final exam am(1)

Upload: novelly-sionita-simanjuntak

Post on 03-Apr-2018

214 views

Category:

Documents


0 download

TRANSCRIPT

  • 7/29/2019 Final Exam AM(1)

    1/44

    UNIVERSITY OF INDONESIA

    BLACKBERRY CUSTOMERS SATISFACTION ANALYSIS

    BY USING FACTOR ANALYSIS METHOD

    MULTIVARIATE ANALYSIS REPORT

    ANDRI MUBARAK (0906489555)

    NOVANDRA RHEZZA PRATAMA (0906515793)

    NOVELLY SIONITA SIMANJUNTAK (0906557190)

    RENALDA KRISSALAM (0906636560)

    FACULTY OF ENGINEERING UNIVERSITAS INDONESIA

    DEPARTMENT OF INDUSTRIAL ENGINEERING

    DEPOK

    DECEMBER 2012

  • 7/29/2019 Final Exam AM(1)

    2/44

    ii

    TABLE OF CONTENTS

    TITLE PAGE ........................................................................................................... iTABLE OF CONTENTS ........................................................................................ ii

    LIST OF PICTURES ............................................................................................. iii

    LIST OF TABLES ................................................................................................. iv

    CHAPTER 1 INTRODUCTION ......................................................................... 1

    1.1 Research Background .......................................................................... 11.2 Research Objectives............................................................................. 31.3 Research Information........................................................................... 3

    CHAPTER 2 LITERATURE REVIEW ............................................................. 4

    2.1 Concept of Factor Analysis ................................................................. 42.2 Design Factor Analysis Experiment .................................................... 5

    CHAPTER 3 DATA PROCESSING ................................................................... 7

    3.1 Developing Questionnaires .................................................................. 73.2 Data Collection .................................................................................... 73.3 Steps on Data Tabulation with Factor Analysis .................................. 7

    CHAPTER 4 ANALYSIS ...................................................................................12

    4.1 Factor Analysis .................................................................................. 124.2 Result Data Tabulation and Analysis ................................................ 124.3 Correlation Between Variables .......................................................... 134.4 Factor Analysis Test .......................................................................... 174.5 AntiImage Analysis ....................................................................... 184.6 Process Data after Variable X3 Turn Out From Tabulation Data ..... 224.7 Communalities of Variables .............................................................. 264.8 Determine Number of Factor ............................................................. 274.9 Interpreting Factor ............................................................................. 304.10Component Transformation Matrix ................................................... 35

    CHAPTER 5 CONCLUSION ............................................................................ 36

    REFERENCES .................................................................................................... 37

    THE ATTACHMENTS CHAPTER ................................................................. 38

  • 7/29/2019 Final Exam AM(1)

    3/44

    iii

    LIST OF PICTURES

    Figure 2.1 The Common Factor Model ............................................................... 5

    Figure 3.1 Data Reduction .................................................................................... 9

    Figure 3.2 Insert Variabel .................................................................................... 9

    Figure 3.3 Descriptive Window .......................................................................... 10

    Figure 3.4 Extraction Window ........................................................................... 10

    Figure 3.5 Rotation Window .............................................................................. 11

    Figure 3.6 Score Window ................................................................................... 11

    Figure 4.1 Scree Plot for the Data...................................................................... 29

    Figure 4.2 Component Plot in Rotated Space ................................................... 35

  • 7/29/2019 Final Exam AM(1)

    4/44

    iv

    LIST OF TABLES

    Table 4.1 Correlation Between Variables ......................................................... 14Table 4.2 KMO and Bartletts Test ................................................................... 17

    Table 4.3 Anti-Image Metrics ............................................................................ 19

    Table 4.4 The New KMO and Bartlett's Test ................................................... 23

    Table 4.5 The New Anti-Image Metrics ............................................................ 24

    Table 4.6 Communalities .................................................................................... 26

    Table 4.7 Total Variance Explained .................................................................. 27

    Table 4.8 Initial Eigen Values ............................................................................ 28

    Table 4.9 Extraction Sum of Squared Loadings .............................................. 29

    Table 4.10 Rotation Sum of Squared Loadings ................................................ 30

    Table 4.11 Component Matrix ........................................................................... 31Table 4.12 Rotated Component Matrix ............................................................ 32

    Table 4.13 Classified Factor from Variables .................................................... 33

    Table 4.14 Component Score Coefficient Matrix ............................................. 34

    Table 4.15 Component Transformation Matrix ............................................... 35

  • 7/29/2019 Final Exam AM(1)

    5/44

    1

    Universitas Indonesia

    CHAPTER 1

    INTRODUCTION

    1.1 Research Background

    According to Cronin and Taylor in Baharet al (2009:972), customer

    satisfaction is a major factor in assessing the quality of products and services.

    Customer assesses the performance of the products and services, they directly

    comparing what they received and what they perceived to services and products.

    Quality of services or products is determined by the level of compatibility

    between the services given to those expected by the consumer. The higher the

    received service equality will be higher levels of customer satisfaction. Marketing

    managers recognize that customer satisfaction proved to be very useful in the

    business so they often held management survey on customer satisfaction with the

    product produced by a company.

