Description
This study investigates the brand choice behaviour of Indian consumers for toothpaste. Data has been collected through survey method from the major cities of Punjab.
Global Advanced Research Journal of Management and Business Studies (ISSN: 2315-5086) Vol. 2(10) pp. 511-517, October, 2013
Available onlinehttp://garj.org/garjmbs/index.htm
Copyright © 2013 Global Advanced Research Journals
Full Length Research Paper
An Analysis of Brand Choice Behaviour of Indian
Consumers for Toothpaste
Rashmi*
*Assistant professor in D.A.V. College, Bathinda (Punjab).
Email : [email protected]
Accepted 23 September 2013
This study investigates the brand choice behaviour of Indian consumers for toothpaste. Data has been
collected through survey method from the major cities of Punjab. The various variables examined are
sales promotion variables (discount, free gifts, feature advertisement, celebrity advertisement and store
display), brand-specific attributes (functional aspect, ingredients, value and easiness) and consumer
demographics (gender, marital status, age, income, education and occupation). Logit model is used to
predict the choice of most preferred brand. Results of the study are useful for marketing managers to
make their brand as leading brand.
Keywords: Brand, toothpaste, most preferred brand, sales promotion variables, brand-specific attributes,
consumer demographics.
INTRODUCTION
In the present competitive scenario and increasing
global market where there is large scale of brand
proliferation, marketing managers are interested in
determining – how the consumer decides which product
to buy. Products are what the company makes. But the
features of a product can be easily copied. It means that
what the consumer buys is a brand. Thus the brands are
considered a marketer’s tool for creating product
differentiation. The most important function of a brand is
that it distinguishes the goods of one producer from the
other. The American Marketing Association (AMA)
define brand as, “A name, term, design, symbol, or any
other feature that identifies one seller's good or service
as distinct from those of other sellers. The legal term for
brand is trademark. A brand may identify one item, a
family of items, or all items of that seller. If used for the
firm as a whole, the preferred term is trade name.” In
fact, brand is comprised of all such elements that identify
the brand in general as well as differentiate it from other
brands. This can be a name, a logo, a character, a
slogan, a jingle, or the packaging. Here the question
arises, when the consumer has large number of
alternative brands available, then what factors he/she
considers while choosing a brand? Further “what is the
relative influence of various factors?” Marketing
managers look for an answer to these questions before
developing a comprehensive marketing strategy.
Consumer evaluates the brands on the basis of their
attributes and selects that brand which proves best on
his/her evaluative criteria. The concept of evaluating a
decision, product or service as a function of its attributes
is a universally accepted approach. It has been applied
in various other fields such as economics (Fishburn,
1967, 1968; Mcguire and Weiss, 1976; Theil, 1969),
engineering (Turban and Metersky, 1971) and finance
(Slovic et al. 1972). A detailed review of literature
reveals that while making a brand choice decision,
consumers evaluates the brnad on the basis of its
512 Glo. Adv. Res. J. Manag. Bus. Stud.
attributes as depicted by (Kraft et al. 1973, Coskunoglu
et al. 1985, Panda, 2005, Banerjee et al. 2005).
Further, the choice of brand is affected due to the
difference in sales promotion variables like discount, free
gifts, store display and feature advertisement, etc. as
examined by Reilbstein (1978), Guadagni and little
(1973), Chintagunta et al. (1991), Banerjee et al. (2005),
Singh et al. (2005).
Further on, a detailed review with regard to the brand
choice behaviour of the consumers reveals that
demographics also play an important role in brand
choice decision of the consumers. Krishnamurthi and
Raj (1988) examine the brand choice probabilities of
consumers for frequently purchased products. The
demographic variable used in the analysis is income.
Results depict a positive coefficient value for income.
Kalyanam and Putler (1997) include demographic
variables like income in their analysis to examine brand
choice behaviour. They find the positive coefficient value
for income which shows that households are more likely
to buy that brand which has a larger size in the product
category. Murthi and Srinivasan (1999) estimate the
brand choice probabilities for ketchup category. Results
show that income has a negative effect whereas
education positively affects the brand choice decision.
Degeratu et al. (2000) examine the role of income which
has a positive effect. Banerjee et al. (2005) evaluate the
brand preferences for toiletries category. The
demographic variables used in the study are age,
education, marital status, occupation, income, etc. As
per the results, significant demographic variables (age,
occupation and income) show preference for a particular
brand.
The above discussion shows that a large number of
variables put an influence on brand choice decision of
consumers. However most of the studies are conducted
outside India. Thus in order to examine the influence of
above discussed variables on brand choice decision,
logit model is used to predict the brand choice behaviour
of Indian consumers while buying toothpaste. The choice
model proposed in this study is stochastic in nature
since it acknowledges the uncertainty of choice
outcome. It is different from most stochastic choice
models that currently appear in the marketing literature
because it explains a single choice.
Objectives of the Study
The overall objective of this study is to determine the
brand choice behaviour of consumers for toothpaste.
The sub-objectives of the study are:
1. To determine which brands are frequently
chosen by the consumer while buying toothpaste.
2. To examine the relative influence of various
factors in the brand choice behaviour of the consumer
while buying toothpaste.
3. To examine which factor plays the most
important role while making a choice between various
brands of toothpaste.
METHODOLOGY
The behaviour of the consumers is analysed for
toothpaste. The reason behind choosing this product is
that
a) this product has a large number of brands,
b) this product is used by end consumer
c) this is purchased frequently. Thus it is deemed that
information provided by the consumers will be correct.
In order to examine the brand choice behaviour of the
consumers for toothpaste, data is collected through
survey method. People from Punjab are approached
through a field survey. Survey is conducted in various
areas like Amritsar (26%), Jalandhar (24%), Ludhiana
(26%) and Chandigarh (24%). The information needed is
collected from the consumers of the product selected for
the study.
A sample of 550 respondents is selected on the basis
of judgement cum convenience sampling. As far as
possible, the respondents were approached in the
market place outside the major shopping centres of the
four cities. They were requested to participate in the ‘not
for profit’ survey. If they agreed, they were asked to fill
the questionnaire. Due care was taken to give
appropriate representation to gender and age. Finally,
543 questionnaires have been used in the analysis.
Biased and incomplete questionnaires have been
removed from the study.
The sample shows that out of the total respondents,
44% are male and 56% are females. The actual age of
the consumers has been recorded. Thus no age wise
description is made. However age of the respondents
varies between 18-59 years of age. Out of the total
sample, the percentage of married respondents is 46%
and that of unmarried is 54%. The respondents were
asked their educational qualification. Education level of
the respondents shows that 8% of the respondents have
passed secondary school, 44% are graduates, 31% are
post graduates and 17% are professionally qualified.
