SUR/EY METHOD
%omparison of Survey Methods
Criteria Personal
survey
'ail
survey
Tele$hone
survey
0
.
4robable response rate Fair 4oor Qood
1
.
Time re&uired *low *low Fast
3
.
8ost High #ow #ow
5
.
8ontrol of sample Qood 4oor D+cellent
<
.
Luantity of information Qood #imited #imited
>
.
Luality of information Qood Fair D+cellent
J
.
*upervision of field work Fair -one D+cellent
:
.
Hersatility D+cellent Fair Qood
COLLECTION OF SECONDARY DATA
*econdary data is data that has been collected by others for another purpose. It is important to us because it
can save considerable time and effort in solving the research problem at hand. 6esearchers should always
conduct a thorough data search in the development and e+ecution of any research programme.
USES OF SECONDARY DATA
The uses of secondary data can be conveniently arranged into the following four categories/
0. Pro%le" recognition/ ) constant monitoring of secondary data can provide the impetus for problem
recognition. "oreover, e+ploratory studies maybe initiated to identify future opportunities or weaknesses in
the firm=s current operations.
1. Pro%le" clari*ication/ *econdary data can also fruitfully be used to help clarify the specific problem that
we may be facing. Here, clarification usually means making the decision problem more reachable by
delineating the components of the situation. )nother related use of secondary data is that they can help plan
the study design and provide information to write the research proposal.
3. For"ulation o* *easi%le alternatives/ )lternatives must e+ist before decision,making can take place.
*econdary data are very useful in generating viable alternatives to solve problems. The multiplicity of data
sources, research approaches and managerial styles usually lead to a number of possibilities which should
be e+amined by the researcher.
4age 30
5. Pro%le" solution/ -ot only are secondary data helpful in the definition and development of a problem,
they are often sufficient in and of themselves, to generate a problem=s solution.
AD/ANTAGES OF SECONDARY DATA
0. The ma!or advantage of secondary data is economy. )s the data are already available, they can be
obtained at a relatively low cost.
1. The secondary data can be obtained &uickly.
3. The secondary data enable the researchers to identify the deficiencies in the data and make
primary data collection specific.
5. They are useful in the case of e+ploratory researches as they provide increased understanding of
the problem.
<. They act as a basis for comparison after primary data are collected.
>. They provide for data which cannot obtained by the typical organi$ation like census reports.
LIMITATIONS OF SECONDARY DATA
0. The available data may not suit the current purpose of research, due to incompleteness,
generalities and so on.
1. Information may be outdated or obsolete.
3. The methodology used in collecting the data such as the sample si$e, date of the research, etc.,
may be unknown.
5. )ll the findings of a research study may not be made public.
<. 8onflicting data may e+ist.
>. It may be difficult to determine the accuracy of the secondary data.
E/ALUATING SECONDARY DATA
The researcher before using any secondary data has to satisfy himself as to their accuracy and reliability.
He has to evaluate secondary data on the following considerations/
0. )vailability/ )t times, one may find that secondary data are !ust not available on the problem at
hand. The secondary data must be available for use.
1. 6elevance/ 6elevance means that the data available must fit the re&uirements of that problem.
This would cover several aspects. First, the unit of measurement should be the same as that in the
marketing problem. *econd, the concepts should be the same as are envisaged in the problem.
)lso, the data should not be obsolete.
3. )ccuracy/ In this regard, one should consult the original source. This would not only give
comprehensive information but would also indiacte the conte+t in which data have been collected,
the procedure followed and the e+tent of care e+ercised in their collecction.
5. *ufficiency/ Finally, the data should be sufficient. If the data are inade&uate, then compliance with
the preceding re&uirements will be in vain.
4age 31
SOURCES OF SECONDARYDATA

Internal *ources D+ternal *ources
)ccounting *alesforce Internal "iscellaneous
6ecords 6eports D+perts 6eports
Qovernment -on,government
4ublications 4ublications
6egular publications )d hoc publications Aniversities
#ike census, !ournals like en&uiry reports 8onsumer
Industry Z research
Trade association agencies

*yndicate International
*ervices publications

Internal sources
Internal sources can be classified under four broad categories,accounting records, sales force reports,
miscellaneous reports and internal e+perts.
Accounting records( These generate a good deal of data. )s profits are based on sales, sales invoice is a
good source. -ormally, a sales invoice includes name of the customer, address of the customer, items
ordered, &uantities ordered, &uantities shipped, discounts allowed, price charged, total amount of sales and
the date of sales. It often contains information on sales territory, sales representative and warehouse from
which the product was dispatched. *uch information can be e+tremely useful in undertaking a detailed
analysis of sales by product, customer, industry, geographic area, sales territory and sales representative.
8ompared with corresponding data on costs, it can indicate the level of profits (or loss for each product.
This apart, data on advertising e+penditure along with the time period would also be available.
Sales *orce re$orts( This source can provide very useful marketing information but somehow it has
remained largely untapped. This is because sales persons may not be giving detailed reports. In order to
ensure that this source is more useful, it is necessary to organi$e the system properly. It should be a simple
process of reporting the information. *ales persons may be encouraged to provide accurate and
comprehensive information. *ome incentive may be given to those who report accurately and ade&uately.