    Customer satisfaction can be used to measure the market a product or

    service so that company can further expand or repair their existing production.

    company use customer satisfaction as input for subsequent products and services,

    because if a customer satisfied with the products or services it wishes to purchase

    very high, after the purchase, the customer would recommend the product or

    service to other people starting families and the closest so they will be motivated

    to buy products or services.

    After purchasing the products or services and the customer is satisfied

    its possible to buy the products or services back the next time and it could be a

    repeat for several times if the level of satisfaction is high. If customers are not

    satisfied then it is not going to happen so there should be measurement ofcustomer satisfaction possible by asking their customers and if customers are not

    satisfied, the customer will provide a critical commentary.

    Many companies believe that customer satisfaction is important. As

    reflected from the vision, mission and business plan that they formulated.

    Measuring customer satisfaction is a concrete manifestation of the company's

    commitment to customer satisfaction. Based on the experience of doing research

    for many companies in Indonesia, there are 5 things to optimize the benefits of the

  • 7/29/2019 Final Exam AM(1)

    6/44

    2

    Universitas Indonesia

    research. The first step is to involve top management. Prior research conducted,

    very important or even should, to seek approval and support from top

    management. If top management is still in doubt, it is better to delay the conduct

    customer satisfaction research. The second step is to involve the relevant

    department. Measuring customer satisfaction often involves many departments

    such as marketing department, service, human resources, production, business

    development and even financial. In the manufacturing industry, R&D department

    is also important to be involved. The higher the company's orientation towards the

    customers, the more departments should be involved. Customer satisfaction is the

    responsibility of all departments and not just the departments that deal with

    customers on a daily basis.

    Third, research should have high credibility. Therefore, the measurement

    of customer satisfaction research must be carried out to the correct method. The

    company can actually do the research yourself if measurements have sufficient

    resources, i.e. at least some researchers who controlled research methodology,

    statistics and basic concepts of customer satisfaction. When internally is not

    possible, it is better to outsource the research firm that has credibility. Without

    credibility, let alone take advantage of the results, to simply listen to the

    presentation of the results of his research-also, they will be reluctant. Fourth, the

    results of this survey should be presented and disseminated. It is important; the

    results of this research are communicated to the level of front-line staff.

    Fifth, the preparation of a new program based on the results of this

    research should be done as quickly as possible. At the latest, one month after the

    research is completed; the company should be ready with the customer

    satisfaction program after learning how effective customer satisfaction programlength based on research. Making this program should be initiated with the

    objective to be achieved such as how much the rate of increase of customer

    satisfaction will be achieved. Sixth, scheduling customer satisfaction measure-

    ment. Time, depending on the speed of change in the business environment and

    how much of the planned program has been implemented. In general, the

    measurement of large-scale customer satisfaction can be done once or twice a

    year. Seventh, to use the results of customer satisfaction measurement is to

  • 7/29/2019 Final Exam AM(1)

    7/44

    3

    Universitas Indonesia

    connect the measurement of customer satisfaction and the reward of the

    employee.

    In this case, we use Blackberry as our product, we choose this product

    because its widely used in Indonesia so that our customer is sufficient for the

    survey, and we especially survey Industrial Engineering students. This research is

    useful to measure Blackberry next product market especially among industrial

    engineering students.

    1.2 Research Objectives

    The research is done to realize the objectives mentioned below:

    Determine the factors or variables that influence customer satisfactions orcustomer preference in Blackberry.

    Find the most significant factors that influence the customer preferencefor Blackberry.

    Determine the level of customer satisfaction especially Blackberry usersin Industrial Engineering.

    1.3 Research Information

    The detail informations about this research is mentioned below:

    Title : Blackberrys customer satisfaction using factor analysismethod.

    Methods : Factor Analysis. Samples : Blackberry users especially Industrial Engineering students

    University of Indonesia.

    Time : Tuesday, 4th December 2012 Tuesday, 18th December2012.

    Place : Mailing list Industrial Engineering 2009 (online),Department of Industrial Engineering, University of

    Indonesia.

  • 7/29/2019 Final Exam AM(1)

    8/44

    4

    Universitas Indonesia

    CHAPTER 2

    LITERATURE REVIEW

    2.1 Concept of Factor Analysis

    Factor analysis is a statistical method used to study the dimensionality of

    a set of variables. In factor analysis, latent variables represent unobserved

    constructs and are referred to as factors or dimensions. Factor analysis searches

    for such joint variations in response to unobserved latent variables. The observed

    variables are modeled as linear combinations of the potential factors, plus "error"

    terms. The information gained about the interdependencies between observed

    variables can be used later to reduce the set of variables in a dataset.

    Computationally this technique is equivalent to low rank approximation of the

    matrix of observed variables. Factor analysis originated in psychometrics, and is

    used in behavioral sciences, social sciences, marketing, product management,

    operations research, and other applied sciences that deal with large quantities of

    data.

    There are basically two types of factor analysis: exploratory and

    confirmatory.