The respondents are segregated on the basis of income
also. Income level shows that 10% of the respondents
are earning up to Rs.15,000/-, 34% are earning between
Rs.15,001/- to Rs.25,000/-, 35% are having income from
Rs.25,001/-to Rs.35,000/- and 21% are earning
`35,001/- and above. The sample collects information
from almost all types of the consumers who are engaged
in different occupations. Self-employed means
businessmen as well as professionals having their own
practice like doctors and chartered accountant doing
their own practice. The percentage of respondents who
are self employed is 28%. Salaried people cover
serviceman as well as respondents working as
Rashmi, 513
Table 1. Brands being used by the respondents for Toothpaste and their respective share
Toothpaste
Brand Share
Colgate 54.3
Pepsodent 24.9
Close up 8.1
Dabur Red 3.2
Miswak 2.4
Babool 1.9
Cibacaa 1.4
Glister (Amway) 1.4
Anchor 0.8
Thermoseal 0.8
Promise 0.5
Vicco 0.3
Total 100
executives or on other higher posts. The percentage of
salaried people is 35%. Housewives account for 17%,
students 15% and retired personnel are only 5%.
The Structure
A large number of factors persuade the brand choice
behaviour of the consumers. Brand specific attributes of
toothpaste based on experts’ advice and own intuition,
various sales promotion variables extracted from the
previous literature and demographic characteristics of
the consumers are observed to determine their effect on
brand choice decision of the consumers.
Brand choice
Brand choice of the consumers is taken as the
dependent variable of the study. Consumers are asked
to state their brand choice in terms of present brand
used for toothpaste. As per Table 1, total 12 brands are
mentioned by the respondents with the highest share for
Colgate (54.3%). Thus Colgate is taken as the most
preferred brand for this analysis. In actual market,
Colgate is also the market leader with 63%
1
www.smh.com.au/articles/2003/1/26/1069825840957.ht
ml?from=storyrhs of market share in the toothpaste
market.
In order to examine the brand choice behaviour of the
consumers for the brand with highest share, or most
preferred brand by the respondents, this variable is
converted into a dummy variable. The value for those
respondents is taken as 1 who mentioned Colgate as
their present choice of brand and 0 for those
respondents who mentioned any other brand of
toothpaste as the present brand being used.
Sales promotion variables
Sales promotion variables are also included in this study
to make the results more useful for the marketing
managers so that they could design their marketing
strategy as per the needs and aspirations of the
consumers. The sales promotion variables examined in
this study is discount, free gifts, feature advertisement,
celebrity advertisement, and store display. Discount
means when the brand is available at a price less than
its actual price. Free gifts means when the brand offers
such gifts/items with the product for which it charges
nothing. Feature advertisement refers to that brand
whose advertisement discloses or concentrates upon the
features of the brand rather than other irrelevant
aspects. Celebrity advertisement means when the brand
is being endorsed by some celebrity. Store display
means the brand is prominently displayed in the store.
The above mentioned sales promotion variables are
measured with the help of some questions. Discount is
measured by asking the respondents, ‘The present
brand of toothpaste when purchased was available on
discount’. For measuring free gifts, respondents were
asked, ‘The present brand of toothpaste when
purchased, offered some free gifts.’ To measure feature
advertisements, respondents were asked, ‘The present
brand of toothpaste when purchased was feature
advertised (the attributes of the product were
advertised’. Celebrity advertisement was measured by
asking, ‘I purchased the present brand of toothpaste
because some celebrity advertised it’. Store display is
measured by asking, ‘The present brand of toothpaste
when purchased was on display (prominently placed in
store)’
In the above mentioned questions, respondents are
asked the condition of sales promotion variables
(discount, free gifts, feature advertisement, celebrity
514 Glo. Adv. Res. J. Manag. Bus. Stud.
Table 2. Results of Factor analysis
Statements Functional aspect Ingredients Value Easiness
It has a good flavour. 0.517
It prevents tooth decay. 0.667
It gives me a feeling of freshness. 0.888
It has a good whitening power. 0.809
It is sweet in taste. 0.547
It is vegetarian toothpaste. 0.526
It has herbal ingredients. 0.843
The price of this brand is reasonable. 0.828
The manufacturer is reputed one. 0.649
My doctor recommended it. 0.729
It is easily available. 0.827
I am familiar with this brand. 0.806
Table 3. Description of demographic variables
Demographic variables Description
Gender 1 if the respondent is male, 0 otherwise.
Marital Status 1 if the respondent is married, 0 otherwise.
Age Actual age of respondent
Income 1, if income is up to `15,000/-.
2, if income is between `15,001/- to `25,000/-.
3, if income is between `25,001/- to `35,000/-.
4, if income is `35001/- or above.
Education 1, if the respondent is under graduate.
2, if the respondent is graduate.
3, if the respondent is postgraduate/professionally qualified.
Occupation (self employed as base
category)
1, if the respondent is salaried, 0 otherwise.
1, if the respondent is housewife, 0 otherwise.
1, if the respondent is student, 0 otherwise.
1, if the respondent is retired personnel, 0 otherwise.
advertisement and store display) one by one at the time
of purchase. The value of the variable is taken as 1 if it is
present at the time of purchase and 0 otherwise. If the
respondent has purchased the brand at a ‘discount’ it
has been allotted a value of 1, and 0 otherwise. Further
if the respondent has purchased that brand which offers
certain free gift with it then the variable ‘free gifts’ has a
value of 1, otherwise 0. If according to the respondent,
the brand is ‘feature advertised’, its value is 1, 0
otherwise. If the brand is endorsed by some celebrity, its
value is 1, 0 otherwise. Lastly if the respondent has
purchased the brand because the brand is prominently
displayed in the store, its value is 1, otherwise 0.
In all the above mentioned questions, the respondents
are supposed to answer in ‘yes’ or ‘no’ only.
Brand-specific attributes
Brand specific attributes play an important role in the
choice of the brand. Thus specific attributes of
toothpaste are included in this study which are based on
experts’ advice and own intuition. The respondents were
to rate these attributes on a seven-point likert scale
ranging from ‘highly satisfied’ to ‘highly dissatisfied’.
Total twelve attributes are examined which are in
relation to the particular brand of toothpaste being used.
Thus, actually these scale items are asked in the context
of the present brand being used by the respondents.
These scale items are factor analysed. Rotated
component matrix with varimax rotation has been
employed to extract factors appropriate for representing
brand specific attributes. The total variance explained is
60%. This percentage of variance is acceptable since
the satisfactory percentage of variance explained in
social sciences is 60% (Hair, et al. 2005). The scale
items result into four factors. Factor score of the
resultant factors is used for further analysis. Only those
items could be considered whose factor loading is
greater than .05 (Hair et al. 2005). Thus all the items are
considered for further analysis since no item has value
less than .05. The results of factor analysis are
presented in Table 2.
