MEASUREMENT & SCALING
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 1/41
MEASUREMENT &SCALING
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 2/41
2
MEASUREMENTIn our daily life we use measurement to describe a specific
feature of a given product or service.
Example: Movie : We may either like it or dislike it.
In Marketing Decision Making Measurement has a crucial Role.
Example: Management wants to know what is the response in the
market for its recently introduced ball-pen.
Response should be in terms of sales or, if firm is not able to
reach its break-even then he has to find out the reasons for
poor sales performance.
Management-Researcher-Collect data-Use Scales under his
process of Investigation.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 3/41
3
Definition ofMeasurement/Scale
Measurement/Scale may be defined as Assignment of
Numbers to characteristics of object or event according
to the rules.
We do not measure the buyer of product but we
measure the characteristics of a buyer or a product.
Buyers :preference usage rate, Income, Attitude can be
measured.
Product: Size, Colour, suitability for particular purpose.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 4/41
4
TYPES OF SCALES
Nominal Scale
Ordinal Scale
Interval Scale
Ratio Scale.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 5/41
5
NOMINAL SCALE• It is simply a system of assigning number symbols to events in order
to label them
• Example: Assigning numbers to football players in order to identify
them – just for convenience, The numbers may be assigned to
students in a class or patients in the hospital.
• We may use Nominal scale by counting students with certain
characteristics or attribute such as those who reside in the university
hostel and other.
• We split a set into subsets which are mutually exclusive and
collectively-exhaustive.
• The only operation involved in this scale is the counting of numbers
in each group.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 6/41
6
NOMINAL SCALE… It is simplest method and least powerful.
The scale doesn’t show any order or distance relationship nor does it
have any arithmetical origin.
It is used for preliminary and exploratory work, where it is sufficient to
know broad dimension of a certain phenomenon.
Number of Students Undergraduates Post Graduates Total
Day Scholars 800 200 1000 Hostellers 400 150 550 Total 1000 350 1550
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 7/41
7
Ordinal Scale
This scale places events in order –
Eg: Rank orders represent ordinal scales – a student’s
rank in his graduation class
These scales have no absolute values – all that we can
say is that one person is higher or lower in rank on the
scale
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 8/41
8
Ordinal Scale…
• Example 1: Rank Households according to their Annual Income.
Household A B C D E
Income 6000 10000 4800 12000 11000
If the household with the highest income is to be given No.1 and so onthen the following order will emerge:-
Household Order of Household on the basis of Income
A D
B E
C B
D A
E C
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 9/41
9
Ordinal Scale… It mere indicates ordering and does not indicate the differences in
annual income among the five households.
Example 2: Likes and Dislikes or acceptability of Softdrink on the
basis of attributes such as colour,flavor etc.
Five point scales can be usedI strongly like it: +2
I like it somewhat:+1
I am indifferent: 0
I dislike it somewhat:-1
I strongly dislike it:-2
One can then combine the individual rankings and get a collective
ranking of the group.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 10/41
10
Interval ScaleIt has the power of nominal and ordinal scale plus one additional
strength, the concept of equality of intervals – eg. The interval
between 1 and 2 equals the difference between 2 and 3.
Example:The Fahrenheit & Centigrade scale & Mean is the
appropriate measure of central tendency
Example: One can say on the basis of this scale, that a
temperature of 100 degrees is 20 degree warmer than 80
degrees and 20 degrees cooler than 120 degrees.
Temperature on scale is a multiple of each other thus differences
can be found out.
Lead prompt completion of measurement and more convenient.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 11/41
11
Ratio ScaleRation Scale possess the power of the preceding three scales as also
the concept of absolute zero or origin.Measurement of physical dimensions like height, weight, distance
and area- geometric mean or harmonic mean are the measures of
central tendency
Equal ratio on the ratio scale indicates the equal ratio among the
elements being measured.
Example: 9lbs and 45lbs are in the ratio of 1:5.
One can change over from one unit to another by using conversion
factor.
Thus facility of conversion from one unit of measurement to another
is available in the case of ratio scale alone.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 12/41
12
Criteria for a Good Scale
Reliability and Validity
Reliability
Test-Retest Reliability
The Alternate Forms reliability
Split-Half Reliability
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 13/41
13
Test-Retest Reliability
• Repeated measurement of the same respondent or group
using the same scaling technique under similar condition
• Administering a test at two points of time to the same person
or a group of persons
• The scores of two tests would be correlated
• If the correlation is low, then reliability too is less
• Reliability is a necessary contribution to validity but it is not
sufficient condition for validity_Ex: Weigh M/c.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 14/41
14
Alternative forms reliability
Same respondents given set of two forms, The forms
should be equivalent but not identical.