    Exploratory Factor Analysis (EFA)Used to explore the dimensionality of a measurement instrument by

    finding the smallest number of interpretable factors needed to explain the

    correlations among a set of variables exploratory in the sense that it places no

    structure on the linear relationships between the observed variables and on the

    linear relationships between the observed variables and the factors but only

    specifies the number of latent variables. Confirmatory Factor Analysis (CFA)

    Used to study how well a hypothesized factor model fits a new sample

    from the same population or a sample from a different populationcharacterized

    by allowing restrictions on the parameters of the model.

  • 7/29/2019 Final Exam AM(1)

    9/44

    5

    Universitas Indonesia

    Figure 2.1 The Common Factor Model

    Both types of factor analyses are based on the Common Factor Model,

    illustrated in figure 1.1. This model proposes that each observed response

    (measure 1 through measure 5) is influenced partially by underlying common

    factors (factor 1 and factor 2) and partially by underlying unique factors (E1

    through E5). The strength of the link between each factor and each measure

    varies, such that a given factor influences some measures more than others.

    2.2 Design Factor Analysis Experiment

    In designing experiment which uses factor analysis, we need several

    steps. Principles of experimental design, following Ronald A. Fisher. Generally

    there are many steps to conduct factor analysis experiment. There are steps will be

    explained below:

    ComparisonIn some fields of study it is not possible to have independent

    measurements to a traceable standard. Comparisons between treatments are much

    more valuable and are usually preferable. Often one compares against a scientific

    control or traditional treatment that acts as baseline.

    RandomizationRandom assignment is the process of assigning individuals at random to

    groups or to different groups in an experiment. The random assignment of

  • 7/29/2019 Final Exam AM(1)

    10/44

    6

    Universitas Indonesia

    individuals to groups (or conditions within a group) distinguishes a rigorous,

    "true" experiment from an adequate, but less-than-rigorous, "quasi-experiment".

    ReplicationMeasurements are usually subject to variation and uncertainty.

    Measurements are repeated and full experiments are replicated to help identify the

    sources of variation, to better estimate the true effects of treatments, to further

    strengthen the experiment's reliability and validity, and to add to the existing

    knowledge of the topic.

    BlockingBlocking is the arrangement of experimental units into groups (blocks)

    consisting of units that are similar to one another. Blocking reduces known but

    irrelevant sources of variation between units and thus allows greater precision in

    the estimation of the source of variation under study.

    OrthogonalityOrthogonality concerns the forms of comparison (contrasts) that can be

    legitimately and efficiently carried out. Contrasts can be represented by vectors

    and sets of orthogonal contrasts are uncorrelated and independently distributed if

    the data are normal.

    Factorial experimentsUse of factorial experiments instead of the one-factor-at-a-time method.

    These are efficient at evaluating the effects and possible interactions of several

    factors (independent variables).

  • 7/29/2019 Final Exam AM(1)

    11/44

    7

    Universitas Indonesia

    CHAPTER 3

    DATA PROCESSING

    3.1 Developing Questionnaires

    To get customer satisfaction data to the Blackberry, we conduct research

    with the use of a detailed questionnaire. In developing the questionnaires, weshould decide the characteristics and the variables needed to make good

    questionnaires. A good questionnaire should be valid and reliable. Valid means

    the study is able to scientifically answer the questions it is intended to answer.

    Reliable means the study is consistent and the study may not measure what it

    wants to be measuring. Both validity and reliability should be fulfilled so that we

    could analyze the real situation accurately.

    3.2 Data Collection

    We gathered data by distributing questionnaire to Blackberry users. From

    the gathered data, we already collect 50 (fifty) data that can represent the

    Blackberry users. The distributed questionnaires and the data result will be

    attached in The Attachments Chapter.

    3.3 Step on Data Tabulation with Factor Analysis

    We use SPSS 17.0 software for tabulated data in factor analysis.

    Furthermore, step for doing factor analysis using SPSS 17.0 are shown below:

    1. Open SPSS 17.02. Create new data.

    a.

    Click on file toolbar.b. Choose new and the data.c. Insert the data in data sheet.

    (by typing data on the data sheet or import data from Ms.Excel)

  • 7/29/2019 Final Exam AM(1)

    12/44

    8

    Universitas Indonesia

    1. If data already exist, open the data file.(click Analyze toolbar, choose data reduction and then factor)

  • 7/29/2019 Final Exam AM(1)

    13/44

    9

    Universitas Indonesia

    Figure 3.1 Data Reduction

    2. Select the variables you want the factor analysis to be based on and movethem into the variable(s) box.

    Figure 3.2 Insert Variabel

  • 7/29/2019 Final Exam AM(1)

    14/44

    10

    Universitas Indonesia

    3. Click descriptive button.a. On statistic cluster, check initial solution.b. On correlation matrix cluster, check coefficients, significance levels,

    determinant, KMO and Bartletts test of sphericity and anti-image.

    c. Click continue button.

