Measurement : Assignment of
numbers or
symbols to
Characteristics of objects
Example: Consumers’ Perception
Attitude
Preferences
Nominal
Ordinal
Interval
Ratio
Measurement Scales: examples
Nominal:
– Categories, Identification, roll nos.
Ordinal:
– Rankings., Preferences, Most popular
brand(s)
Interval :
– Temperature, Grading in exams, User
categories
Ratio:
– Length, Marks in exams, Distance, Salaries
Primary Scales of Measurement
Scale
Nominal Numbers Finish
Assigned
to Runners
Interval Performance
Rating on a 8.2 9.1 9.6
0 to 10 Scale
15.2 14.1 13.4
Ratio Time to
Finish, in
Primary Scales of Measurement
Scale Basi c Common M arketing Permissible Statistics
Characteristics E xam ples Exampl e s Descriptive Inferential
N om inal N u m b e r s identify Social B r a n d nos., Percentages, C hi-square,
& classify objects Security nos., store types m ode binomial test
numbering o f
football
players
Ordinal N os. indicate t h e Quality P reference P ercentile, R ank-order
relative positions rankings, rankings, median correlation,
o f objects but not rankings o f market Friedman
t h e magnitude o f teams i n a position, A N O VA
differences tournam ent social class
betw een t h e m
Interval Differences Tem perature Attitudes, Range, mean, Product-
betw een objects (Fahrenheit, opinions, standard moment
c a n b e compared, Celsius) index nos. deviation correlation,
z e r o point is t tests,
arbitrary regression
Interval
N ominal Ordinal Ratio
Scale
Scale Scale Scale
Preference Ratings
Preference Rs spent last
1-7 11-17
No. Store Rankings
1. Ahuja Sons 7 79 5 15 0
2. Home Saaz 2 25 7 17 200
3. K-mart 8 82 7 17 0
4. Nanz 3 30 6 16 100
5. Aastha 1 10 7 17 250
6. Ebony 5 53 5 15 35
7. Target 9 95 4 14 0
8. Jagdish Store 6 61 5 15 100
9. BigJos 4 45 6 16 0
Comparison
Interval : Y= a + b X
– Here difference between two measurement
is equal, the ratio is not equal
Example:
Ratio : Y= b X
Assigning scores to interval Scale
A five point interval scale could be as
follows:
– Very happy
– Happy
– Neither happy nor unhappy
– Unhappy
– Very unhappy
Now the question is what numerical scores
could be assigned to the five categories ?
Does it really matter ?
Measurement Scales
What measurement scale will you use for
the following?
– Do you own a washing machine?
– What is your marital status ?
– How many books are there in your library?
– How satisfactory is bank’s service?
– Describe your preference for brands of
shampoo?
– Your attendance in the Business Research class
?
– How many members watch DD-1 during the
prime time in your household?
A Classification of Scaling Techniques
Scaling Techniques
Comparative Non-comparative
Scales Scales
Semantic Stapel
Likert
Differential
Scaling Techniques: Attitude and Rating scales
Scaling
Monadic
Comparative Or
Non-Comparative
Scaling Techniques: Attitude and Rating scales
Comparative
Dunlop - 80 59 52 77
Modi 20 - 60 46 56
Ceat 41 40 - 61 60
G.Year 48 54 39 - 67
MRF 23 44 40 33 -
All the frequencies shown in the previous table are divided
from the total number of respondents( 100 here).
A paired comparison
taste test
Other issues related to ranking
A ranking-grid is always better than
simple ranking
In a grid , a frequency table is prepared
for all objects ranked
Only ‘top’ ranking will lead to partial
picture only
An analysis of ‘first two’ is more explicit
Continuous Itemized
Rating Scales Rating Scales
Pictorial
Likert Semantic
Staple
Type Differential
Attitude and Rating scales
Form:
1 2 3 4 5
Attitude and Rating scales
Example ?
Summary of Itemized Scale Decisions
1) Number of categories Although there is no single, optimal number,
traditional guidelines suggest that there should
be between five and nine categories
2) Balanced vs. unbalanced In general, the scale should be balanced to obtain
objective data
3) Odd/ even no. of categories If a neutral or indifferent scale response is possible
from at least some of the respondents, an odd
number of categories should be used
4) Forced vs. non-forced In situations where the respondents are expected
to have no opinion, the accuracy of the data may
be improved by a non-forced scale
5) Verbal description An argument can be made for labeling all or many
scale categories. The category descriptions should
be located as close to the response categories as
possible
6) Physical form A number of options should be tried and the best
selected
Attitude and Rating scales:
Some standard Scales
Thurstone:
– Agreement /Disagreement on a large
number of statements ( usually 11)
– Time consuming , Tiring, dissuading ,
– Long process of identifying
Likert:
– Five point( usually)
– Levels of Agreement /Disagreement
– Easy to explain /develop
Attitude and Rating scales
Stapel:
– Unipolar,10 point interval, rating from
+5 to -5
– Balanced without neutral point
Semantic Differential
– ( usually) 7 point scale
– Bi polar, with a neutral point,
– rating +3 to -3
Summative scale
– Attitude /Dimension to be used is divided into
several statements and Likert scale of 5 points
( or 4/6/7 etc.) to be used for every
statement.
– The scores of each statement are then
summed up to measure the total score on that
dimension
– Example: to measure brand loyalty of
consumers or organizational commitment of
employees
Loyalty:
I always use this brand Strongly Agree Cant Dis Strongly
Agree say Agree disagree
5 4 3 2 1
I am willing to pay more for
this brand
………..
……..
Balanced Scores
To make a comparison between two
attributes /dimensions
Example:
– For me ………. Comes first
Profit Customer
satisfaction
5 4 3 2 1 0 1 2 3 4 5
Basic Non-comparative Scales
Scale Basic Examples Advantages Disadvantages
Characteristics
Continuous Place a mark on a Reaction to TV Easy to construct Scoring can be
Rating continuous line commercials cumbersome
Scale unless
computerized
Itemized Rating
Scales
Likert Scale Degrees of Measurement Easy to construct, More
agreement on a 1 of attitudes administer, and time-consuming
(strongly disagree) understand
to 5 (strongly
agree) scale
3) . Very harsh
. Cheer
.
. Neither harsh nor gentle
.
.
. Very gentle
4)
Very Somewhat Neither harsh Very
harsh Harsh harsh nor gentle Somewhat Gentle gentle
gentle
5) -3 -2 -1 0 +1 +2 +3