'iscellaneous re$orts( )ny studies done earlier on marketing problems of the company, special audit, etc.
come in this category. *uch reports on varying sub!ects should be properly maintained and easily accessible
when re&uired.
E-$erts( D+perts working in the company can also be a good source of internal data. D+ecutives working as
product managers, marketing research managers, public relations personnel and advertising personnel have
speciali$ed knowledge relevant to marketing problems. However, this source is least tapped. ) limitation of
4age 33
this source is that information is in the e+perts mind and not on paper. The e+perts can provide useful
information or ideas on a given marketing problem whenever a serious discussion is held in a meeting.
#8ternal Sources
The e+ternal secondary data do not originate in the firm and are obtained from outside sources. It may be
noted that secondary data can be collected from the originating sources or from, secondary sources. For
e-a"$le) the 2ffice of the Dconomic )dviser, Qovernment of India, is the originating source for the data
on wholesale prices. In contrast, a publication such as the &eserve /an1 o* #ndia /ulletin containing some
parts of the series of wholesale prices is a secondary source.
Qenerally, the originating source of e+ternal secondary data should be preferred on account of several
reasons. First, the originating source is more likely to e+plain the ob!ect and procedure of data collection.
*econd, the originating source is more likely to present all the data, whereas a secondary source may
present a part of such data, depending on its re&uirement or convenience. Finally, the originating source
would be more accurate as each additional repeating source of secondary data presents another possible
source of error.
Bovern"ent Pu%lications( ) large bulk of secondary data useful to a marketing researcher is found in
various government publications. To begin with, the 6egistrar Qeneral of India conducts a population
census throughout the country every ten years and brings out demographic data in voluminous reports.
These publications provide perhaps the most basic source of information useful to the marketing researcher.
The data relate to various characteristic such as the break,up of population by se+, rural urban residence,
age, education and occupation. %hile it is true that these statistics are available only decennially, they are
the most authentic and are often used as the basis for pro!ection for future years.
The Central Statistical =rganiCation DCS=3 brings out statistics of national income. Its ma!or
publication, '-ational )ccounts *tatistics is brought out once a year and contains estimates of
national income for several years. The figures are given separately for ma!or economic activities
such as agriculture, industry, trade, transport, etc. ?esides this, the estimates of saving, capital
formation and consumption e+penditure, together with national and public sector accounts are
given.
The .irector Beneral o* Co""ercial #ntelligence, Qovernment of India, brings out from 8alcutta,
monthly statistics of the foreign trade of India. The statistics are contained in two separate
volumes one for the e+port trade and the other for the import trade. These statistics are compiled
on a very comprehensive basis, covering a very large number of products and are e+tremely useful
in undertaking region wise, country wise or product wise studies on the prospects of foreign trade.
They also provide historical data over a long period, thus enabling the researchers to study the
changing composition of Indias foreign trade over a chosen period.
The >holesale Price #nde- nu"%ers are constructed by the 2ffice of the Dconomic )dviser,
"inistry of 8ommerce and Industry, Qovernment of India. This is a weekly series and it is revised
from time to time so as to make it representative of all the products. The present series uses 0KK3,
K5 as the base year. The products covered are food articles, food grains, non,food grains, minerals,
fuel, power, light and lubricants, various manufactured products such as te+tiles, chemicals, metal,
machinery and transport e&uipment.
All+#ndia Consu"er Price #nde- nu"%ers for (a industrial workers, (b urban non,manual
employees, and (c agricultural labourers.
4age 35
*ome other official publications include the/
(i /asic Statistics relating to the #ndian Econo"y) which is an annual publication of the
4lanning 8ommission. It contains data on various aspects of the economy for several
years.
(ii &eserve /an1 o* #ndia /ulletin) which is a monthly !ournal dealing with all aspects of the
economy in general, and currency and finance in particular.
(iii Currency Finance &e$ort) which is an annual publication brought out by the 6eserve
?ank of India.
(iv The Econo"ic Survey) which is an annual publication of the 9epartment of Dconomic
)ffairs, "inistry of Finance, Qovernment of India. It is published on the eve of the
presentation of the national budget and contains a detailed review of the different sectors
of the economy. 9etailed statistics are also given in the publication.
(v Agricultural Situation in #ndia) which is a monthly !ournal of the 9irectorate of
Dconomics and *tatistics of the "inistry of )griculture, Qovernment of India. It contains
current statistics and articles on the assessment of the agricultural situation in the country
and the states.
(vi The #ndian La%our Eournal) which is a monthly !ournal of the #abour ?ureau, publishes
statistics on price indices, employment, wages and earnings, absenteeism, etc.
(vii The #ndian La%our Aear /oo1) which is an annual publication, contains detailed statistics
on wages and earnings, cost of living, industrial relations, labour welfare and industrial
housing.
(viii *tate Statistical A%stracts and .istrict Statistical 4and%oo1s) which are published by
various *tate *tatistical ?ureau, publish statistical abstracts for their states.