    Figure 3.3 Descriptive Window

    4. Choose extraction button.a. On display cluster, check unrotated factor solution, scree plot.b. Then continue.

    Figure 3.4 Extraction Window

    5. Click rotation button.

  • 7/29/2019 Final Exam AM(1)

    15/44

    11

    Universitas Indonesia

    a. On method cluster, choose varimax.b. Click continue.

    Figure 3.5 Rotation Window

    6. Choose scores button.a. Check save as variable.b. On method cluster, choose Bartlett.c. Check display factor score coefficient matrix.d. Then continue.

    Figure 3.6 Score Window

    7. Click OK button

  • 7/29/2019 Final Exam AM(1)

    16/44

    12

    Universitas Indonesia

    CHAPTER 4

    ANALYSIS

    4.1 Factor Analysis

    Factor analysis is a multivariate method used for data reduction purposes.

    The basic idea is to represent a set of variables by a smaller number of variables.

    In this case they are called factor.

    There are two kinds of factor analysis that often used, which are:

    a. Data reduction analysis. Use to remove redundant (highly correlatd)variables from the data file, perhaps replacing the entire data file with a

    smaller number of uncorrelated variables.

    b. Structure detection analysis. Use to examine the underlying (or latent)relationships between the variables.

    We would use data reduction analysis as the procedure in accomplishing

    this research, considering the objective of this factor analysis is to reduce 10

    attributes into fewer numbers, which are the linear combination of those attributes.

    We also used structure detection in order to see the correlation between variables.

    4.2 Result Data Tabulation and Analysis

    Determinant using to show something or in this case factor that establish.

    KMO Bartlett used to determine if data was suitable or not. In the Descriptives

    window above, we select KMO and Bartletts test of sphericity. KMO is a statistic

    which tells whether you have sufficient items for each factor. Barletts test test is

    used to check that the original variables are sufficiently correlated. Whereas, anti

    image can be using to seeing when factor model are good or not from the valuenumber of element.

    On extraction, correlation matrix and eigenvalue over 1 are checked, it

    use to be based on extraction. Raw data are reducted into factor that amount

    appropriate with number of data with eigenvalue more than 1. Unrotated factor

    solution displayed extraction on side of communalities. It indicates how well a

    variable representing on the factor. Scree plot used to display graph of eigenvalue

    from each component in the initial solution. The scree plot also helps you to

  • 7/29/2019 Final Exam AM(1)

    17/44

    13

    Universitas Indonesia

    determine the optimal number of components. Data then classified into that factor

    based on correlation of each variable to factor created. Usually, from the

    correlation matrix, there is a data which hard to classified because the correlation

    to factor almost same. While, correlation matrix used because variables on our

    analysis are measured on different scales.

    To solve this, we can used rotation. Rotation method that used in this

    data tabulation are Variamax. Variamax simplifying the columns of the factor

    matrix so there tend to be some high loading (i.e, close to -1 or +1 ) and some

    laodings near to 0 in each column of matrix. The reason to choose this method are

    Variamax The reason to choose this method are Variamax has a simple structure

    (near -1 or +1 high correlation and near 0 indicating lack of correlation).

    Compared with Quartimax which has simpler analytical. Varimax give a clearer

    separation of the factor. When different subsets of variables are analyzed, factor

    pattern that obtained from Varimax tend to be more invariant than that obtained

    by Quartimax method.

    Furthemore, data that has been tabulated saved as Bartlett. This factor

    will be use again on the develop linear regression as an independent variable.

    4.3 Correlation Between Variables

    The underlying statisticlal assumptions impact factor analysis to the

    extent that they affect the derived correlation. Departures from normality,

    homoscedasticity and linearity can diminish correlations between variables.

    Examining the correlations among the survey items reveals that there is significant

    overlap among various subgroups of items.

  • 7/29/2019 Final Exam AM(1)

    18/44

    14

    Universitas Indonesia

    Table 4.1 Correlation Between Variables

    Correlation Matrix

    X1 -

    kepuasananda

    terhadapBB

    X2 - kecepatanoperasi/penggunaan

    program BB yangbaik

    X3 -kehandalanBB, dalam

    menangani

    program dan

    gangguaneksternal

    sepertiterjatuh

    X4 -kesesuaian

    service

    (program

    yang

    dimasukkan)dengan

    kebutuhananda

    X5 -

    kesesuaianharga

    denganproduk

    X6 -service

    pascapenjualan

    X7 -

    kemudahandalam

    menggunakanuntuk pemula

    X8 -kemudahan

    dalammaintenance

    X9 -

    promosiproduk

    yangefektif

    X10 -design

    yangmenarik

    Correlation X1 - kepuasan anda

    terhadap BB

    1.000 .678 .049 .157 .087 .399 .534 .507 .177 -.002

    X2 - kecepatanoperasi/penggunaan

    program BB yang

    baik

    .678 1.000 .153 .167 .176 .525 .319 .343 .165 -.063

    X3 - kehandalan BB,

    dalam menanganiprogram dan

    gangguan eksternalseperti terjatuh

    .049 .153 1.000 .321 .097 .120 -.020 .054 .091 .394

    X4 - kesesuaian

    service (programyang dimasukkan)dengan kebutuhan

    anda

    .157 .167 .321 1.000 .514 .136 .169 .115 .443 .432

    X5 - kesesuaian

    harga dengan produk

    .087 .176 .097 .514 1.000 .349 .128 .301 .408 .304

  • 7/29/2019 Final Exam AM(1)

