Messung und statistische Analyse von Kundenzufriedenheit KF Qualitätsmanagement Vertiefungskurs V

Messung und statistische Analyse von Kundenzufriedenheit

KF Qualitätsmanagement Vertiefungskurs V

3.12.2004 Messung & Analyse von Kundenzufriedenheit

2

Outline

Customer satisfaction measurement The Structural Equation Model (SEM) Estimation of SEMs Evaluation of SEMs Practice of SEM-Analysis


3

The ACSI Model

Ref.: http://www.theacsi.org/model.htm


4

ACSI-Model: Latent Variables Customer Expectations: combine customers’

experiences and information about it via media, advertising, salespersons, and word-of-mouth from other customers

Perceived Quality: overall quality, reliability, the extent to which a product/service meets the customer’s needs

Customer Satisfaction: overall satisfaction, fulfillment of expectations, comparison with ideal

Perceived Value: overall price given quality and overall quality given price

Customer Complaints: percentage of respondents who reported a problem

Customer Loyalty: likelihood to purchase at various price points


5

Baseline

*

Q2 1995

Q2 1996

Q2 1997

Q2 1998

Q2 1999

Q2 2000

Q2 2001

Q2 2002

Q2 2003

Q22004

% Changes

% Changes

MANUFACTURING/DURABLES

79.2 79.8 78.8 78.4 77.9 77.3 79.4 78.7 79.0 79.2 78.3 -1.1% -1.1%

Personal Computers 78 75 73 70 71 72 74 71 71 72 74 2.8% -5.1%

Apple Computer, Inc. 77 75 76 70 69 72 75 73 73 77 81 5.2% 5.2%

Dell Inc. NM NM NM 72 74 76 80 78 76 78 79 1.3% 9.7%

Gateway, Inc. NM NM NM NM 76 76 78 73 72 69 74 7.2% -2.6%

All Others NM 70 73 72 69 69 68 67 70 69 71 2.9% 1.4%

Hewlett-Packard Company – HP 78 80 77 75 72 74 74 73 71 70 71 1.4% -9.0%

Hewlett-Packard Company – Compaq 78 77 74 67 72 71 71 69 68 68 69 1.5% -11.5%


6

The European Customer Satisfaction Index (ECSI)

Ref.: http://www.swics.ch/ecsi/index.html


7

ACSIe-Model for Food Retail

Custo-mer Satis-

faction

LoyaltyExpec-tations

PerceivedQuality

Value

EmotionalFactor

Hackl et al. (2000)Latent variables and path coefficients


8

Austrian Food Retail Market Pilot for an Austrian National CS Index (Zuba, 1997) Data collection: December 1996 by Dr Fessel & GfK

(professional market research agency) 839 interviews, 327 complete observations Austria-wide active food retail chains (1996: ~50%

from the 10.5 B’EUR market)Billa: well-assorted medium-sized outletsHofer: limited range at good pricesMerkur: large-sized supermarkets with comprehensive range Meinl: top in quality and service


9

The Data

Indicators Latent

total expected quality (EGESQ), expected compliance with demands (EANFO), expected shortcomings (EMANG)

Expectations (E)

total perceived quality (OGESQ), perceived compliance with needs (OANFO), perceived shortcomings (OMANG)

Perceived Quality (Q)

value for price (VAPRI), price for value (PRIVA) Value (P)

total satisfaction (CSTOT), fulfilled expectations (ERWAR), comparison with ideal (IDEAL)

Customer Sa-tisfaction (CS)

number of oral complaints (NOBES), number of written complaints (NOBRI)

Voice (V)

repurchase probability (WIEDE), tolerance against price-change (PRVER)

Loyalty (L)


10

The Emotional Factor

Principal component analysis of satisfaction driversstaff (availability, politeness) outlet (make-up, presentation of merchandise, cleanliness)range (freshness and quality, richness)price-value ratio (value for price, price for value)customer orientation (access to outlet, shopping hours, queuing time for checkout, paying modes, price information, sales, availability of sales)

identifies (Zuba, 1997)staff, outlet, range: “Emotional factor”price-value ratio: “Value”customer orientation: “Cognitive factor”


11

Structural Equation Models

Combine three concepts Latent variables

Pearson (1904), psychometrics Factor analysis model

Path analysis Wright (1934), biometrics Technique to analyze systems of relations

Simultaneous regression models Econometrics


12

Customer Satisfaction

Is the result of the customer‘s comparison of his/her expectations with his/her experiences

has consequences on loyalty future profits of the supplier


13

Expectation vs. Experience

Expectation reflects customers‘ needs offer on the market image of the supplier etc.