The results obtained on the basis of these forms are
compared to ascertain whether there is considerable
difference between two scores.
Time consuming in comparison
Expensive
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 15/41
15
Split-Half Reliability
Administering the measurement instrument only once to test
the internal consistency
It can be used in case of multi-item instrument
It involves splitting of a multi-item measurement instruments
into two equivalent groups,the item responses of the two
groups then correlated to estimate reliability.
A high degree of correlation indicates that there is similarity
or homogeneity among the items.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 16/41
16
Validity
Are we measuring what we set out to measure?
It refers to the extent to which a test measures what we actually wishto measure – it can also be thought of as Utility – validity is theextent to which differences found with a measuring instrumentreflect true differences among those being tested.
Differences between individuals’ scores can be taken as representingtrue differences in the characteristics under study.
Example:
We want to know if people will buy our new brand of
champagne. We ask people how they like it.
We haven’t measured purchase intention; we have measuredattitude toward the product!
OUR MEASURE IS NOT VALID
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 17/41
17
Approaches of Validity
Content Validity
Construct Validity
Predictive Validity
Con current Validity
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 18/41
18
Content ValidityContents of the scale correspond to the content of the attitude
system and that they are comprehensive enough to cover the fullrange of the attitude.
The researcher should define the problem clearly
Identify the items to be measured
Evolve a suitable scale for the purpose
Criticism: it can be criticized on the ground whether or not given
scale contains the content validity will depend on the judgment of
the researcher and this likely to vary from individual to individual. To
avoid this it may be preferable to approach a group of
knowledgeable person, rather than leaving it to one person alone.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 19/41
19
Construct ValidityIn order to apply construct validity researcher postulates the nature
and extent of association between the attitude and other specified
variables. He then examines whether these relationship exist.
If not, there could be two situation either scale might be invalid as it
does not satisfactorily measure what it set out to measure and
second his theory might be deficient in some way and it may be
difficult for him to identify it.
Example: Individual current status in society
Dependent on education, car own, money etc, to consider Elite:
proper association of these variables.
The existence of high degree of correlation indicates test of validity.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 20/41
20
Predictive Validity
It indicates how best the researcher can guess the futureperformance ,from his knowledge of the attitude score.
Example: Opinion questionnaire forms the basis for correctly
forecasting the demand for a product has predictive validity
Procedure of predictive validity is first measure the attitude and then
predicts future behavior on the basis of measurement.
Same obtained scores compared with the earlier predictive scores.
High correlation have high predictive validity.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 21/41
21
Concurrent Validity
An attitude scale on one variable can be used to estimate the scoreon the other variable.
Ex: Social status of respondents on the basis of attitude of hissavings.
Practicality:It should be practical, economical, convenient and results should beinterpretable.
Economy:
There should be balance between ideal research and budget. Wheredata are not freely available economic factors become dominant. Sochoice of data collection method should take into a/c the economicfactors.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 22/41
22
cntd
Convenience:
It is always advised that measurement should not be inconvenient to
administer. There should be some set of clear instruction
regd.questionaire or measurement scale to elicit better response.
The more complex the construct great need for very clear and
complete instructions.
Interpretability:
It is one of the important aspect that the person other than the test
designer have to interpret the results .
There must be important and relevant information to facilitate
interpretation in proper perspective to understand the actual fact.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 23/41
23
Sources of Error
Sourcesof Errors
TheRespondent
TheSituation
TheMeasurer
TheInstrument
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 24/41
24
The Respondents
Different opinion of the respondent on a given issueIt Arises on a/c of status of respondent,level of edu.,social
class,nature of job etc.
Some may be reluctant to express their negative opinion on some
issue
Some other may express deliberately different from others
Some respondents responds without having knowledge on the topic
of issue,answer in vague manner.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 25/41
25
The Situation
Situation of interview influences the measurement
Ex: Investigator approaches a respondent who is
discussing some matter with his friends.
He may refuse to cooperate with investigator or he may
give answer in casual manner to cut short the interview.
When the respondents do not want their identitity
disclosed they may not express actual feelings.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 26/41
26
The Measurer
The measurer is no one but the interviewer.He conductsinterview or questionnaire.