@ational Sa"$le Survey D@SS3, which was set up by the Qovernment of India in the "inistry of
4lanning in 0K<;. The ob!ective of setting up this organi$ation was to obtain social, economic,
demographic, industrial and agricultural statistics on a comprehensive and continuing basis.
@on+govern"ent Pu%lications(
@ational Council o* A$$lied Econo"ic &esearch D@CAE&3, which is a premier research institution
in applied economics in India. The -8)D6 studies are EThe Qreat Indian "arketG (1;;< and the
Qreat Indian "iddle 8lass (1;;5. The QI" gives information on consumer purchasing patterns
and product preferences such as/
%ho owns what, where and why(
%hat products are consumers buying(
%hat does the future hold for your market(
How are preferences likely to change in urban and rural areas by 1;0;(
%here can you source the most recent information on products after census 1;;0(
The 'ar1eting >hite%oo1 published by ?usiness %orld. The publication covers a wide variety of
marketing data under eight different sections, each of which contains the latest data relating to that
section. For e+ample, data on consumer psychographics, emerging consumer markets, product
penetration in Indian households, emergence of rural markets, etc. are given in relevant sections.
The &.F S,a"y //.= Buide to Ur%an 'ar1ets provides a rigorous e+planatory framework. It
provides a simple way of inde+ing market potential across different towns, aggregated into states
and regions. It has developed the market intensity inde+ ("Il, which is a good indicator of the
character of the market.
?esides these three ma!or sources of marketing information, there are a number of private
organi$ations which bring out statistics. For e+ample, the Indian 8otton "ills Federation brings
4age 3<
out statistics on the cotton te+tile industry. #ikewise, the annual report of the ?ombay "ill
2wners )ssociation contains the latest statistics on the working of the member units.
Federation o* #ndian Cha"%ers o* Co""erce and #ndustry DF#CC#3 is associated 8hamber of
8ommerce and Industry of India, Indo,)merican 8hamber of 8ommerce, etc. These chambers or
their federations occasionally bring out memoranda dealing with a specific industry and its
problems.
Syndicated Services( *yndicated services are provided by certain organi$ations, which collect and tabulate
marketing information on a continuing basis. 6eports based on the marketing information collected by such
organi$ations are sent periodically (weekly, monthly, or &uarterly to clients who are subscribers.
*yndicated services are normally designed to suit the re&uirements of many individuals or firms. *uch
services are particularly useful in the spheres of TH viewing, maga$ine readership, and the movement of
consumer goods through retail outlets.
*yndicated services may be regarded as an 'intermediate source falling between the primary and
secondary sources. This is because they possess the characteristics of both types of sources. )s such
services are based on data collected specifically by the organi$ation from original sources and since they
are current in nature, the data may be called primary. 2n the other hand syndicated data may be regarded as
secondary data as they are collected by an independent organi$ation for purposes of sale to a large number
of firms.
Consu"er &esearch Services( It goes to the credit of these agencies that some of the studies conducted by
them were of pioneering nature, involving the development of appropriate concepts as well as sampling and
measurement techni&ues. They have a large field set,up supported by full,time investigators in different
parts of the country having several years of professional e+perience.
Pu%lications o* #nternational =rganiCation( *o far the discussion was confined to national organi$ations.
There are several international organi$ations that publish statistics in their respective areas. *ome of these
organi$ations publish data on India along with that of other countries. The main advantage of such data is
that international comparisons can be drawn.
REPORT WRITING
)fter the data has been analy$ed and the conclusions reached, the marketing researcher has to report his
findings to the management. The reporting has to be done in a clear manner so that the chances of any
misunderstanding can be minimi$ed.
SIGNIFICANCE OF REPORT WRITING
6esearch report is considered a ma!or component of the research study for the research task remains
incomplete till the report has been presented and.or written. )s a matter of fact even the most brilliant
hypothesis, highly well designed and conducted research study, and the most striking generali$ations and
findings are of little value unless they are effectively communicated to others. The purpose of research is
not well served unless the findings are made known to others. )ll this e+plains the significance of writing
research report.
WAYS OF REPORTING
!" Oral reporting
4age 3>
)n oral report is any presentation of information through the spoken words. If the researcher has
been asked to make an oral presentation of his research findings, he should bear in mind a few
ma!or considerations/
a First, he should know the audience to whom he is to report.
b 2ral reporting should be properly planned. To begin with, the reporter should be clear
about the ob!ectives of the report.
c )fter the reporter has decided the ob!ective of his report, he should gather the necessary
information.
d 2nce the information is ready, the reporter has to organi$e it in a logical manner. The
sub!ect matter should be divided into meaningful and comparable parts. *imple ideas
should precede comple+ and difficult ones.
e )nother point worth considering is the suitability of language. The reporting should be
done in a simple and convincing manner.
f In order to make the communication effective, it may be advisable for the reporter to use
visual aids such as charts, diagrams, tables, pictures, posters, blackboards, slides and
movies. However, e+cessive use of visual aids may sometimes spoil the overall &uality of
presentation, therefore, it is necessary to ensure their !udicious use.
g *ometimes, it may be advisable to distribute a handout containing statistical data or
charts to the audience. This would facilitate both the listener and the reporter. The latter
can refer to the figures or charts in the handout as he proceeds with his reporting.
h %hen an oral summary of a written report is to be presented, the reporter should e+ercise
great care in the preparation of the summary. )ll ma!or points must be covered and
e+cessive details avoided otherwise the focus will be lost. The reporter should ensure that
within the time allotted to him, he covers all ma!or points including his
recommendations.