Experiences include perceived performance/quality subjective assessment etc.


14

CS-Model: Path Diagram

Custo-mer Satis-

faction

LoyaltyPerceived

Quality

Expecta-tions


15

A General CS-Model

Custo-mer Satis-

faction

LoyaltyPerceived

Quality

Expecta-tions

Voice

Profits


16

CS-Model: Structure

tofrom

EX PQ CS LY

EX X X 0

PQ 0 X 0

CS 0 0 X

LY 0 0 0

EX: expectationPQ: perceived qualityCS: customer satisfactionLY: loyalty

Recursive structure: triangular form of relations


17

CS-Model: Equations

PQ = 1 + 11EX + 1

CS = 2 + 21PQ + 21EX + 2

LY = 3 + 32CS + 3

Simultaneous equations model in latent variables

Exogenous: EXEndogenous: PQ, CS, LYError terms (noises): 1, 2, 3


18

Simple Linear RegressionModel: Y = + X + Observations: (xi, yi), i=1,…,n

Fitted Model: Ŷ = a + cXOLS-estimates a, c:

minimize the sum of squared residuals

sxy: sample-covariance of X and Y

2,

ˆ( ) ( , ) mini iiy y S

2 ,xy

x

s

sc a y cx s


19

Criteria of Model FitR2: coefficient of determination

the squared correlation between Y and Ŷ: R2 = ryŷ

2

t-Test: Test of H0: =0 against H1:≠0 t=c/s.e.(c)

s.e.(c): standard error of cF-Test: Test of H0: R2=0 against H1: R2≠0

follows for large n the F-distribution with n-2 and 2 df

2

2

2

1 2

RF

R n


20

Multiple Linear RegressionModel: Y = + X1+ + Xk+= + x’ + Observations: (xi1,…, xik, yi), i=1,…,nIn Matrix-Notation: y = + X + y, : n-vectors, :k-vector, X: nxk-matrixFitted Model: ŷ = a + XcOLS-estimates a, c:

R2 = ryŷ2

F-Testt-Test

11 1( ' ) ' , ... k kc X X X y a y c x c x


21

Simultaneous Equations ModelsA 2-equations model:

PQ = 1 + 11EX + 1

CS = 2 + 21PQ + 21EX + 2

In matrix-notation: Y = BY + X + with 1

2

1 11

21 2 21

, ,

0 0,

0

PQY X EX

CS

B

path coefficients


22

Simultaneous Equations ModelsModel: Y = BY + X +

Y, : m-vectors, B: (mxm)-matrix :(mxK)-matrix, X: K-vector

Problems: Simultaneous equation bias: OLS-estimates of

coefficients are not consistentIdentifiability: Can coefficients be consistently

estimated?

Some assumptions:

: E()=0, Cov() =

Exogeneity: Cov(X,) = 0


23

Path Analytic Model

CS

PQ

EX

Var(1) = EX2

21 1

22 2

0

0Var

PQ = 11EX + 1

CS = 21PQ + 21EX + 2


24

Path Analysis Wright (1921, 1934) A multivariate technique Model: Variables may be

structurally related structurally unrelated, but correlated

Decomposition of covariances allows to write covariances as functions of structural parameters

Definition of direct and indirect effects


25

Example

CS

PQ

EX

CS,EX = 21EX + 21PQ,EX

= 212EX + 11212

EX

CS,EX = 21 + 1121 with standardized variable EX:

( )YX Yi iXi Y X


26

Direct and Indirect Effects

CS,EX = 21 + 1121

Direct effect: coefficient that links independent with dependent variable; e.g., 21 is direct effect of EX on CS

Indirect effect: effect of one variable on another via one or more intervening variable(s), e.g., 1121

Total indirect effect: sum of indirect effects between two variables

Total effect: sum of direct and total indirect effects between two variables


27

Decomposition of Covariance yx

( )YX YI IXI Y X

( )I Y X : variable on path from X to Y

YI: path coefficient of variable I to Y


28

First Law of Path Analysis

Decomposition of covariance xy between Y and X:

Assumptions: Exogenous (X) and endogenous variables (Y) have

mean zero Errors or noises ()

have mean zero and equal variances across observations

are uncorrelated across observations are uncorrelated with exogenous variables are uncorrelated across equations

( )YX Yi iXi Y X


29

Identification

PQ = 11EX + 1 Y1 = 11X + 1

CS = 21PQ + 21EX + 2 Y2 = 21Y1 + 21X + 2

In matrix-notation: Y = BY + X +

Number of parameters: p=6Model is identified, if all parameters can be expressed

as functions of variances/covariances of observed variables

211 2 1

221 21 2

0 0 0, , ( ) ,

0 0EXB


30

Identification, cont’dY1 = 11X + 1

Y2 = 21Y1 + 21X + 2

1X =11 X2

2X = 211X + 21X2

21 = 2112 + 211X

X2 = X

2

y12 = 111X+1

2

y22 = 2121 + 212X+2

2

p=6

first 3 equations allow unique solution for pathcoefficients, last three forvariances of and


31

Condition for Identification Just-identified: all parameters can be uniquely

derived from functions of variances/covariances Over-identified: at least one parameter is not

uniquely determined Under-identified: insufficient number of

variances/covariances

Necessary, but not sufficient condition for identification: number of variances/covariances at least as large as number of parameters

A general and operational rule for checking identification has not been found


32

Latent variables and Indicators

Latent variables (LVs) or constructs or factors are unobservable, but

We might find indicators or manifest variables (MVs) for the LVs that can be used as measures of the latent variable

Indicators are imperfect measures of the latent variable


33

Indicators for “Expectation”

EX

E3

E2

E1

E1: When you became a customer of AB-Bank, you probably knew something about them. How would you grade your expectations on a scale of 1 (very low) to 10 (very high)?

E2: Now think about the different services they offer, such as bank loans, rates, … Rate your expectations on a scale of 1 to 10?

E3: Finally rate your overall expectations on a scale of 1 to 10?

1

2

3

From: Swedish CSB Questionnaire, Banks: Private Customers

E1, E2, E3: „block“ of LVs for Expectation


34

Notation

X3

X2

X1

1

2

3

1

2

3

X1=1+1

X2=2+2

X3=3+3

: latent variable, factorXi: indicators, manifest

variablesi: factor loadingsi: measurement errors, noise

Some properties: LV: unit variancenoise i: has mean zero, variance i

2, uncorrelated with other noises

“reflective” indicators


35

Notation

X3

X2

X1

1

2

3

1

2

3

X1=1+1

X2=2+2

X3=3+3

X = +

In matrix-notation:

with vectors X, , and e.g., X = (X1, X2, X3)‘

: latent variable, factorXi: indicators, manifest

variablesi: factor loadingsi: measurement error, noise


36

CS-Model: Path Diagram

CS

PQ

EX

Q1

Q2

Q3

E3

E2

E1

C1

C2

C3

1

3

2

3

2

1

4

5

6


37

SEM-Model: Path Diagram

1

Y1

Y2

Y3

X3

X2

X1

Y4

Y5

Y6

1

3

2

3

2

1

4

5

6

=++

X = x+Y= y+


38

SEM-Model: Notation

211 2 1

221 21 2

0 0 0, , ( ) ,

0 0EX

=++

X = x+Y= y+

11 12 1311 12 13

11 12 13

0 0 0, , ,

0 0 0x y

X, : 3-component vectorY, : 6-component vector

Inner relations, inner model

Outer relations, measurement model


39

Statistical Assumptions Error terms of inner model () have

zero means constant variances across observations are uncorrelated across observations are uncorrelated with exogenous variables

Error terms of measurement models () have zero means constant variances across observations are uncorrelated across observations are uncorrelated with latent variables and with each

other Latent variables are standardized


40

Covariance Matrix of Manifest VariablesUnrestricted covariance matrix (order: K = kx+ky)

= Var{(X’,Y’)’}

Model-implied covariance matrix 1 2

2 3

1 11

12

3

( ) , ( , , , , , , , )

( ) ( )[( ) ]

( ) [ ]