Measurement error may crop on a/c of method used by
interviewer
The carelessness of interviewer during recording of
responses
He may change wordings and he may manipulate the
responses or misinterprets the meaning of responses.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 27/41
27
The Instrument
Defective instrument of data collection cause distortion innumerous ways-
Too lengthy questionnaire
Element of ambiguity in questions
language used in questions is suggestive of a
particular response.
Layout of instruments ,the extent of space available
for response, the type of printing quality and paper
quality may be inadequate or of poor quality.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 28/41
Attitude
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 29/41
29
Attitude???
• An enduring disposition(nature , temperament) toconsistently respond in a given matter
• The term attitude refers to the predisposition/mental state
of individuals/users towards a product/idea/attributes of an
object.
• It also implies the mental readiness to act in a particular
manner and influences the individuals' behavior towards
the object/group/organization/person under consideration.
• In attitude measurement, the researcher is primarily
interested in measuring the "state of mind" of the
respondent (s).
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 30/41
30
Attitudes directly affect purchasedecisions and these in turn,
directly affect attitudes.
Attitude Action/
Behavior
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 31/41
31
Continued…. • The salient factors that go into the building of the overall
attitude of the individual towards the object are
• a) his/her beliefs about the attributes possessed by the
object,
• b) his/her preference or otherwise for those attributes, and
• c) the relative importance of each attribute to the
individual's decision making process.
• It may include factors such as awareness, attitudes and
decision processes.
• The dictionary meaning of Attitude is “settled behavior ,as
indicating opinion”
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 32/41
32
What Authors says…. • Thurstone ,Louis in his article “Attitude can be Measured” in
American Journal of Sociology,Vol.33 in 1928 defines
attitude as “the sum of total of man’s inclinations and
feelings ,prejudices or bias, preconceived notion, ideas,
fear, threats and convictions about any specific topic”
• Katz in his article “The functional approaches to the study
of Attitudes” in public opinion quarterly in 1960 has defined
attitude as “Attitude is predisposition of the individual toevaluate some symbol or object or aspect of his world in a
favorable manner.”
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 33/41
33
Issues in Attitude Measurement
• Attitude measurement relates to the process of measuring
an individual's attitude towards an object. When we go for
measurement of attitudes or any other parameter, one has
to clearly sort out the following :
– • "what" has to be measured ?
– • "who" is to be measured ?
– • the accuracy desired in the measurement
– • the costs permissible
– • the choices available in the measurement/data
collection techniques.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 34/41
34
Components of Attitude
•Affective
•Cognitive
•Behavioral
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 35/41
35
Affective
• The feelings or emotions toward an object
• It indicates the respondents’ liking and
preferences of an object or phenomenon. When
the respondent states “ I like this product” “
Advertising for product X is not good” he is
showing affective component.
• It is quite important in knowing the feelings and
preferences of the respondent towards a certain
product or service.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 36/41
36
Cognitive• It indicates that the respondent is aware of and what knows
about a given object or phenomenon.
• This is related to knowledge ,information having with
respondent
• In marketing research we wants to know whether the
respondent has some idea of product features, advertising
campaigns ,pricing of the product as also its availability
,competitive product and such other aspects.
• “ I have no idea about the teaching of this college”
• Knowledge and beliefs
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 37/41
37
Behavioral
• It refers to what the respondent has done or isdoing in the field of marketing
• This includes the respondent’s purchase of a
product or service, the pattern of purchase, the
quantum of product purchased as also the main
characteristics of the buyer.
– Predisposition(tendency, inclination) to action
– Intentions
– Behavioral expectations
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 38/41
38
General Procedure in AttitudeScale..
• There are numerous methods available for measurement of
attitude.
• Self report-where as person is asked directly how he feels
about an object
• Observation of behavior
• Indirect techniques such as –Word association test,
Sentence completion test, Story telling, performance of
objective task and physiological reactions.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 39/41
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 40/41
40
Continued… • While discarding the unsuitable item ,it should be ensured
that the items retained are such as comprehensively cover
the attitude dimension.
• For this purpose sometimes exploratory study is
undertaken in which some persons are asked to respond to
all the items of the pool. Another approach ,as is used in
thurstone scales, is to ask a group of judges to assess the
suitability of items for the final scale.• Finally the scale once formed should be tested in regard to
its reliability and validity .If the scale satisfies these criteria
,it can be used in survey.
8/3/2019 Rm05 Scaling
http://slidepdf.com/reader/full/rm05-scaling 41/41
www.acnielsen.com