$" ,ritten reports
%ritten reports are preferred over oral reports because of the following few reasons/
a First, unlike written reports, oral reports lack visual advantages. 8harts, diagrams or
pictures can be used to a limited e+tent, though, to stress particular points in oral
presentation, the reporter can use pauses and volume emphasis.
b *econd, the pace of presentation cannot be controlled and regulated by anyone from an
audience, who is being presented an oral report. In contrast7 whenever a certain point is
not clear in a written report the reader can read it two or more times, if need be slowly
and more carefully.
c Third, since a written report is bound to receive considerable attention and scrutiny from
the readers the report,writer is likely to work hard to produce an accurate report of a high
&uality. In contrast, an oral report will not be so precise nor will the reporter give as much
time in its presentation since it cannot be sub!ected to the same degree of scrutiny as
written reports because of the e+tremely limited time at the disposal of the audience for
indicating its reaction.
d %ritten reports can be used for future references. In contrast, oral reports can also be
stored if recorded, but the recorded tapes become unusable after a limited time if not used
regularly.
e ?ecause written reports can be stored, they can also be used as evidences at times, if
necessary.
Ty$es o* ,ritten re$orts
B)sis T3!es
4age 3J
Time interval %eekly reports
"onthly reports
Luarterly reports
)nnual reports etc.
Functional basis Informational report is a factual report on a particular sub!ect and it
neither contains an analysis nor conclusions.
D+amination report in addition to giving facts on a sub!ect matter, it
analyses these facts. It does not specify conclusions or recommendations.
)nalytical reports not only presents and analyses data but it also draws
conclusions and, if necessary, makes recommendations.
*ub!ect matter or field "arketing reports
Finance reports
H6 reports
IT reports
2perations reports etc.
#ength *hort,length reports like memoranda, letters etc.
#ong,length reports
6elationship between
the reader and writer
)dministrative report is written within an organi$ation.
4rofessional report is submitted by an outside e+pert to an organi$ation.
Independent reports are generally prepared by non,profit research
organi$ations and published for the benefit of the public.
Dmployment status of
their authors
4rivate reports that are prepared in a business organi$ation.
4ublic reports that are prepared in a government or public institution.
Formality ) formal report is written befitting the re&uirements of a formal occasion
like a report prepared by an e+ecutive for the top management may be a
formal report.
)n informal report is worded to suit an informal occasion like a report
prepared by one employee for the information of his colleagues may be
an e+ample of an informal report.
PREPARATION FOR REPORT WRITING
Research Report 4ormat
) research report can be written in a number of ways. However, three formats are generally followed. First,
the report may use a logical $attern, which implies that the findings are presented in inductive order, i.e.,
moving from specific to general. *econd, the report may follow a $sychological $attern which is almost
inverse of the preceding pattern. In this format, the most critical information, i.e., the conclusion, is
provided first, after which follow the findings supporting the conclusion. Third, the report may use a
chronological *or"at wherein information is given along the time dimension, i.e., things which happened
earlier precede those which happened later. Having decided the format to be used in a research report, the
researcher has now to prepare the report outline in accordance with that format.
Report Outline
?efore attempting to write any report, the researcher must prepare a report outline. 2utlines are the
framework upon which long written works are constructed. )n outline to the writer is what the blueprint is
to the construction engineer. %hile preparing an outline, it must be noted that it should not be restrictive
and rigid. There should be fle+ibility built into it so that if subse&uently a change is needed, it can be easily
introduced.
) typical report outline based on the logical format is as follows/
1. Title $age
The title page should indicate the topic on which the report has been prepared, the person or agency
who has prepared it, the person or agency for whom it has been prepared and the date of submission
(or completion of the report.
4age 3:
2. Letters o* authoriCation
This indicates the official who has authori$ed or sponsored the research and the official to whom the
report is addressed.
. Ta%le o* contents) statistical ta%les) charts and illustrations
The table of contents should indicate the main parts, divisions or sections of the report, the chapter
headings along with the page numbers. %hen a report contains a number of statistical tables, charts,
figures or illustrations, a separate table for each of these categories should be given immediately after
the table of contents.
!. #ntroduction
The introduction deals with the genesis of the report. It e+plains why the study was undertaken, the
statement and formulation of the problem, formulation of hypotheses, if any, and the scope of the
study. It is desirable to give a review of the related studies done earlier so that the reader may take the
present report in the proper perspective.