[ ]

x y

x y

y x

x x

A A

A A

A I I

A I

A


41

Estimation of the Parameters Covariance fitting methods

search for values of parameters so that the model-implied covariance matrix fits the observed unrestricted covariance matrix of the MVs

LISREL (LInear Structural RELations): Jöreskog (1973), Keesling (1972), Wiley (1973)

Software LISREL by Jöreskog & Sörbom PLS techniques

partition of in estimable subsets of parameters iterative optimizations provide successive

approximations for LV scores and parameters Wold (1973, 1980)


42

Discrepancy FunctionThe discrepancy or fitting function

F(S;) = F(S; )is a measure of the “distance” between the model-implied covariance-matrix and the estimated unrestricted covariance-matrix S

Properties of the discrepancy function: F(S;) ≥ 0; F(S;) = 0 if S=


43

Covariance Fitting (LISREL) Estimates of the parameters are derived by

F(S;) min

Minimization of (K: number of indicators)

F(S;) = log|| – log|S| + trace (S-1) – K

gives ML-estimates, if the manifest variables are independently, multivariate normally distributed

Iterative Algorithm (Newton-Raphson type) Identification Choice of starting values is crucial Other choices of F result in estimation methods like OLS and

GLS; ADF (asymptotically distribution free)


44

PLS Techniques Estimates factor scores for latent variables Estimates structural parameters (path coefficients,

loading coefficients), based on estimated factor scores, using the principle of least squares

Maximizes the predictive accuracy “Predictor specification”, viz. that E(|) equals the

systematic part of the model, implies E(|)=0: the error term has (conditional) mean zero

No distributional assumptions beyond those on 1st and 2nd order moments


45

The PLS-AlgorithmStep 1: Estimation of factor scores

1. Outer approximation2. Calculation of inner weights3. Inner approximation4. Calculation of outer weights

Step 2: Estimation of path and loading coefficients by minimizing Var() and Var()

Step 3: Estimation of location parameters (intercepts) Bo from = Bo + B + + o from X = o + x+


46

Estimation of Factor Scores

Factor i: realizations Yin, n=1,…,NYin

(o): outer approximation of Yin

Yin(i): inner approximation of Yin

Indicator Yij: observations yijn; j=1,…,Ji; n=1,…,N1. Outer approximation: Yin

(o)=jwijyijn s.t. Var(Yi(o))=1

2. Inner weights: vih=sign(rih), if i and h adjacent; otherwise vih=0; rih=corr(i,h) (“centroid weighting”)

3. Inner approximation: Yin(i)=hvihYhn

(o) s.t. Var(Yi(i))=1

4. Outer weights: wij=corr(Yij,Yi(i))

Start: choose arbitrary values for wij

Repeat 1. through 4. until outer weights converge


47

Example

CS

PQ

EX

Q1

Q2

Q3

E3

E2

E1

C1

C2

C3

1

3

2

3

2

1

4

5

6


48

Example, cont’dStarting values wEX,1,…,wEX,3,wPQ,1,…,wPQ,3,wCS,1,…,wCS,3 Outer approximation:

EXn(o) = jwEX,jEjn; similar PQn

(o), CSn(o);

standardizedInner approximation:

EXn(i) = + PQn

(o) + CSn(o)

PQn(i) = + EXn

(o) + CSn(o)

CSn(i) = + EXn

(o) + PQn(o)

standardizedOuter weights:

wEX,j = corr(Ej,EX(i)), j=1,…,3; similar wPQ,j, wCS,j


49

Choice of Inner WeightsCentroid weighting scheme: Yin

(i)=hvihYhn(o)

vij=sign(rih), if i and h adjacent, vij=0 otherwise

with rih=corr(i,h); these weights are obtained if vih are chosen to be +1 or -1 and Var(Yi

(i)) is maximized

Weighting schemes:h predecessor h successor

centroid sign(rih) sign(rih)

factor, PC rih rih

path bih rih

bih: coefficient in regression of i on h


50

Measurement Model: ExamplesLatent variables from Swedish CSB Model1. Expectation

E1: new customer feelings

E2: special products/services expectations

E3: overall expectation

2. Perceived QualityQ1: range of products/services Q2: quality of serviceQ3: clarity of information on products/services Q4: opening hours and appearance of locationQ5: etc.