Consumer demographics
Consumer demographics are also taken into account to
analyse brand choice behaviour. Consumers differ from
one another because of the demographic traits. Hence
their choice process is also likely to be different. Thus it
becomes necessary to include the demographic traits as
explanatory variables in the brand choice model. The
various demographic characteristics examined in this
analysis along with their measurement criteria are listed
in Table 3.
Hypothesis Development
Three categories of variables are examined to determine
their influence on brand choice behaviour of the
consumers. Significance level of these variables is
checked through the following hypothesis.
Sales promotion variables
Various sales promotion variables examined in the study
are discount, free gifts, feature advertisement, celebrity
advertisement and store display. Hypothesis set are as
under:
H
1a
: Discount and brand choice have significant
relationship.
H
1b
: Free gifts and brand choice have significant
relationship.
H
1c
: Feature advertisement and brand choice have
significant relationship.
H
1d
: Celebrity advertisement and brand choice have
significant relationship.
H
1e
: Store display and brand choice have significant
relationship.
Brand-specific attributes
The following hypotheses are built to check the
significance of brand-specific attributes on brand choice
of toothpaste.
H
2a
: Functional aspect and brand choice have significant
relationship.
H
2b
: Ingredients and brand choice have significant
relationship.
H
2c
: Value and brand choice have significant
relationship.
H
2d
: Easiness and brand choice have significant
relationship.
Consumer demographics
The hypothesis for demographic traits of gender, marital
status, age, income, education and occupation are as
under:
Rashmi, 515
H
3a
: Gender and brand choice have significant
relationship.
H
3b
: Marital Status and brand choice have significant
relationship.
H
3c
: Age and brand choice have significant relationship.
H
3d
: Income and brand choice have significant
relationship.
H
3e
: Education and brand choice have significant
relationship.
H
3f
: Salaried (Occupation) and brand choice have
significant relationship.
H
3g
: Housewife (Occupation) and brand choice have
significant relationship.
H
3h
: Student (Occupation) and brand choice have
significant relationship.
H
3i
: Retired Personnel (Occupation) and brand choice
have significant relationship.
Logit Analysis
Logit analysis has been used to determine the
relationship between a binary dependent variable and
multiple independent variables. As already discussed, if
the respondent has purchased the most preferred brand
he/she is allotted a value of 1, 0 otherwise. This variable,
by taking a binary value of either 1 or 0, becomes the
dependent variable. Further, factor scores of brand-
specific attributes consisting of functional aspect,
ingredients, value and easiness act as the independent
variables of the study. Dummy value of sales promotion
variables consisting of discount, free gifts, feature
advertisement, celebrity advertisement and store display
are other independent variables. Moreover demographic
traits of gender, marital status, age, income, education
and occupation are also added to the analysis as
independent variables.
The model of consumer’s decision of buying the most
preferred brand takes the following form:
Log [P
i
/1 – P
i
] = ? + ß X
j
+ ?U
ij
+ ?Z
i
+ µ
i
Equation: 1
where i (i=1,2, _ _ _, n)) refers to the individual; j refers
to the current brand used by individual i; ?
is constant; X
j
is a vector of sales promotion variables; U
ij
is a vector of
brand-specific attributes; Z
i
is a vector of demographic
variables; ß, ? and ? are the coefficients of X
j
, U
ij
and Z
i
respectively; µ
ij
refers to the error term which captures all
misspecifications associated with a given individual and
brand being used. P
i
is the probability of buying the
most preferred brand; and 1 - P
i
is the probability of not
buying the most preferred brand. Log [P
i
/1 – P
i
] is log-
odds ratio, that is, the natural logarithm of the odds that
most preferred brand will be bought by a particular
individual.
After specifying the values for explanatory variables
the above specified model is tested. The method of
516 Glo. Adv. Res. J. Manag. Bus. Stud.
Table 4. Logit results
Description of variables Variable Toothpaste Coefficient (T ratio)
Constant 10.91 (6.61)*
Sales promotion Variables Discount -0.99 (-2.02)****
Free gifts 1.24 (1.28)
Feature advertisement -0.52 (-2.57)**
Celebrity advertisement -0.70 (-0.54)
Store display 0.17 (0.04)
Brand-specific attributes Functional aspect 1.81 (2.95)**
Ingredients 0.74 (2.87)**
Value 0.96 (2.59)***
Easiness -0.56 (-0.32)
Consumer demographics Gender 0.29 (2.14)****
Marital Status 0.47 (1.95)****
Age 0.73 (1.17)
Income 1.52 (1.94)****
Education 0.89 (3.45)*
Salaried (Occupation) -0.99 (-1.35)
Housewife(Occupation) 0.48 (2.54)***
Student(Occupation) 0.74 (0.75)
Retired Personnel (Occupation) 0.99 (1.12)
Correct predictions 88%
Note: *,**,***,**** represent highly significant, significant at 1%, 5% and 10% respectively.
maximum likelihood has been used wherein that value of
the coefficients is taken which maximizes the following
log likelihood function (Madnani, 1994):
}] Z U X { F 1 [ log ) Yi 1 ( 1 ] } Z U X { F [ log Yi L
i ij j
n
1 i
i ij j
n
1 i
? + ? + ? + ? ? ? ? + ? + ? + ? + ? ? =
= =
Equation: 2
where L = the log- likelihood function to be maximized
and Yi is dummy variable taking value 1 if most preferred
brand is purchased and 0 otherwise.
The outcome of above stated model would give the
estimated values of coefficients of explanatory variables
that is, sales promotion variables (discount, free gifts,
feature advertisement, celebrity advertisement, store
display), brand-specific attributes (functional aspect,
ingredients, value, easiness) and demographic variables
(gender, marital status, age, income, education,
occupation) which will depict the value of change in log
odds ratio of the purchase probability of most preferred
brand.
Thus the above stated model depicts the probability of
buying the most preferred brand for each consumer.
DATA ANALYSIS
The logit model is used to estimate the probability of
buying the most preferred brand by the respondents.
Table 4 depicts that 88% of the observations are
correctly predicted.
Table 4 reveals that all variables except free gifts,
celebrity advertisement, store display, age, salaried,
student and retired personnel do not play a significant
role in brand choice decision. Thus H
1a
(discount), H
1c
(feature advertisement), H
2a
(functional aspect), H
2b
(ingredients), H
2c
(value), H
3a
(gender), H
3b
(marital
status), H
3d
(income), H
3e
(education), H
3g
(housewife)
are accepted whereas H
1b
(free gifts), H
1d
(celebrity
advertisement), H
1e
(store display), H
2d
(easiness), H
3c
(age), H
3f
(salaried), H
3h
(student), H
3i
(retired personnel)
are rejected.