G. 'ethodology
The report should describe the methodology used in studying the problem. If a sample survey has been
undertaken, the report should indicate the type of sample design used, its si$e and the procedure used
to draw the sample, and the instrument used to collect the data with the number of research workers
employed in the field work, supervisors. ) mention of the e+tent of precision achieved and the
methods used for handling any special problems during the course of the study is advisable.
H. Findings
This section of the report contains the result of the research and is perhaps the largest section of the
report. The researcher should ensure that the findings are presented in proper se&uence and in such a
manner that the reader has no difficulty in understanding them.
I. Li"itations
The report should also point out the main limitations of the research report therein. This will be helpful
to the reader who can form his own opinion as to how far the results are reliable.
J. Su""ary and Conclusions
This section contains a summary of the report as well as the main conclusions. The latter part is based
on the findings given earlier. To facilitate the reader, the conclusions reported in this section should
give references to pages, paragraphs or tables in the findings section.
K. &eco""endations
It may not be necessary for all reports to contain recommendations. However, when a study has been
commissioned by management which e+pects that the researcher will recommend certain measures, it
is necessary that these be made in clear language.
1;. A$$endices
)n appendi+ gives supplementary information which supports the body of the report but which cannot
be given within it. If it is given within the main report, it may distort the focus on the main theme and
confuse the reader. Qenerally, large and comple+ statistical tables, technical notes, copies of the
&uestionnaires used, instructions to field staff and any other material used as supporting evidence for
the write,up in the report are given in the form of appendices.
11. /i%liogra$hy
?ibliography shows the title of the books.reports used in their preparation, names of authors,
publishers, the year of publication and, if possible, the page numbers. In addition, similar information
is given in respect of articles.papers. The bibliography should be arranged in an alphabetical order.
4age 3K
12. #nde-
Finally, an inde+ showing the various topics and the relevant page numbers in the report should also be
given. It should be prepared on an alphabetical basis. However, an inde+ may not be necessary in case
the report is very short.
Preparation of a rough draft
This follows the logical analysis of the sub!ect and the preparation of the final outline. *uch a step is of
utmost importance for the researcher now sits to write down what he has done in the conte+t of his research
study. He will write down the procedure adopted by him in collecting the material for his study along with
various limitations faced by him, the techni&ue of analysis adopted by him, the broad findings and
generali$ations and the various suggestions he wants to offer regarding the problem concerned.
Re+riting and polishing of the rough draft
The careful revision makes the difference between a mediocre and a good piece of writing. %hile rewriting
and polishing, one should check the report for weaknesses in logical development or presentation. In
addition the researcher should give due attention to the fact that in his rough draft he has been consistent or
not. He should check the mechanics of writing,grammar, spelling and usage.
Preparation of the final bibliography
-e+t in order comes the task of the preparation of the final bibliography. The bibliography, which is
generally appended to the research report, is a list of books in some way pertinent to the research which has
been done. It should contain all those works which the researcher has consulted. The bibliography should
be arranged alphabetically and may be divided into two parts7 the first part may contain the names of books
and pamphlets, and the second part may contain the names of maga$ine and newspaper articles.
,riting the final draft
This constitutes the last step. The final draft should be written in a concise and ob!ective style and in simple
language, avoiding vague e+pressions. %hile writing the final draft, the researcher must avoid abstract
terminology and technical !argon. Illustrations and e+amples based on common e+periences must be
incorporated in the final draft as they happen to be most effective in communicating the research findings
to others. ) research report should not be dull, but must enthuse people and maintain interest and must
show originality.
PRECAUTIONS FOR WRITING RESEARCH REPORTS
%hile preparing the report, points to be kept in mind/
a 6eports must be written ob!ectively. 2b!ectivity is a prere&uisite of good research work. )bsolute
words such as 'always and 'never, for e+ample, should be avoided. The researcher should not impose
his viewpoint on the reader7 he should narrate the facts and draw such inferences as can be drawn.
b The report must be written in a concrete style. It must contain necessary statistics to support the thesis.
2ne should avoid the use of such vague words as 'several people, 'a large number of respondents,
and =often=.
c The report must be organi$ed and coherent, which can be ensured by formulating a research outline
prior to writing the report. 9ifferent sections of the report must be given in a proper se&uence so that
the reader is taken logically from beginning to end.
d %hile determining the length of the report one should keep in view the fact that it should be long
enough to cover the sub!ect but short enough to maintain interest.
e There must be absolute clarity in the presentation of ideas. 8larity in writing is not possible in !ust one
draft. The writer may have to revise the draft once, twice or even several times to make it lucid and
understandable.
4age 5;
f The report should be neat and tidy. %herever possible, charts and diagrams should be given. The
write,up should be divided into different paragraphs and suitable headings should be provided.
g 6eaders are often interested in ac&uiring a &uick knowledge of the main findings and as such the report
must provide a ready availability of the findings. For this purpose, charts, graphs and the statistical
tables may be used for the various results in the main report in addition to the summary of important
findings.
h ) research report should show originality and should necessarily be an attempt to solve some
intellectual problem. It must contribute to the solution of a problem and must add to the store of
knowledge.
E/ALUATION OF THE RESEARCH REPORT
)fter the report has been submitted by the researcher, he should try to get the feedback on the same.