51

Measurement ModelsReflective model: each indicator is reflecting the latent

variable (example 1)Yij = iji + ij

Yij is called a reflective or effect indicator (of i)Formative model: (example 2)

i = y'Yi + i

y is a vector of ki weights; Yij are called formative or cause indicators

Hybrid or MIMIC model (for “multiple indicators and multiple causes”)

Choice between formative and reflective depends on the substantive theory

Formative models often used for exogenous, reflective and MIMIC models for endogenous variables


52

Estimation of Outer Weights “Mode A” estimation of Yi

(o): reflective measurement model weight wij is coefficient from simple regression of Yi

(i) on Yij: wij = corr(Yij,Yi(i))

“Mode B” estimation of Yi(o): formative

measurement model weight wij is coefficient of Yij from multiple regression of Yi

(i) on Yij, j=1,…,Jimulticollinearity?!

MIMIC model


53

Properties of EstimatorsA general proof for convergence of the PLS-algorithm

does not exists; practitioners experience no problems

Factor scores are inconsistent but “consistent at large”: consistency is achieved with increasing sample size and block size

Loading coefficients are inconsistent and seem to be overestimated

Path coefficients are inconsistent and seem to be underestimated


54

ACSI Model: Results

Custo-mer Satis-

faction


PerceivedQuality

Value

Voice

EQS-estimates

PLS-estimates


55

Evaluation of SEM-Models Depends on estimation method

Covariance-fitting methods: distributional assumptions, optimal parameter estimates, factor indeterminacy

PLS path modeling: non-parametric, optimal prediction accuracy, LV scores

Step 1: Inspection of estimation results (R2, parameter estimates, standard errors, LV scores, residuals, etc.)

Step 2: Assessment of fit Covariance-fitting methods: global measures PLS path modeling: partial fitting measures


56

Inspection of Results Covariance-fitting methods: global optimization

Model parameters and their standard errors; do they confirm theory?

Correlation residuals: sij-sij() Graphical methods

PLS techniques: iterative optimization of outer models and inner model Model parameters Resampling procedures like blindfolding or jackknifing

give standard errors of model parameters LV scores Graphical methods


57

Fit Indices Covariance-fitting methods: covariance fit

measures such as Chi-square statistics Goodness of Fit Index (GFI), AGFI Normed Fit Index (NFI), NNFI, CFI Etc. Basis is the discrepancy function

PLS path modeling: prediction-based measures Communality Redundancy Stone-Geisser’s Q2


58

Chi-square Statistic Test of H0: = () against non-specified alternative Test-statistic X2=(N-1)F(S;( )) If model is just identified (c=p): X2=0 [c=K(K+1)/2, p:

number of parameters in ] Under usual regularity conditions (normal distribution,

ML-estimation), X2 is asymptotically 2(c-p)-distributed Non-significant X2 indicate: the over-identified model

does not differ from a just-identified version Problem: X2 increases with increasing N Some prefer X2/(c-p) to X2 (has reduced sensitivity to

sample size); rule of thumb: X2/(c-p) < 3 is acceptable


59

Goodness of Fit IndicesGoodness of Fit Index (Jöreskog & Sörbom):

Portion of observed covariances explained by the model-implied covariances

“How much better fits the model as compared to no model at all”

Ranges from 0 (poor fit) to 1 (perfect fit) Rule of thumb: GFI > 0.9 AGFI penalizes model complexity:

ˆ[ , ( )]1

[ , ( )]

F SGFI

F S O

ˆ[ , ( )]1

[ , ( )]

c F SAGFI

c p F S O


60

Other Fit Indices Normed Fit Index, NFI (Bentler & Bonett)

Similar to GFI, but compares with a baseline model, typically the independence model (indicators are uncorrelated)

Ranges from 0 (poor fit) to 1 (perfect fit) Rule of thumb: NFI > 0.9

Comparative Fit Index, CFI (Bentler) Less depending of sample size than NFI

Non-Normed Fit Index, NNFI (Bentler & Bonett) Also known as Tucker-Lewis Index Adjusted for model complexity

Root mean squared error of approximation, RMSEA (Steiger): ˆ[ , ( )] /( )RMSEA F S c p