    19/44

    15

    Universitas Indonesia

    X6 - service pasca

    penjualan

    .399 .525 .120 .136 .349 1.000 .359 .569 .189 .136

    X7 - kemudahan

    dalam menggunakanuntuk pemula

    .534 .319 -.020 .169 .128 .359 1.000 .554 -.038 .121

    X8 - kemudahandalam maintenance

    .507 .343 .054 .115 .301 .569 .554 1.000 .397 .261

    X9 - promosi produk

    yang efektif

    .177 .165 .091 .443 .408 .189 -.038 .397 1.000 .460

    X10 - design yang

    menarik

    -.002 -.063 .394 .432 .304 .136 .121 .261 .460 1.000

    Sig. (1-

    tailed)

    X1 - kepuasan anda

    terhadap BB

    .000 .368 .137 .274 .002 .000 .000 .110 .493

    X2 - kecepatan

    operasi/penggunaan

    program BB yangbaik

    .000 .144 .123 .111 .000 .012 .007 .126 .333

    X3 - kehandalan BB,

    dalam menangani

    program dangangguan eksternalseperti terjatuh

    .368 .144 .012 .252 .202 .444 .356 .264 .002

    X4 - kesesuaian

    service (program

    yang dimasukkan)

    dengan kebutuhananda

    .137 .123 .012 .000 .173 .120 .214 .001 .001

    X5 - kesesuaian

    harga dengan produk

    .274 .111 .252 .000 .006 .188 .017 .002 .016

  • 7/29/2019 Final Exam AM(1)

    20/44

    16

    Universitas Indonesia

    X6 - service pasca

    penjualan

    .002 .000 .202 .173 .006 .005 .000 .094 .174

    X7 - kemudahandalam menggunakanuntuk pemula

    .000 .012 .444 .120 .188 .005 .000 .398 .201

    X8 - kemudahandalam maintenance

    .000 .007 .356 .214 .017 .000 .000 .002 .034

    X9 - promosi produkyang efektif

    .110 .126 .264 .001 .002 .094 .398 .002 .000

    X10 - design yang

    menarik

    .493 .333 .002 .001 .016 .174 .201 .034 .000

  • 7/29/2019 Final Exam AM(1)

    21/44

    17

    Universitas Indonesia

    Table above shows the correlation between the factors. It shows the

    relative value between the variable. As we can see, we cannot conclude anything

    since too much variability within the numbers above. Thats why continue our

    analysis to the next figure.

    4.4 Factor Analysis Test

    After examine the correlation between variable, we assess the overall

    significance of the correlation matrix using KMO ( Kaiser-Meyer-Olkin) and

    Bartletts Test of Sphericity. This test is also effective to check whether factor

    analysis is suitable to our data or not. Factor analysis is suitable if there is

    correlation between variable that being measured. If correlation between all

    variables is low, factor analysis can not be used.

    The Kaiser - MeyerOlkin measure of sampling adequacy tests whether

    the partial correlations among variables are small. Bartletts test of sphericity tests

    whether the correlation matrix is an identity matrix, which would indicate that the

    factor model is inappropriate. Use of these method is related to structure detection

    analysis.

    The result of KMO (Kaiser Meyer Olkin) and Bartletts Test of

    Sphericity is shown in table below.

    Table 4.2 KMO and Bartletts Test

    KMO and Bartlett's Test

    Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .628

    Bartlett's Test of Sphericity Approx. Chi-Square 177.936

    df 45

    Sig. .000

    Value range of KMO test is 0-1. High values (close to 1.0) generally

    indicate that a factor analysis may be useful with your data. If the value is less

    than 0.50, the results of the factor analysis probably wont be very useful. In table

  • 7/29/2019 Final Exam AM(1)

    22/44

    18

    Universitas Indonesia

    above, we get KMO 0.628. It means that factor analysis is very useful to our

    variables.

    For Bartlett Test, small values (less than 0.05) of the significance level

    indicate that a factor a factor analysis may be useful with our data. In table above,

    we get significance level 0.000 which is less than 0.05, so we may conclude that

    factor analysis may be useful with our data.

    4.5 AntiImage Analysis

    The anti image correlation matrix contains the negatives of the partial

    correlation coefficient, and the image covariance matrix contains the negatives of

    the partial covariance. In a good factor model, most of the off- diagonal elements

    will be small. The measure of sampling adequacy for a variable is displayed on

    the diagonal of the anti-image correlation matrix. If true factors exist in the data,

    the partial correlation should be small, because the variables can be explained by

    the factors (variates with loadings for each variable). If partial correlation are

    high, then there are no true underlying factors, and factor analysis is

    inappropriate.