The most important variable affecting brand choice
decision is functional aspect of toothpaste with the
highest coefficient value (1.81). The least influencing
factor is marital status (0.47)
DISCUSSION
Among the sales promotion variables, the most
influencing variable is discount (-0.99). Discounted
brands can attract the attention of the consumer and
he/she can swing from his/her last purchased brand
which may be even the top brand of the market. A price
conscious consumer buys the discounted brands to save
money or to store a large quantity for future. The product
category chosen in this study is largely affected due to
discount strategy of the consumers. However the
marketing managers must be wide awake while drawing
discount strategy. While explaining about promotion
techniques, Alvarez and Casielles (2005) suggest the
marketing managers that the sales promotion techniques
(discount mainly) will be more result oriented when
consumers are not expecting it.
In case of toothpaste, feature advertisement (-0.52)
affect brand choice decision which means that
consumers are willing to know about the features of
various brands of toothpaste so that they could take a
right brand choice decision. As celebrity advertisement is
not important for the consumers, hence consumers are
required to be informed, no matter who provides this
information. Further free gifts and display are also not
important for the consumer while buying toothpaste.
Bhusghan and Daftari (2007) have suggested one new
mechanism – modern trade that offers better displays
and ambience as compared to kiryana stores to push the
sales of brand up.
Regaining brand specific attributes, in case of
toothpaste, all the factors work well except easiness.
Consumers are willing to buy the most preferred brand
because of its functional aspect (1.81), ingredients (0.74)
and value (0.96). However easiness is not effective in
influencing the purchase of most preferred brand. The
most important attribute for the consumers is functional
aspect of toothpaste. The reason may be the utilitarian
nature of the product as any wrong purchase could
result in serious long lasting dental problems. While
discussing the role of advertising in generation of brand
attitudes, Baker (2001) suggests that advertising must
maximise specific brand information that can be used by
the consumers to make a differentiation between various
alternatives so that they could choose the best one.
In this study various demographic variables are also
examined and significant results are found. Results
depict that male (0.29), married (0.47) and housewives
(0.48) are willing to buy the most preferred brand.
Increase in income and increase in education also
increases the probability of buying the most preferred
brand.
CONCLUSION
This article investigates the reasons that why consumers
buy a particular brand more because of which it
becomes the top or leading brand. Influencing factors
extracted from the results could be used by marketing
managers to make their brand as the leading brand.
REFERENCES
Alvarez BA, Casielles RV (2005). Consumer Evaluation of Sales
Promotion: The Effect on Brand Choice. European Journal of
Marketing 39 (½): 54-70.
Baker WE (2001). “The Diagnosticity of Advertising Generated Brand
Attitudes in Brand Choice Contexts”, Journal of Consumer
Psychology, Vol 11 No 2, pp. 129-139.
Banerjee A, Awasthy D, Gupta V (2005). A Choice Modelling
Approach to Evaluate Effectiveness of Brand Development
Initiatives. International Journal of Management and Decision
Making, 6(2): 180-198.
Rashmi, 517
Bhusghan R, Daftari I (2007). Regional Brand to Cash in on Modern
Trade. The Economics Times (Business of Brands), p 3.
Chintagunta PK, Jain DC, Vilcassim NJ (1991). Investigating
Heterogeneity in Brand Preferences in Logit Model for Panel Data.
Journal of Marketing Research, 28(4): 417-428.
Coskunoglu O, Hansotia BJ, Shaikh MA (1985). A New Logit Model for
Decision-Making and Its Application. Journal of the Operational
Research Society, 36(1): 35-41.
Degeratu AM, Rangaswamy A, Wu J (2000). Consumer Choice
Behaviour in Online and Traditional Supermarkets: The Effects of
Brand Name, Price and Other Search Attributes. International
Journal of Research in Marketing, 17: 55-78.
Fishburn P (1967). “Methods of Estimating Additive Utilities”,
Management Science, Vol 13 No 7, pp. 435-453.
Fishburn P (1968). “Utility Theory”, Management Science, Vol 14 No 5
(Theory Series), pp. 335-378.
Guadagni PM, Little JDC (1983). A Logit Model of Brand Choice
Calibrated on Scanner Data. Marketing Science, 2(3): 203-238.
Hair, Joseph F Jr, Anderson, Rolph E, Tatham, Ronald L, Black,
William C (2005). Multivariate Data Analysis, Sixth Edition, Pearson
Education.
Kalyanam K, Putler DS (1997). Incorporating Demographic Variables
in Brand Choice Models: An Invisible Alternatives Framework.
Marketing Science, 16 (2): 166-181.
Kraft FB, Granbois DH, Summers JO (1973). Brand Evaluation and
Brand Choice: A Longitudinal Study. Journal of Marketing Research,
10 (3): 235-241.
Krishnamurthi L, Raj SP (1988). A Model of Brand Choice and
Purchase Quantity Price Sensitivities. Marketing Science, 7 (1): 1-
20.
Madnani GMK (1994). Introduction to Econometrics, 6
th
Ed., Oxford
and IBH Publishing Co. Private. Ltd.
McGuire TW, Weiss DL (1976). Logically Consistent Market Share
Models. Journal of Marketing Research, 13(3): 296-302.
Murthi BPS, Srinivasan K (1999). Consumers’ Extent of Evaluation in
Brand Choice. Journal of Business, 72 (2): 229-256.
Panda TK (2005). “Predicting Behavioral Intention for Purchase and
Positioning of a New Brand”, Journal of Marketing and
Communication, Vol 1 No 2, pp. 53-67.
Reibstein DJ (1978) “The Prediction of Individual Probabilities of Brand
Choice”, Journal of Consumer Research, Vol 5 No 3, pp. 163-168.
Singh VP, Hansen KT, Gupta S (2005). Modeling Preferences for
Common Attributes in Multicategory Brand Choice. Journal of
Marketing Research, 42(2): 195-209.
Slovic P, Fleisnner D, Bauman WS (1972). Analyzing the Use of
Information in Investment Decision-Making: A Methodological
Proposal. Journal of Business, 45(2): 283-301.
Theil H (1969). “A Multinomial Extension of the Linear Logit Model”,
International Economic Review, Vol 10 No 3, pp. 251-259.
Turban E, Metersky ML (1971). Utility Theory Applied to Multi-variable
System Effectiveness Evaluation. Management Science, 17(12)
Application Series: B817-B828.
doc_977516843.pdf
This study investigates the brand choice behaviour of Indian consumers for toothpaste. Data has been collected through survey method from the major cities of Punjab.