61

Assessment of PLS Results Not a single but many optimization steps; not a

global measure but many measures of various aspects of results

Indices for assessing the predictive relevance Portions of explained variance (R2) Communality, redundancy, etc. Stone-Geisser’s Q2

Reliability indices NFI, assuming normality of indicators Allows comparisons with covariance-fitting

results


62

Some IndicesAssessment of diagonal fit (proportion of explained

variances) SMC (squared multiple correlation coefficient) R2:

(average) proportion of the variance of LVs that is explained by other LVs; concerns the inner model

Communality H2: (average) proportion of the variance of indicators that is explained by the LVs directly connected to it; concerns the outer model

Redundancy F2: (average) proportion of the variance of indicators that is explained by predictor LVs of its own LV

r2: proportion of explained variance of indicators


63

Some Indices, cont’dAssessment of non-diagonal fit Explained indicator covariances

rs = 1 c/swith c = rms(C), s = rms(S); C: estimate of Cov()

Explained latent variable correlation

rr = 1 q/rwith q = rms(Q), r = rms(Cov(Y)); Q: estimate of Cov()

reY = rms (Cov(e,Y)), e: outer residuals reu = rms (Cov(e,u)), u: inner residuals

rms(A) = (ij aij2)1/2: root mean squared covariances (diagonal elements of

symmetric A excluded from summation)


64

Stone-Geisser’s Q2

Similar to R2

E: sum of squared prediction errors; O: sum of squared deviations from mean

Prediction errors from resampling (blindfolding, jackknifing)

E.g., communality of Yij, an indicator of i

2 1E

QO

22

2

ˆ[ ( )]1

[ ]ijn ij inn

ijcijn ijn

y YQ

y y


65

Lohmöller’s Advice Check fit of outer model

Low unexplained portion of indicator variances and covariances

High communalities in reflective blocks, low residual covariances

Residual covariances between blocks close to zero Covariances between outer residuals and latent

variables close to zero Check fit of inner model

Low unexplained portion of latent variable indicator variances and covariances

Check fit of total model High redundancy coefficient Low covariances of inner and outer residuals


66

ACSI Model: Results

Custo-mer Satis-

faction


PerceivedQuality

Value

Voice

EQS-estimates

PLS-estimates


67

Diagnostics: EQS

ACSI ACSIe

247.5 378.7

df 81 173

NNFI 0.898 0.930

RMSEA 0.079 0.060


68

Diagnostics: PLS (centroid weighting)

ACSI ACSI e Hui Schenk

R2 0.29 0.35 0.43 0.40

Q2 0.36 0.41 0.58 0.49

rr 0.47 0.55 0.58 0.59

H2 0.71 0.64 0.64 0.64

F2 0.22 0.24 0.30 0.26

r2 0.63 0.63 0.57 0.60

reY 0.26 0.24 0.19 0.09

reu 0.19 0.17 0.16 0.08


69

Practice of SEM Analysis Theoretical basis Data

Scaling: metric or nominal (in LISREL not standard) Sample-size: a good choice is 10p (p: number of

parameters); <5p cases might result in unstable estimates; large number of cases will result in large values of X2

Reflective indicators are assumed to be uni-dimensional; it is recommended to use principal axis extraction, Cronbach’s alpha and similar to confirm the suitability of data

Model Identification must be checked for covariance fitting

methods Indicators for LV can be formative or reflective; formative

indicators not supported in LISREL


70

Practice of SEM Analalysis cont’d

Model LISREL allows for more general covariance structures

e.g., correlation of measurement errors

Estimation Repeat estimation with varying starting values

Diagnostic checks Use graphical tools like plots of residuals etc. Check each measurement model Check each structural equation Lohmöller’s advice Model trimming Stepwise model building (Hui, 1982; Schenk, 2001)


71

LISREL vs PLS Models

PLS assumes recursive inner structure PLS allows for higher complexity w.r.t. B, , and ; LISREL

w.r.t. and Estimation method

Distributional assumptions in PLS not needed Formative measurement model in PLS Factor scores in PLS PLS: biased estimates, consistency at large LISREL: ML-theory In PLS: diagnostics much richer

Empirical facts LISREL needs in general larger samples LISREL needs more computation


