Diapositivo 1 - PowerPoint PPT Presentation

1 / 25
About This Presentation
Title:

Diapositivo 1

Description:

perceived quality, satisfaction, trust, symbolic image, and ... AVE1/2 corr LV. Structural model. Perceived quality. Satisfaction. R2 = 54.2% Q2 = 0.389 ... – PowerPoint PPT presentation

Number of Views:64
Avg rating:3.0/5.0
Slides: 26
Provided by: nini157
Category:
Tags: ave1 | diapositivo

less

Transcript and Presenter's Notes

Title: Diapositivo 1


1
Segmentation Using Finite Mixture Partial Least
Squares loyalty and satisfaction perceived by
customers of the supermarkets in Portugal
Sandra M. C. Loureiro
2
Objective Understand the relationship among
perceived quality, satisfaction, trust, symbolic
image, and how they impact loyalty in Portuguese
supermarket service.
Using PLS approach Applying the finite mixture
partial least squares (FIMIX-PLS)
Segmentation
Combines a finite mixture procedure with an
expectation-maximization (EM)-algorithm
3
Method
  • Questionnaire
  • Items of the latent variables
  • Socio-demographic variables
  • (ten-point agree/disagree response format)

4
Variables
The overall judgment made by the consumer
(Crompton Love 1995 Cronin Taylor 1994).
Perceived quality
An emotional or cognitive judgment of the
supermarket service (Brady et al., 2002 Loureiro
Miranda, 2008).
Satisfaction
Confidence in supermarket service delivered,
fulfill its promises, and has a real concern with
the consumers needs (Morgan Hunt, 1994
Garbarino Johnson, 1999 Singh Sirdeshmukh,
2000).
Trust
Symbolic benefits and elements of brand
personality to build the symbolic image of
supermarket perceived by customers (Aaker, 1996
Biel, 1992 Keller, 1993) .
Symbolic image
Components behavior and attitude (Zeithaml et
al., 1996 Baker Crompton, 2000).
Loyalty
5
Social concern
Proposed model
Attendance
Store organization
Prestige
Solidity
Perceived quality
Honesty
I intend to continue to buy with the same
frequency in the store
Symbolic image
Dynamism
I speak well about this store to other people

Loyalty
Variety of products and brands





Satisfies my needs
I will recommend the store if someone ask for my
advice
Satisfaction
Trust

It is one of the best stores
I trust on service delivered from store
There are a real concern to my needs
The promises are fulfilled
Delivers an excellent service
I feel confidence in the quality of the products
6
Results
First model estimation
Measurement model
Structural model
Second segmentation (FIMIX-PLS)
Model selection
Global model and disaggregate results for two
latent segments
7
Measurement model
Prestige
Social concern
Honesty
Attendance
Solidity
0.745
0.764
Store organization
0.877
0.805
0.793
0.880
Perceived quality
Dynamism
I intend to continue to buy with the same
frequency in the store
Symbolic image
0.842
0.772
0.781
Variety of products and brands
Item loading gt 0.707
Loyalty
0.880
I speak well about this store to other people
0.896
Satisfaction
Satisfies my needs
Trust
I will recommend the store if someone ask for my
advice
0.825
0.831
0.872
0.836
0.763
0.895
I trust on service delivered from store
There are a real concern to my needs
0.847
The promises are fulfilled
It is one of the best stores
Delivers an excellent service
I feel confidence in the quality of the products
8
AVE gt 0.5
Discriminant validity AVE1/2 gt corr LV
Reliability gt 0.8
9
Structural model
Symbolic image R2 56.3 Q2 0.343
Perceived quality
0.491
0.430
34.5
Loyalty R2 53.9 Q2 0.381
30.7
0.456
0.736
32.2
0.356
54.2
0.300
24.1
19.4
Trust R2 58.0 Q2 0.392
Satisfaction R2 54.2 Q2 0.389
0.387
27.2
GoF 0.622
p lt 0.01 p lt 0.001
Good fit Tenenhaus et al. (2005)
10
Second segmentation (FIMIX-PLS)
1st seg. 68 2ndseg. 32
Model selection
All relevant evaluation criteria considerably
decrease in the ensuing numbers of segments
Each additional segment has only a small size,
11
Multi-group comparison test
This test uses the path coefficients and the
standard errors of the structural paths
calculated by PLS with the samples of the two
segments, using the following expression of
t-value for multi-group comparison test
12
Global model and disaggregate results for two
latent segments
p lt 0.01 p lt 0.001 ns no significant.
tmgp t-value for multi-group comparison test
13
Conclusions
1st segment Customers live mainly in Lisbon and
other urban cities in coast. Most is young
adult, married, with medium or higher education.
The average number of members in family is 3.
Satisfaction and trust contribute significantly
to explain symbolic image of the supermarkets.
Symbolic image and trust explain 72.1 of the
variance in intention to recommend and to
continue to buy in the supermarket store that
they usually frequent.
14
2nd segment Customers live mainly in inland
(interior North of Portugal). Most is married
adults and 44 has more than 50 years, with
medium education. The average number of member
in family is three or four. For these
customers, satisfaction doesnt contribute
significantly to explain symbolic image of the
supermarkets. Quality contributes more to
explain trust than satisfaction. Symbolic image
and trust explain only 30.7 of the variance in
intention to recommend and to continue to buy in
the supermarket store that they usually frequent.
15
Portugal
Porto
2nd Segment
1st Segment
Married adults 44 has more than 50 years, with
medium education 3 or 4 members.
Lisbon
Young adult, married, with medium or higher
education 3 members
16
Thank you
17
(No Transcript)
18
Descriptive statistics
All sample
19
(No Transcript)
20
All sample
The demographic profile of the interviewed
inhabitants of the Portugal
Simple size 1 043 Not completed 39 Completed 1
004
CATI (computer assisted telephone survey) was
used to conduct the survey Quota
sample Surveys were conducted at different times
of day to ensure working and non-working members
of the households had equal chances of being
present.
universe data of General Recenseamento of the
Population of 2001 were used supplied by INE
(CENSUS 2001)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
Write a Comment
User Comments (0)
About PowerShow.com