    From the result of data tabulation we can look at diagonal from left-up to

    right down the data on anti-image correlation matrix which also measures of

    sampling adequacy. If the value less than 0.5 then the data must be trun out from

    the data tabulation and then repeat the process. As seen in the table below,

    Variable X3 - kehandalan BB, dalam menangani program dan gangguan

    eksternal seperti terjatuhis less than 0.5, so the data must be turn out from

    the data tabulation and we must repeat the process.

  • 7/29/2019 Final Exam AM(1)

    23/44

    19

    Universitas Indonesia

    Table 4.3 Anti-Image Metrics

    Anti-image Matrices

    X1 -

    kepuasananda

    terhadapBB

    X2 -

    kecepatan

    operasi/penggunaan

    program BByang baik

    X3 -kehandalanBB, dalam

    menangani

    program dan

    gangguaneksternal

    sepertiterjatuh

    X4 -

    kesesuaian

    service

    (program yangdimasukkan)

    dengankebutuhan anda

    X5 -

    kesesuaianharga

    denganproduk

    X6 -service

    pascapenjualan

    X7 -

    kemudahandalam

    menggunakanuntuk pemula

    X8 -kemudahan

    dalammaintenance

    X9 -

    promosiproduk

    yangefektif

    X10 -design

    yangmenarik

    Anti-image

    Covariance

    X1 - kepuasan anda

    terhadap BB

    .381 -.227 .018 -.038 .077 .048 -.123 -.087 -.031 .036

    X2 - kecepatanoperasi/penggunaan

    program BB yang baik

    -.227 .407 -.106 -.002 -.028 -.183 .002 .065 -.054 .105

    X3 - kehandalan BB,

    dalam menangani

    program dan gangguaneksternal seperti

    terjatuh

    .018 -.106 .710 -.159 .081 -.013 .118 -.044 .138 -.246

    X4 - kesesuaian

    service (program yang

    dimasukkan) dengankebutuhan anda

    -.038 -.002 -.159 .497 -.232 .023 -.133 .135 -.163 -.073

    X5 - kesesuaian harga

    dengan produk

    .077 -.028 .081 -.232 .595 -.120 .022 -.062 -.049 -.017

    X6 - service pasca

    penjualan

    .048 -.183 -.013 .023 -.120 .504 .002 -.165 .063 -.032

  • 7/29/2019 Final Exam AM(1)

    24/44

    20

    Universitas Indonesia

    X7 - kemudahan

    dalam menggunakan

    untuk pemula

    -.123 .002 .118 -.133 .022 .002 .458 -.187 .214 -.088

    X8 - kemudahandalam maintenance

    -.087 .065 -.044 .135 -.062 -.165 -.187 .357 -.180 -.021

    X9 - promosi produkyang efektif

    -.031 -.054 .138 -.163 -.049 .063 .214 -.180 .461 -.177

    X10 - design yang

    menarik

    .036 .105 -.246 -.073 -.017 -.032 -.088 -.021 -.177 .551

    Anti-imageCorrelation

    X1 - kepuasan andaterhadap BB

    .697a

    -.575 .034 -.088 .163 .109 -.295 -.236 -.074 .079

    X2 - kecepatan

    operasi/penggunaan

    program BB yang baik

    -.575 .629a -.197 -.006 -.056 -.403 .004 .171 -.125 .223

    X3 - kehandalan BB,dalam menangani

    program dan gangguaneksternal seperti

    terjatuh

    .034 -.197 .450a

    -.268 .125 -.021 .207 -.087 .241 -.394

    X4 - kesesuaian

    service (program yangdimasukkan) dengan

    kebutuhan anda

    -.088 -.006 -.268 .598a

    -.427 .045 -.279 .321 -.340 -.140

    X5 - kesesuaian hargadengan produk

    .163 -.056 .125 -.427 .724a -.220 .043 -.135 -.093 -.030

    X6 - service pascapenjualan

    .109 -.403 -.021 .045 -.220 .733a

    .003 -.390 .131 -.060

    X7 - kemudahan

    dalam menggunakanuntuk pemula

    -.295 .004 .207 -.279 .043 .003 .569a -.463 .465 -.176

  • 7/29/2019 Final Exam AM(1)

    25/44

    21

    Universitas Indonesia

    X8 - kemudahan

    dalam maintenance

    -.236 .171 -.087 .321 -.135 -.390 -.463 .632a

    -.444 -.048

    X9 - promosi produkyang efektif

    -.074 -.125 .241 -.340 -.093 .131 .465 -.444 .525a

    -.351

    X10 - design yang

    menarik

    .079 .223 -.394 -.140 -.030 -.060 -.176 -.048 -.351 .658a

    a. Measures of Sampling Adequacy(MSA)

  • 7/29/2019 Final Exam AM(1)

    26/44

    22

    Universitas Indonesia

    4.6 Process Data After Variable X3 Turn Out From Tabulation Data

    The result of KMO (Kaiser - Meyer Olkin) and Bartletts Test of

    Sphericity after we removed Variable X3 is shown in table below, This is the

    processing of the data.