72

The Extended Model

Custo-mer Satis-

faction


PerceivedQuality

Value

EmotionalFactor

EQS-estimates

PLS-estimates


73

Diagnostics: EQS

ACSI ACSI e

247.5 378.7

df 81 173

NNFI 0.898 0.930

RMSEA 0.079 0.060


74

Diagnostics: PLS (centroid weighting)

ACSI ACSI e Hui Schenk

R2 0.29 0.35 0.43 0.40

Q2 0.36 0.41 0.58 0.49

rr 0.47 0.55 0.58 0.59

H2 0.71 0.64 0.64 0.64

F2 0.22 0.24 0.30 0.26

r2 0.63 0.63 0.57 0.60

reY 0.26 0.24 0.19 0.09

reu 0.19 0.17 0.16 0.08


75

Model Building: Hui’s Approach

Custo-mer Satis-

faction

Loyalty

Expec-tations

PerceivedQuality

Value

EmotionalFactor


76

Model Building: Schenk’s Approach

Custo-mer Satis-

faction

Expec-tations

PerceivedQuality

Value

EmotionalFactor


77

The end


78

Data-driven Specification

No solid a priori knowledge about relations among variables

Stepwise regressionSearch of the “best” modelForward selectionBackward eliminationProblem: omitted variable bias

General to specific modeling


79

Stepwise SE Model Building

Hui (1982): models with interdependent inner relations

Schenk (2001): guaranties causal structure, i.e., triangular matrix B of path coefficients in the inner model

η = B η + ζ


80


Hui’s algorithm

Stage 11. Calculate case values Yij for LVs ηi as principal

component of corresponding block, calculate R = Corr(Y) 2. Choose for each endogenous LV the one with highest

correlation to form a simple regression3. Repeat until a stable model is reached

a. PLS-estimate the model, calculate case values, and recalculate R

b. Drop from each equation LVs with t-value |t|<1,65c. Add in each equation the LV with highest partial

correlation with dependent LV


81


Hui’s algorithm, cont’d

Stage 2

1. Use rank condition for checking identifiability of each equation

2. Use 2SLS for estimating the path coefficients in each equation


82

Hui’s vs. Schenk’s Algorithm

Hui’s algorithm is not restricted to a causal structure; allows cycles and an arbitrary structure of matrix B

Schenk’s algorithm uses an iterative procedure similar to that used

by Hui makes use of a priori information about the

structure of the causal chain connecting the latent variables

latent variables are to be sorted


83


Schenk’s algorithm

1. Calculate case values Yij for LVs ηi as principal component of corresponding block, calculate R = Corr(Y)

2. Choose pair of LVs with highest correlation3. Repeat until a stable model is reached

a. PLS-estimate the model, calculate case values, and recalculate R

b. Drop LVs with non-significant t-valuec. Add LV with highest correlation with already included

LVs


84

Data, special CS dimensions

Staff 2 availability1 (PERS), politeness1 (FREU)

Outlet 3 make-up1 (GEST), presentation of mer-chandise1 (PRAE), cleanliness1 (SAUB)

Range 2 freshness and quality (QUAL), richness (VIEL)

Customer-orientation

7 access to outlet (ERRE), shopping hours (OEFF), queuing time for checkout1 (WART), paying modes1 (ZAHL), price information1 (PRAU), sales (SOND), availability of sales (VERF)

1 Dimension of “Emotional Factor”


85

ReferencesC. Fornell (1992), “A National Customer Satisfaction Barometer:

The Swedish Experience”. Journal of Marketing, (56), 6-21.C. Fornell and Jaesung Cha (1994), “Partial Least Squares”, pp.

52-78 in R.P. Bagozzi (ed.), Advanced Methods of Marketing Research. Blackwell.

J.B. Lohmöller (1989), Latent variable path modeling with partial least squares. Physica-Verlag.

H. Wold (1982), “Soft modeling. The basic design and some extensions”, in: Vol.2 of Jöreskog-Wold (eds.), Systems under Indirect Observation. North-Holland.

H. Wold (1985), “Partial Least Squares”, pp. 581-591 in S. Kotz, N.L. Johnson (eds.), Encyclopedia of Statistical Sciences, Vol. 6. Wiley.

Documents

Messung und statistische Analyse von Kundenzufriedenheit KF Qualitätsmanagement Vertiefungskurs V