Feedback will enable him to know the deficiencies of his report, both in regard to the sub!ect,matter and the
write,up.
)part from this feedback, a detailed evaluation of the research pro!ect should be undertaken. To begin with,
one has to ask a very important &uestion/ 9oes the report fulfill the ob!ectives set out for research(
*ometimes, it has been noticed that while taking up a research pro!ect some high,sounding ob!ectives are
included in it, but by the time the research is completed, one or more such ob!ectives are either overlooked
or diluted to a large e+tent. The subse&uent &uestions to be asked should be in conformity with the research
process.
For e-a"$le(
%ere the data sought in conformity with the actual re&uirements of the study(
%as the selection of the instrument used for the collection of data appropriate(
%as the data collection instrument properly designed(
%as the sample design used in the survey appropriate( %as the sample si$e
ade&uate(
%as there sufficient control in the survey when data collection was in progress(
%ere the techni&ues used in the analysis of data appropriate(
%as the interpretation of data ob!ectively done(
How was the written report( 9id it faithfully report the findings of the study(
ONE OF THE WIDELY USED RESEARCH DESIGN IN SOCIAL SCIENCES TO COLLECT
DATA IS
SUR/EYS
Ho# &o -o**e-& ifo"m)&io ...
) researcher can collect information
either
by observation
or
by asking.
%hen he.she ass for information, we say that he.she is conducting a survey.
SUR/EY ...
4age 50
) 'survey is a techni&ue of gathering information/
by &uestioning,
those in individuals who are the ob!ect of the research,
belonging to a representative sample,
through standardi$ed or otherwise &uestioning procedure,
with the aim of studying the relationship among the variables and.or collecting information that
probably describe the whole population.
SUR/EYS ...
*urveys can be conducted
(i either
i STRUCTURED MANNER
or
i UNSTRUCTURED MANNER.
(ii either
Se*f>A%miis&e"e%
or
A%miis&e"e% &4"o674 o&4e"s.
T4e"e m)3 5e %iffe"e& #)3s &o -o%6-& s6"ve3s@ -4)"& 4e"e
1UESTIONNAIRE
A 86es&io)i"e s4o6*% 5e %esi7e% i s6-4 ) m)e" &4)&
responses can be easily tabulated and translated into findings and recommendations that satisfy the
re.uired ob9ectives of the research pro9ect"
W4i*e %esi7i7 ) 86es&io)i"e, oe m6s& $ee! &4e fo**o#i7 -osi%e")&ios i mi%2
FIRST, %oes i& !"ovi%e &4e e-ess)"3 ifo"m)&io &4)& is "e86i"e% fo" &4e "ese)"-4 !"oAe-&B
SECOND, is i& %esi7e% fo" &4e -oveie-e of &4e "es!o%e&sB
THIRD, %oes i& mee& -o%i7, e%i&i7 )% %)&) !"o-essi7 "e86i"eme&s of &4e "ese)"-4
!"oAe-&B
1UESTIONNAIRE @CCONTINUEDD
I s4o"&, ) 7oo% 86es&io)i"e s4o6*%
)--ommo%)&e )** &4e "ese)"-4 o5Ae-&ives i s6ffi-ie& %e!&4 )% 5"e)%&4E
Fs!e)$G &o "es!o%e&s i &4e 6%e"s&)%)5*e *)76)7e )% )& &4e )!!"o!"i)&e i&e**e-&6)*
*eve*E
5e -oveie& &o 5e )%miis&e"e% > i &e"ms of !osi7 86es&ios )% "e-o"%i7 "es!osesE
5e fi**e% i -omfo"&)5*e &ime %6")&io 5efo"e &4e 5o"e%om se&s iE )%
f)-i*i&)&e -o%i7, e%i&i7, %)&) e&"3 )% %)&) !"o-essi7.
S&e!s ivo*ve% i &4e %eve*o!me& of ) 86es&io)i"e...
F*o#-4)"& 4e"e
SPECIFY WHAT INFORMATION WILL BE SOUGHT
@ m6s& 5e -*e)" )5o6& &4e ifo"m)&io 3o6 ee% &o )s#e" &4e iss6es &o 5e )%%"esse% i &4e
"ese)"-4 !"oAe-&.
@ %o o& &"3 &o e=&")-& &4)& ifo"m)&io #4i-4 "es!o%e&s 4esi&)&e i !"ovi%i7 ei&4e" i& is
&oo -ofi%e&i)* o" &oo !"iv)&e )% !e"so)* &o 5e s4)"e% #i&4 )o&4e" !e"so.
4age 51
@ &"3 &o )s$ only and the only ifo"m)&io &4)& )"e "e*ev)& fo" 3o6" "ese)"-4E fo" &4)& oe
s4o6*% 4)ve )5so*6&e -*)"i&3 )5o6& &4e "ese)"-4 !"o5*em.
DETERMINE TYPE OF 1UESTIONNAIRE AND METHOD OF ADMINISTRATION.