Global Advanced Research Journal of Management and Business Studies (ISSN: 2315-5086) Vol. 2(10) pp. 511-517, October, 2013
Available onlinehttp://garj.org/garjmbs/index.htm
Copyright © 2013 Global Advanced Research Journals
Full Length Research Paper
An Analysis of Brand Choice Behaviour of Indian
Consumers for Toothpaste
Rashmi*
*Assistant professor in D.A.V. College, Bathinda (Punjab).
Email : [email protected]
Accepted 23 September 2013
This study investigates the brand choice behaviour of Indian consumers for toothpaste. Data has been
collected through survey method from the major cities of Punjab. The various variables examined are
sales promotion variables (discount, free gifts, feature advertisement, celebrity advertisement and store
display), brand-specific attributes (functional aspect, ingredients, value and easiness) and consumer
demographics (gender, marital status, age, income, education and occupation). Logit model is used to
predict the choice of most preferred brand. Results of the study are useful for marketing managers to
make their brand as leading brand.
Keywords: Brand, toothpaste, most preferred brand, sales promotion variables, brand-specific attributes,
consumer demographics.
INTRODUCTION
In the present competitive scenario and increasing
global market where there is large scale of brand
proliferation, marketing managers are interested in
determining – how the consumer decides which product
to buy. Products are what the company makes. But the
features of a product can be easily copied. It means that
what the consumer buys is a brand. Thus the brands are
considered a marketer’s tool for creating product
differentiation. The most important function of a brand is
that it distinguishes the goods of one producer from the
other. The American Marketing Association (AMA)
define brand as, “A name, term, design, symbol, or any
other feature that identifies one seller's good or service
as distinct from those of other sellers. The legal term for
brand is trademark. A brand may identify one item, a
family of items, or all items of that seller. If used for the
firm as a whole, the preferred term is trade name.” In
fact, brand is comprised of all such elements that identify
the brand in general as well as differentiate it from other
brands. This can be a name, a logo, a character, a
slogan, a jingle, or the packaging. Here the question
arises, when the consumer has large number of
alternative brands available, then what factors he/she
considers while choosing a brand? Further “what is the
relative influence of various factors?” Marketing
managers look for an answer to these questions before
developing a comprehensive marketing strategy.
Consumer evaluates the brands on the basis of their
attributes and selects that brand which proves best on
his/her evaluative criteria. The concept of evaluating a
decision, product or service as a function of its attributes
is a universally accepted approach. It has been applied
in various other fields such as economics (Fishburn,
1967, 1968; Mcguire and Weiss, 1976; Theil, 1969),
engineering (Turban and Metersky, 1971) and finance
(Slovic et al. 1972). A detailed review of literature
reveals that while making a brand choice decision,
consumers evaluates the brnad on the basis of its
512 Glo. Adv. Res. J. Manag. Bus. Stud.
attributes as depicted by (Kraft et al. 1973, Coskunoglu
et al. 1985, Panda, 2005, Banerjee et al. 2005).
Further, the choice of brand is affected due to the
difference in sales promotion variables like discount, free
gifts, store display and feature advertisement, etc. as
examined by Reilbstein (1978), Guadagni and little
(1973), Chintagunta et al. (1991), Banerjee et al. (2005),
Singh et al. (2005).
Further on, a detailed review with regard to the brand
choice behaviour of the consumers reveals that
demographics also play an important role in brand
choice decision of the consumers. Krishnamurthi and
Raj (1988) examine the brand choice probabilities of
consumers for frequently purchased products. The
demographic variable used in the analysis is income.
Results depict a positive coefficient value for income.
Kalyanam and Putler (1997) include demographic
variables like income in their analysis to examine brand
choice behaviour. They find the positive coefficient value
for income which shows that households are more likely
to buy that brand which has a larger size in the product
category. Murthi and Srinivasan (1999) estimate the
brand choice probabilities for ketchup category. Results
show that income has a negative effect whereas
education positively affects the brand choice decision.
Degeratu et al. (2000) examine the role of income which
has a positive effect. Banerjee et al. (2005) evaluate the
brand preferences for toiletries category. The
demographic variables used in the study are age,
education, marital status, occupation, income, etc. As
per the results, significant demographic variables (age,
occupation and income) show preference for a particular
brand.
The above discussion shows that a large number of
variables put an influence on brand choice decision of
consumers. However most of the studies are conducted
outside India. Thus in order to examine the influence of
above discussed variables on brand choice decision,
logit model is used to predict the brand choice behaviour
of Indian consumers while buying toothpaste. The choice
model proposed in this study is stochastic in nature
since it acknowledges the uncertainty of choice
outcome. It is different from most stochastic choice
models that currently appear in the marketing literature
because it explains a single choice.
Objectives of the Study
The overall objective of this study is to determine the
brand choice behaviour of consumers for toothpaste.
The sub-objectives of the study are:
1. To determine which brands are frequently
chosen by the consumer while buying toothpaste.
2. To examine the relative influence of various
factors in the brand choice behaviour of the consumer
while buying toothpaste.
3. To examine which factor plays the most
important role while making a choice between various
brands of toothpaste.
METHODOLOGY
The behaviour of the consumers is analysed for
toothpaste. The reason behind choosing this product is
that
a) this product has a large number of brands,
b) this product is used by end consumer
c) this is purchased frequently. Thus it is deemed that
information provided by the consumers will be correct.
In order to examine the brand choice behaviour of the
consumers for toothpaste, data is collected through
survey method. People from Punjab are approached
through a field survey. Survey is conducted in various
areas like Amritsar (26%), Jalandhar (24%), Ludhiana
(26%) and Chandigarh (24%). The information needed is
collected from the consumers of the product selected for
the study.
A sample of 550 respondents is selected on the basis
of judgement cum convenience sampling. As far as
possible, the respondents were approached in the
market place outside the major shopping centres of the
four cities. They were requested to participate in the ‘not
for profit’ survey. If they agreed, they were asked to fill
the questionnaire. Due care was taken to give
appropriate representation to gender and age. Finally,
543 questionnaires have been used in the analysis.
Biased and incomplete questionnaires have been
removed from the study.
The sample shows that out of the total respondents,
44% are male and 56% are females. The actual age of
the consumers has been recorded. Thus no age wise
description is made. However age of the respondents
varies between 18-59 years of age. Out of the total
sample, the percentage of married respondents is 46%
and that of unmarried is 54%. The respondents were
asked their educational qualification. Education level of
the respondents shows that 8% of the respondents have
passed secondary school, 44% are graduates, 31% are
post graduates and 17% are professionally qualified.