  • 7/29/2019 Final Exam AM(1)

    27/44

    23

    Universitas Indonesia

    Table 4.4 The New KMO and Bartlett's Test

    Table The New KMO and Bartlett's Test

    Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .657

    Bartlett's Test of Sphericity Approx. Chi-Square 163.760

    Df 36

    Sig. .000

    From the result of data tabulation, we can see the new result of KMO and

    Bartletts Test. As seen in the table below, all of them more than 0.5, so all of

    data can be included on data tabulation and classified into factors on

    extraction.

  • 7/29/2019 Final Exam AM(1)

    28/44

    24

    Universitas Indonesia

    Table 4.5 The New Anti-Image Metrics

    Anti-image Matrices

    X1 -kepuasan

    anda

    terhadap

    BB

    X2 -kecepatanoperasi/

    |penggunaan

    program BB

    yang baik

    X4 - kesesuaianservice (program

    yang dimasukkan)

    dengan kebutuhan

    anda

    X5 -kesesuaian

    harga

    dengan

    produk

    X6 -

    service

    pasca

    penjualan

    X7 - kemudahan

    dalam

    menggunakan

    untuk pemula

    X8 -

    kemudahan

    dalam

    maintenance

    X9 -promosi

    produk

    yang

    efektif

    X10 -

    design

    yang

    menarik

    Anti-imageCovariance

    X1 - kepuasan andaterhadap BB

    .382 -.233 -.037 .077 .048 -.132 -.087 -.037 .050

    X2 - kecepatan

    operasi/penggunaanprogram BB yang baik

    -.233 .423 -.029 -.016 -.192 .021 .061 -.037 .085

    X4 - kesesuaian service(program yang

    dimasukkan) dengankebutuhan anda

    -.037 -.029 .535 -.234 .021 -.120 .136 -.151 -.163

    X5 - kesesuaian harga

    dengan produk

    .077 -.016 -.234 .604 -.121 .009 -.059 -.069 .013

    X6 - service pascapenjualan

    .048 -.192 .021 -.121 .504 .004 -.168 .070 -.043

    X7 - kemudahan dalam

    menggunakan untukpemula

    -.132 .021 -.120 .009 .004 .479 -.190 .212 -.058

    X8 - kemudahan dalammaintenance

    -.087 .061 .136 -.059 -.168 -.190 .360 -.183 -.043

    X9 - promosi produkyang efektif

    -.037 -.037 -.151 -.069 .070 .212 -.183 .489 -.162

  • 7/29/2019 Final Exam AM(1)

    29/44

    25

    Universitas Indonesia

    X10 - design yang

    menarik

    .050 .085 -.163 .013 -.043 -.058 -.043 -.162 .652

    Anti-image

    Correlation

    X1 - kepuasan anda

    terhadap BB

    .691a

    -.580 -.082 .160 .110 -.309 -.234 -.085 .101

    X2 - kecepatan

    operasi/penggunaan

    program BB yang baik

    -.580 .646a -.062 -.032 -.415 .046 .157 -.081 .161

    X4 - kesesuaian service

    (program yangdimasukkan) dengankebutuhan anda

    -.082 -.062 .605a -.412 .041 -.237 .310 -.295 -.277

    X5 - kesesuaian hargadengan produk

    .160 -.032 -.412 .741a

    -.219 .017 -.126 -.127 .021

    X6 - service pascapenjualan

    .110 -.415 .041 -.219 .723a

    .008 -.394 .140 -.074

    X7 - kemudahan dalammenggunakan untuk

    pemula

    -.309 .046 -.237 .017 .008 .610a -.457 .437 -.105

    X8 - kemudahan dalam

    maintenance

    -.234 .157 .310 -.126 -.394 -.457 .639a -.437 -.090

    X9 - promosi produkyang efektif

    -.085 -.081 -.295 -.127 .140 .437 -.437 .579a

    -.287

    X10 - design yang

    menarik

    .101 .161 -.277 .021 -.074 -.105 -.090 -.287 .731a

    a. Measures of Sampling Adequacy(MSA)

  • 7/29/2019 Final Exam AM(1)

    30/44

    26

    Universitas Indonesia

    4.7 Communalities of Variables

    Communality shows the amount of variance in a variable that is

    accounted for by the two factors taken together. The size of the communality is a

    useful index for assessing how much variance variable is accounted for by the

    factor solution. Small communalities show that a substantial portion of the

    variance in a variable is unaccounted for by the factors. Large communalities

    indicate that a large amount of variance in a variable has been extracted by the

    factor solution.

    Communality table of variables is shown in table below:

    Table 4.6 Communalities

    Communalities

    Initial Extraction

    X1 - kepuasan anda terhadap BB 1.000 .711

    X2 - kecepatan operasi/penggunaan program BB yang baik 1.000 .601

    X4 - kesesuaian service (program yang dimasukkan) dengan

    kebutuhan anda

    1.000 .580

    X5 - kesesuaian harga dengan produk 1.000 .540

    X6 - service pasca penjualan 1.000 .552

    X7 - kemudahan dalam menggunakan untuk pemula 1.000 .511

    X8 - kemudahan dalam maintenance 1.000 .624

    X9 - promosi produk yang efektif 1.000 .601

    X10 - design yang menarik 1.000 .562

    Extraction Method: Principal Component Analysis.