@ s&"6-&6"e% o" 6s&"6-&6"e%B
@ o5&)ii7 ifo"m)&io &4"o674 #e** %esi7e% #"i&&e 86es&ios o" &4"o674 s4o#i7
!i-&6"es )% &e**i7 s&o"ies )5o6& &4ese !i-&6"es, o" -o**e-&i7 ifo"m)&io ve"5)**3 o" i #"i&i7.
@ 86es&io)i"e #i** 5e )%miis&e"e% 53 se*f, 53 4i"e% !e"soe*:)7e-3E &4"o674 m)i*E o"
&e*e!4oi- i&e"vie#s o" &4"o674 i&e"e&:em)i*.
@ #4e&4e" i&e"vie#s #i** 5e -o%6-&e% i !4)ses.
DETERMINE THE CONTENTS OF INDI/IDUAL 1UESTIONS
@ 4o# m)3 86es&ios ee% &o 5e )s$e% )% #4)& )** &o 5e )s$e% i e)-4 86es&ios.
@ #4e&4e" 86es&ios &4)& )"e "e*)&e% )mo7 &4emse*ves Ce.7. 86es&ios "e*)&e% &o !e"so)*
%e&)i*s, o" "e*)&e% &o )&&i&6%e &o#)"%s &4ei" 5osses o" s65o"%i)&esD s4o6*% 5e !6& 6%e" oe 86es&io
#i&4 v)"io6s s65>86es&ios o" s4o6*% &4e3 5e )s$e% )s m)i 86es&ios #i&4o6& )3 s65>86es&ios.
@ o& &"3 &o )s$ so m)3 &4i7s A6s& i oe 86es&ioH
DETERMINE THE CONTENTS OF INDI/IDUAL 1UESTIONS C-o&i6e%@D
To %e&e"mie &4e -o&e&s of 86es&ios, ) "ese)"-4e" s4o6*% )%%"ess &4e fo**o#i7 iss6es2
IS THE 1UESTION NECESSARYB
ARE SE/ERAL 1UESTIONS NEEDED INSTEAD OF ONEB
DO THE RESPONDENTS HA/E NECESSARY INFORMATION TO PRO/IDEB
WILL THE RESPONDENTS LIKE TO PRO/IDE INFORMATION ABOUT THE
1UESTIONS ASKEDB
DETERMINE FORM OF RESPONSES TO EACH 1UESTION
S4o6*% &4e 86es&ios )s$e% 5e i &4e fo"m of OPEN>ENDED 1UESTIONS o" CLOSE>
ENDED 1UESTIONSB
S4o6*% -*ose>e%e% 86es&ios 5e i &4e fo"m of %i-4o&omo6s 86es&ios o" m6*&i!*e>-4oi-e
86es&iosB
S-)*e% Res!ose 16es&ios2 &4ese 86es&ios )"e %esi7e% &o -)!&6"e &4e i&esi&3 of &4e
"es!o%e&sG )s#e"s.
EIAMPLE2 O!e>E%e% 16es&ios
OPEN>ENDED 1UESTIONS )"e &4ose 86es&ios &o #4i-4 ) "es!o%e& "e!*ies i 4e" o" 4is
#o"%s. S6-4 )s >
W4)& %o 3o6 &4i$ is &4e mos& i ee% of im!"oveme& i &4e !"ese& s3s&em of 5i**i7 )&
MTNLBJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJ
Give &4e mos& im!o"&)& "e)so #43 3o6 4)ve se*e-&e% &4e !"ese& 5")% of &oo&4
5"6s4JJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJ.
S677es& )s !e" 3o6" o!iio #4i-4 is &4e mos& im!o"&)& f6-&io of P)"e& K Te)-4e"s
Asso-i)&io i S-4oo*s.JJJJJJJJJJJJJJJJJJJJJ
JJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJ
EIAMPLE K Di-4o&omo6s 16es&ios
@ -*ose% e%e% 86es&ios &4)& )s$ &4e "es!o%e&s &o -4oose 5e&#ee &#o )s#e"s. S6-4 )s >
Do 3o6 &4i$ &4)& &4is 3e)" if*)&io is 7oi7 &o 5e mo"e &4) &4)& of &4e *)s& 3e)"B
4age 53
YES
NO
T4e UPA Gove"me& 4)s &)$e se"io6s s&e!s &o im!"ove I%o>P)$ "e*)&ios.
AGREE
DISAGREE
A--o"%i7 &o 3o6" o!iio, &4e 7"o#&4 ")&e i GDP &4is fis-)* is 7oi7 &o 5e
GREATER THAN THAT OF THE PRE/IOUS YEAR
LOWER THAN THAT OF THE PRE/IOUS YEAR
EIAMPLE K M6*&i!*e>C4oi-e 16es&ios
@ -*ose% e%e% 86es&ios &4)& )s$ &4e "es!o%e&s &o -4oose )mo7 seve")* )s#e"s. S6-4 )s
>
P*e)se i%i-)&e &4e )7e 7"o6! 3o6 5e*o7 &o >
I &4e *)s& mo&4, 4)ve 3o6 visi&e% 3o6" 5)$ fo" @@ CTICK THAT APPLYD.