The respondents are segregated on the basis of income
also. Income level shows that 10% of the respondents
are earning up to Rs.15,000/-, 34% are earning between
Rs.15,001/- to Rs.25,000/-, 35% are having income from
Rs.25,001/-to Rs.35,000/- and 21% are earning
`35,001/- and above. The sample collects information
from almost all types of the consumers who are engaged
in different occupations. Self-employed means
businessmen as well as professionals having their own
practice like doctors and chartered accountant doing
their own practice. The percentage of respondents who
are self employed is 28%. Salaried people cover
serviceman as well as respondents working as
Rashmi, 513
Table 1. Brands being used by the respondents for Toothpaste and their respective share
Toothpaste
Brand Share
Colgate 54.3
Pepsodent 24.9
Close up 8.1
Dabur Red 3.2
Miswak 2.4
Babool 1.9
Cibacaa 1.4
Glister (Amway) 1.4
Anchor 0.8
Thermoseal 0.8
Promise 0.5
Vicco 0.3
Total 100
executives or on other higher posts. The percentage of
salaried people is 35%. Housewives account for 17%,
students 15% and retired personnel are only 5%.
The Structure
A large number of factors persuade the brand choice
behaviour of the consumers. Brand specific attributes of
toothpaste based on experts’ advice and own intuition,
various sales promotion variables extracted from the
previous literature and demographic characteristics of
the consumers are observed to determine their effect on
brand choice decision of the consumers.
Brand choice
Brand choice of the consumers is taken as the
dependent variable of the study. Consumers are asked
to state their brand choice in terms of present brand
used for toothpaste. As per Table 1, total 12 brands are
mentioned by the respondents with the highest share for
Colgate (54.3%). Thus Colgate is taken as the most
preferred brand for this analysis. In actual market,
Colgate is also the market leader with 63%
1
www.smh.com.au/articles/2003/1/26/1069825840957.ht
ml?from=storyrhs of market share in the toothpaste
market.
In order to examine the brand choice behaviour of the
consumers for the brand with highest share, or most
preferred brand by the respondents, this variable is
converted into a dummy variable. The value for those
respondents is taken as 1 who mentioned Colgate as
their present choice of brand and 0 for those
respondents who mentioned any other brand of
toothpaste as the present brand being used.
Sales promotion variables
Sales promotion variables are also included in this study
to make the results more useful for the marketing
managers so that they could design their marketing
strategy as per the needs and aspirations of the
consumers. The sales promotion variables examined in
this study is discount, free gifts, feature advertisement,
celebrity advertisement, and store display. Discount
means when the brand is available at a price less than
its actual price. Free gifts means when the brand offers
such gifts/items with the product for which it charges
nothing. Feature advertisement refers to that brand
whose advertisement discloses or concentrates upon the
features of the brand rather than other irrelevant
aspects. Celebrity advertisement means when the brand
is being endorsed by some celebrity. Store display
means the brand is prominently displayed in the store.
The above mentioned sales promotion variables are
measured with the help of some questions. Discount is
measured by asking the respondents, ‘The present
brand of toothpaste when purchased was available on
discount’. For measuring free gifts, respondents were
asked, ‘The present brand of toothpaste when
purchased, offered some free gifts.’ To measure feature
advertisements, respondents were asked, ‘The present
brand of toothpaste when purchased was feature
advertised (the attributes of the product were
advertised’. Celebrity advertisement was measured by
asking, ‘I purchased the present brand of toothpaste
because some celebrity advertised it’. Store display is
measured by asking, ‘The present brand of toothpaste
when purchased was on display (prominently placed in
store)’
In the above mentioned questions, respondents are
asked the condition of sales promotion variables
(discount, free gifts, feature advertisement, celebrity
514 Glo. Adv. Res. J. Manag. Bus. Stud.
Table 2. Results of Factor analysis
Statements Functional aspect Ingredients Value Easiness
It has a good flavour. 0.517
It prevents tooth decay. 0.667
It gives me a feeling of freshness. 0.888
It has a good whitening power. 0.809
It is sweet in taste. 0.547
It is vegetarian toothpaste. 0.526
It has herbal ingredients. 0.843
The price of this brand is reasonable. 0.828
The manufacturer is reputed one. 0.649
My doctor recommended it. 0.729
It is easily available. 0.827
I am familiar with this brand. 0.806
Table 3. Description of demographic variables
Demographic variables Description
Gender 1 if the respondent is male, 0 otherwise.
Marital Status 1 if the respondent is married, 0 otherwise.
Age Actual age of respondent
Income 1, if income is up to `15,000/-.
2, if income is between `15,001/- to `25,000/-.
3, if income is between `25,001/- to `35,000/-.
4, if income is `35001/- or above.
Education 1, if the respondent is under graduate.
2, if the respondent is graduate.
3, if the respondent is postgraduate/professionally qualified.
Occupation (self employed as base
category)
1, if the respondent is salaried, 0 otherwise.
1, if the respondent is housewife, 0 otherwise.
1, if the respondent is student, 0 otherwise.
1, if the respondent is retired personnel, 0 otherwise.
advertisement and store display) one by one at the time
of purchase. The value of the variable is taken as 1 if it is
present at the time of purchase and 0 otherwise. If the
respondent has purchased the brand at a ‘discount’ it
has been allotted a value of 1, and 0 otherwise. Further
if the respondent has purchased that brand which offers
certain free gift with it then the variable ‘free gifts’ has a
value of 1, otherwise 0. If according to the respondent,
the brand is ‘feature advertised’, its value is 1, 0
otherwise. If the brand is endorsed by some celebrity, its
value is 1, 0 otherwise. Lastly if the respondent has
purchased the brand because the brand is prominently
displayed in the store, its value is 1, otherwise 0.
In all the above mentioned questions, the respondents
are supposed to answer in ‘yes’ or ‘no’ only.
Brand-specific attributes
Brand specific attributes play an important role in the
choice of the brand. Thus specific attributes of
toothpaste are included in this study which are based on
experts’ advice and own intuition. The respondents were
to rate these attributes on a seven-point likert scale
ranging from ‘highly satisfied’ to ‘highly dissatisfied’.
Total twelve attributes are examined which are in
relation to the particular brand of toothpaste being used.
Thus, actually these scale items are asked in the context
of the present brand being used by the respondents.
These scale items are factor analysed. Rotated
component matrix with varimax rotation has been
employed to extract factors appropriate for representing
brand specific attributes. The total variance explained is
60%. This percentage of variance is acceptable since
the satisfactory percentage of variance explained in
social sciences is 60% (Hair, et al. 2005). The scale
items result into four factors. Factor score of the
resultant factors is used for further analysis. Only those
items could be considered whose factor loading is
greater than .05 (Hair et al. 2005). Thus all the items are
considered for further analysis since no item has value
less than .05. The results of factor analysis are
presented in Table 2.
Consumer demographics
Consumer demographics are also taken into account to
analyse brand choice behaviour. Consumers differ from
one another because of the demographic traits. Hence
their choice process is also likely to be different. Thus it
becomes necessary to include the demographic traits as
explanatory variables in the brand choice model. The
various demographic characteristics examined in this
analysis along with their measurement criteria are listed
in Table 3.
Hypothesis Development
Three categories of variables are examined to determine
their influence on brand choice behaviour of the
consumers. Significance level of these variables is
checked through the following hypothesis.
Sales promotion variables
Various sales promotion variables examined in the study
are discount, free gifts, feature advertisement, celebrity
advertisement and store display. Hypothesis set are as
under:
H
1a
: Discount and brand choice have significant
relationship.
H
1b
: Free gifts and brand choice have significant
relationship.
H
1c
: Feature advertisement and brand choice have
significant relationship.
H
1d
: Celebrity advertisement and brand choice have
significant relationship.
H
1e
: Store display and brand choice have significant
relationship.
Brand-specific attributes
The following hypotheses are built to check the
significance of brand-specific attributes on brand choice
of toothpaste.
H
2a
: Functional aspect and brand choice have significant
relationship.
H
2b
: Ingredients and brand choice have significant
relationship.
H
2c
: Value and brand choice have significant
relationship.
H
2d
: Easiness and brand choice have significant
relationship.
Consumer demographics
The hypothesis for demographic traits of gender, marital
status, age, income, education and occupation are as
under:
Rashmi, 515
H
3a
: Gender and brand choice have significant
relationship.
H
3b
: Marital Status and brand choice have significant
relationship.
H
3c
: Age and brand choice have significant relationship.
H
3d
: Income and brand choice have significant
relationship.
H
3e
: Education and brand choice have significant
relationship.
H
3f
: Salaried (Occupation) and brand choice have
significant relationship.
H
3g
: Housewife (Occupation) and brand choice have
significant relationship.
H
3h
: Student (Occupation) and brand choice have
significant relationship.
H
3i
: Retired Personnel (Occupation) and brand choice
have significant relationship.
Logit Analysis
Logit analysis has been used to determine the
relationship between a binary dependent variable and
multiple independent variables. As already discussed, if
the respondent has purchased the most preferred brand
he/she is allotted a value of 1, 0 otherwise. This variable,
by taking a binary value of either 1 or 0, becomes the
dependent variable. Further, factor scores of brand-
specific attributes consisting of functional aspect,
ingredients, value and easiness act as the independent
variables of the study. Dummy value of sales promotion
variables consisting of discount, free gifts, feature
advertisement, celebrity advertisement and store display
are other independent variables. Moreover demographic
traits of gender, marital status, age, income, education
and occupation are also added to the analysis as
independent variables.
The model of consumer’s decision of buying the most
preferred brand takes the following form:
Log [P
i
/1 – P
i
] = ? + ß X
j
+ ?U
ij
+ ?Z
i
+ µ
i
Equation: 1
where i (i=1,2, _ _ _, n)) refers to the individual; j refers
to the current brand used by individual i; ?
is constant; X
j
is a vector of sales promotion variables; U
ij
is a vector of
brand-specific attributes; Z
i
is a vector of demographic
variables; ß, ? and ? are the coefficients of X
j
, U
ij
and Z
i
respectively; µ
ij
refers to the error term which captures all
misspecifications associated with a given individual and
brand being used. P
i
is the probability of buying the
most preferred brand; and 1 - P
i
is the probability of not
buying the most preferred brand. Log [P
i
/1 – P
i
] is log-
odds ratio, that is, the natural logarithm of the odds that
most preferred brand will be bought by a particular
individual.
After specifying the values for explanatory variables
the above specified model is tested. The method of
516 Glo. Adv. Res. J. Manag. Bus. Stud.
Table 4. Logit results
Description of variables Variable Toothpaste Coefficient (T ratio)
Constant 10.91 (6.61)*
Sales promotion Variables Discount -0.99 (-2.02)****
Free gifts 1.24 (1.28)
Feature advertisement -0.52 (-2.57)**
Celebrity advertisement -0.70 (-0.54)
Store display 0.17 (0.04)
Brand-specific attributes Functional aspect 1.81 (2.95)**
Ingredients 0.74 (2.87)**
Value 0.96 (2.59)***
Easiness -0.56 (-0.32)
Consumer demographics Gender 0.29 (2.14)****
Marital Status 0.47 (1.95)****
Age 0.73 (1.17)
Income 1.52 (1.94)****
Education 0.89 (3.45)*
Salaried (Occupation) -0.99 (-1.35)
Housewife(Occupation) 0.48 (2.54)***
Student(Occupation) 0.74 (0.75)
Retired Personnel (Occupation) 0.99 (1.12)
Correct predictions 88%
Note: *,**,***,**** represent highly significant, significant at 1%, 5% and 10% respectively.
maximum likelihood has been used wherein that value of
the coefficients is taken which maximizes the following
log likelihood function (Madnani, 1994):
}] Z U X { F 1 [ log ) Yi 1 ( 1 ] } Z U X { F [ log Yi L
i ij j
n
1 i
i ij j
n
1 i
? + ? + ? + ? ? ? ? + ? + ? + ? + ? ? =
= =
Equation: 2
where L = the log- likelihood function to be maximized
and Yi is dummy variable taking value 1 if most preferred
brand is purchased and 0 otherwise.
The outcome of above stated model would give the
estimated values of coefficients of explanatory variables
that is, sales promotion variables (discount, free gifts,
feature advertisement, celebrity advertisement, store
display), brand-specific attributes (functional aspect,
ingredients, value, easiness) and demographic variables
(gender, marital status, age, income, education,
occupation) which will depict the value of change in log
odds ratio of the purchase probability of most preferred
brand.
Thus the above stated model depicts the probability of
buying the most preferred brand for each consumer.
DATA ANALYSIS
The logit model is used to estimate the probability of
buying the most preferred brand by the respondents.
Table 4 depicts that 88% of the observations are
correctly predicted.
Table 4 reveals that all variables except free gifts,
celebrity advertisement, store display, age, salaried,
student and retired personnel do not play a significant
role in brand choice decision. Thus H
1a
(discount), H
1c
(feature advertisement), H
2a
(functional aspect), H
2b
(ingredients), H
2c
(value), H
3a
(gender), H
3b
(marital
status), H
3d
(income), H
3e
(education), H
3g
(housewife)
are accepted whereas H
1b
(free gifts), H
1d
(celebrity
advertisement), H
1e
(store display), H
2d
(easiness), H
3c
(age), H
3f
(salaried), H
3h
(student), H
3i
(retired personnel)
are rejected.