DETERMINE WORDING OF EACH 1UESTION
@ 6se sim!*e #o"%s.
@ )voi% )3 $i% of 6)m5i76o6s #o"%s.
@ )voi% )m5i76o6s 86es&ios.
@ )voi% *e)%i7 86es&ios.
@ )voi% im!*i-i& )*&e")&ives.
@ )voi% im!*i-i& )ss6m!&ios.
@ )voi% 7ee")*iL)&ios )% es&im)&es.
@ )voi% )voi% %o65*e 5)""e*e% 86es&ios.
LEADING 1UESTIONS2
A *e)%i7 86es&io is oe #4i-4 )&&em!&s &o 76i%e &4e "es!o%e&Ms )s#e".
If ) "es!ose &o ) 86es&io is F!"e>$o# )% 76i%e% oeG, &4e i& is ) *e)%i7 86es&io.
E=)m!*es2
Wo6*% 3o6 *i$e &o vo&e fo" M". R)) Si74, ) m) #4o 4)s -"imi)* 5)-$7"o6%B
YES:NO
Wo6*% 3o6 !"efe" s!e%i7 ) e=&") Rs. 9,9(( o 4e)*&4 -)"e eve"3 3e)" o" *osi7 .( 3e)"s
off 3o6" *ifes!)B YES:NO
H)ve Co7"essme fo"7o&&e fo" &4e s)$e of &4ei" se*fis4 i&e"es&sB
YES:NO
A"e 3o6 i f)vo6" of ) 5)*)-e% 56%7e&B YES:NO
LOADED 1UESTIONS2
A Lo)%e% 16es&io is ) 86es&io #i&4 ) f)*se o" 86es&io)5*e !"es6!!osi&io, )% i& is
N*o)%e%N #i&4 &4)& !"es6m!&io.
E=)m!*es2
H)ve 3o6 s&o!!e% 5e)&i7 3o6" #ifeB YES:NO
4age 55
T4is 86es&io !"es6!!oses &4)& 3o6 4)ve 5e)&e 3o6" #ife !"io" &o i&s )s$i7, )s #e** )s &4)&
3o6 4)ve ) #ife. If 3o6 )"e 6m)""ie%, o" 4)ve eve" 5e)&e 3o6" #ife, &4e &4e 86es&io is *o)%e%.
Do 3o6 fee* 4)!!3 #i&4 &4e mess 3o6" i&e""6!&io 4)s -"e)&e% i &4e mee&i7B
YES:NO
T4is 86es&io !"es6!!oses &4)& 3o6 4)ve i&e""6!&e% )% 3o6" i&e""6!&io 4)s -"e)&e% mess
i &4e mee&i7.
E=)m!*es@
I 3o6" vie#, #4i-4 %3)% is mos& "es!osi5*e fo" fee*i7s of &"6s& i e)"*3 -4i*%4oo%B
Mo&4e" )% F)&4e"
F)&4e" )% Si5*i7
Mo&4e" )% Si5*i7
Do 3o6 &4i$ &4)& o&4e" !eo!*e #o6*% eAo3 &4e 5oo$B
YES
NO
Do 3o6 &4i$ &4)& ) -om!)3 #i&4 ) PTI of mo"e &4) Rs. 9(( -"o"es i ) 3e)" s4o6*% 5e
-osi%e"e% )s ) *)"7e s-)*e -om!)3B
YES
NO
DONGT KNOW
DETERMINE THE SE1UENCE OF 1UESTIONS
@ 3o6 m)3 s&)"& #i&4 sim!*e )% i&e"es&i7 86es&iosHHH
@ s&)"& #i&4 5"o)% 86es&ios )% )""o# %o# &o mo"e s!e-ifi- 86es&ios > i& is -)**e%
FUNNEL APPROACH.
@ 5e7ii7 )% e%i7 of ) 86es&io)i"e m)3 4)ve sim!*e )% *ess %em)%i7 86es&ios
)% i mi%, 4)ve mo"e %em)%i7 )% se"io6s 86es&ios.
@ es6"e !"o!e" !*)-i7 of 5")-4i7 86es&ios so )s &o es6"e !"o!e" s$i! !)&&e".
@ !*)-e %iffi-6*& )% sesi&ive ifo"m)&io )& &4e e% of &4e 86es&io)i"e.
@ !6& Pe"so)* ifo"m)&io fi"s& so )s &o 7)i &4e f)mi*i)"i&3 #i&4 &4e "es!o%e&s.
DETERMINE PHYSICAL CHARACTERISTICS OF 1UESTIONNAIRE
@ !43si-)* )!!e)")-e of ) 86es&io)i"e m)3 )ffe-& &4e -oo!e")&io )% &46s, &4e 86)*i&3
of "es!oses oe -) o5&)i.
@ es6"e !"o!e" siLe )% *)3o6& of ) 86es&io)i"e.
@ ) 86es&io)i"e s4o6*% o& 7ive ) fee*i7 of -*6&&e"e% o" -"o#%e% )!!e)")-e.
@ !"o!e"*3 s&)!*e%:5o6%e% )% 4)%3.
4age 5<