The most important variable affecting brand choice
decision is functional aspect of toothpaste with the
highest coefficient value (1.81). The least influencing
factor is marital status (0.47)
DISCUSSION
Among the sales promotion variables, the most
influencing variable is discount (-0.99). Discounted
brands can attract the attention of the consumer and
he/she can swing from his/her last purchased brand
which may be even the top brand of the market. A price
conscious consumer buys the discounted brands to save
money or to store a large quantity for future. The product
category chosen in this study is largely affected due to
discount strategy of the consumers. However the
marketing managers must be wide awake while drawing
discount strategy. While explaining about promotion
techniques, Alvarez and Casielles (2005) suggest the
marketing managers that the sales promotion techniques
(discount mainly) will be more result oriented when
consumers are not expecting it.
In case of toothpaste, feature advertisement (-0.52)
affect brand choice decision which means that
consumers are willing to know about the features of
various brands of toothpaste so that they could take a
right brand choice decision. As celebrity advertisement is
not important for the consumers, hence consumers are
required to be informed, no matter who provides this
information. Further free gifts and display are also not
important for the consumer while buying toothpaste.
Bhusghan and Daftari (2007) have suggested one new
mechanism – modern trade that offers better displays
and ambience as compared to kiryana stores to push the
sales of brand up.
Regaining brand specific attributes, in case of
toothpaste, all the factors work well except easiness.
Consumers are willing to buy the most preferred brand
because of its functional aspect (1.81), ingredients (0.74)
and value (0.96). However easiness is not effective in
influencing the purchase of most preferred brand. The
most important attribute for the consumers is functional
aspect of toothpaste. The reason may be the utilitarian
nature of the product as any wrong purchase could
result in serious long lasting dental problems. While
discussing the role of advertising in generation of brand
attitudes, Baker (2001) suggests that advertising must
maximise specific brand information that can be used by
the consumers to make a differentiation between various
alternatives so that they could choose the best one.
In this study various demographic variables are also
examined and significant results are found. Results
depict that male (0.29), married (0.47) and housewives
(0.48) are willing to buy the most preferred brand.
Increase in income and increase in education also
increases the probability of buying the most preferred
brand.
CONCLUSION
This article investigates the reasons that why consumers
buy a particular brand more because of which it
becomes the top or leading brand. Influencing factors
extracted from the results could be used by marketing
managers to make their brand as the leading brand.
REFERENCES
Alvarez BA, Casielles RV (2005). Consumer Evaluation of Sales
Promotion: The Effect on Brand Choice. European Journal of
Marketing 39 (½): 54-70.
Baker WE (2001). “The Diagnosticity of Advertising Generated Brand
Attitudes in Brand Choice Contexts”, Journal of Consumer
Psychology, Vol 11 No 2, pp. 129-139.
Banerjee A, Awasthy D, Gupta V (2005). A Choice Modelling
Approach to Evaluate Effectiveness of Brand Development
Initiatives. International Journal of Management and Decision
Making, 6(2): 180-198.
Rashmi, 517
Bhusghan R, Daftari I (2007). Regional Brand to Cash in on Modern
Trade. The Economics Times (Business of Brands), p 3.
Chintagunta PK, Jain DC, Vilcassim NJ (1991). Investigating
Heterogeneity in Brand Preferences in Logit Model for Panel Data.
Journal of Marketing Research, 28(4): 417-428.
Coskunoglu O, Hansotia BJ, Shaikh MA (1985). A New Logit Model for
Decision-Making and Its Application. Journal of the Operational
Research Society, 36(1): 35-41.
Degeratu AM, Rangaswamy A, Wu J (2000). Consumer Choice
Behaviour in Online and Traditional Supermarkets: The Effects of
Brand Name, Price and Other Search Attributes. International
Journal of Research in Marketing, 17: 55-78.
Fishburn P (1967). “Methods of Estimating Additive Utilities”,
Management Science, Vol 13 No 7, pp. 435-453.
Fishburn P (1968). “Utility Theory”, Management Science, Vol 14 No 5
(Theory Series), pp. 335-378.
Guadagni PM, Little JDC (1983). A Logit Model of Brand Choice
Calibrated on Scanner Data. Marketing Science, 2(3): 203-238.
Hair, Joseph F Jr, Anderson, Rolph E, Tatham, Ronald L, Black,
William C (2005). Multivariate Data Analysis, Sixth Edition, Pearson
Education.
Kalyanam K, Putler DS (1997). Incorporating Demographic Variables
in Brand Choice Models: An Invisible Alternatives Framework.
Marketing Science, 16 (2): 166-181.
Kraft FB, Granbois DH, Summers JO (1973). Brand Evaluation and
Brand Choice: A Longitudinal Study. Journal of Marketing Research,
10 (3): 235-241.
Krishnamurthi L, Raj SP (1988). A Model of Brand Choice and
Purchase Quantity Price Sensitivities. Marketing Science, 7 (1): 1-
20.
Madnani GMK (1994). Introduction to Econometrics, 6
th
Ed., Oxford
and IBH Publishing Co. Private. Ltd.
McGuire TW, Weiss DL (1976). Logically Consistent Market Share
Models. Journal of Marketing Research, 13(3): 296-302.
Murthi BPS, Srinivasan K (1999). Consumers’ Extent of Evaluation in
Brand Choice. Journal of Business, 72 (2): 229-256.
Panda TK (2005). “Predicting Behavioral Intention for Purchase and
Positioning of a New Brand”, Journal of Marketing and
Communication, Vol 1 No 2, pp. 53-67.
Reibstein DJ (1978) “The Prediction of Individual Probabilities of Brand
Choice”, Journal of Consumer Research, Vol 5 No 3, pp. 163-168.
Singh VP, Hansen KT, Gupta S (2005). Modeling Preferences for
Common Attributes in Multicategory Brand Choice. Journal of
Marketing Research, 42(2): 195-209.
Slovic P, Fleisnner D, Bauman WS (1972). Analyzing the Use of
Information in Investment Decision-Making: A Methodological
Proposal. Journal of Business, 45(2): 283-301.
Theil H (1969). “A Multinomial Extension of the Linear Logit Model”,
International Economic Review, Vol 10 No 3, pp. 251-259.
Turban E, Metersky ML (1971). Utility Theory Applied to Multi-variable
System Effectiveness Evaluation. Management Science, 17(12)
Application Series: B817-B828.
doc_977516843.pdf