    Table Communalities

    The first column in table above, which is labeled as initial

    communalities, are estimates of the variance in each variable accounted for by all

    components or factors. For principal components extraction, this is always equal

    to 1.0 for correlation analyses.

    The second column is labeled as extraction. This column show the

    estimation of the variance in each accounted for by the components. The

  • 7/29/2019 Final Exam AM(1)

    31/44

    27

    Universitas Indonesia

    communalities in this table are all high, which indicates that the extracted

    components represent the variables well. If any communalities are very low in a

    principal components extraction, you may need to extract another component. For

    example, lets take a look in variable Kepuasaan Anda terhadap BB. This

    variable has value 0.711. It means that 71.1 % variance of variable. Kepuasaan

    Anda terhadap BB can be determine by the result factor.

    4.8 Determine Number of Factor

    In order to determine the correct number of factor, there are 4 methods

    that can be used, which are latent root criterion, priori criterion, percentage of

    variance, and screen plot criterion. In this step, we focus in latent root method.

    Latent root criterion is the most commonly used technique. This

    method extracts eigen value to determine how many factors should be included in

    the research. Eigen value is value that represents the amount of variance

    accounted for by factor. The rationale of Latent Root Criterion is that any

    individual factor should account for the variance of at least a single variable if it is

    to be retained for interpretation. After the calculation has been done, eigen value

    will continue to decrease, as shown in the table below.

    Table 4.7 Total Variance Explained

    Total Variance Explained

    Component

    Initial Eigenvalues

    Extraction Sums of Squared

    Loadings

    Rotation Sums of Squared

    Loadings

    Total

    % of

    Variance

    Cumulative

    % Total

    % of

    Variance

    Cumulative

    % Total

    % of

    Variance

    Cumulative

    %

    1 3.391 37.679 37.679 3.391 37.679 37.679 2.909 32.324 32.324

    2 1.892 21.018 58.697 1.892 21.018 58.697 2.374 26.373 58.697

    3 .905 10.057 68.754

    4 .786 8.732 77.486

    5 .743 8.254 85.740

    6 .523 5.807 91.547

    7 .336 3.730 95.277

    8 .232 2.581 97.858

    9 .193 2.142 100.000

    Extraction Method: Principal Component Analysis.

  • 7/29/2019 Final Exam AM(1)

    32/44

    28

    Universitas Indonesia

    Table 4.8 Initial Eigen Values

    Component

    Initial Eigenvalues

    Total % of Variance Cumulative %

    1 3.391 37.679 37.679

    2 1.892 21.018 58.697

    3 .905 10.057 68.754

    4 .786 8.732 77.486

    5 .743 8.254 85.740

    6 .523 5.807 91.547

    7 .336 3.730 95.277

    8 .232 2.581 97.858

    9 .193 2.142 100.000

    The Total column gives the eigen value, or amount of variance in the

    original variables accounted for by each component. The % of Variance column

    gives the ratio, expressed as a percentage, of the variance accounted for by each

    component to the total variance in all of the variables. The Cumulative %

    column gives the percentage of variance accounted for by the first n components.

    For example, the cumulative percentage for the second components is the sum of

    the percentage of variance for the first and second components.

    For the initial solution, there are as many components as variables, and in

    a correlation analysis, the sum of the eigen values equals the number of

    components. We requested that eigen values greater than 1 be extracted in SPSS

    step. It means that only the components having eigen values greater than 1 are

    considered significant; all components with eigen values less than 1 are

    considered insignificant and are disagreed. In table above, only two components

    that have eigen values greater than 1. They are first two components with eigen

    values 3.391 and 1.892 respectively. So we may conclude that there are 2 factors

    that will be retained.

    Scree plot will show the calculation graphically. The scree plot for the

    data is shown below:

  • 7/29/2019 Final Exam AM(1)

    33/44

    29

    Universitas Indonesia

    Figure 4.1 Scree Plot for the Data

    From scree plot we can obviously see that there are 2 components

    retained. The second section of the table above show the extracted components.

    Table 4.9 Extraction Sum of Squared Loadings

    Extraction Sums of Squared Loadings

    Total % of Variance Cumulative %

    3.391 37.679 37.679

    1.892 21.018 58.697

    The % of Variance column gives the ratio, expressed as a percentage;

    between the eigenvalues with overall eigenvalues (the value of overall

    eigenvalues is equal to number number of variables, 9). Factor 1 (component 1)

  • 7/29/2019 Final Exam AM(1)

    34/44

    30

    Universitas Indonesia

    has 37.679% ratio, and factor 2 has 21.018% ratio. The Cumulative % column

    gives the cumulative of % variance. The last row of this column contains value of

    58.697%. It means 58.697% variability of level of customer satisfaction in the

    original 9 variables can be described b this factor analysis. So we can

    considerably reduce the complexity of the data set by using these components,

    with 41.303% loss of information.

    The third section of the table above contains rotation sum of squared

    loading. The rotation maintains the cumulative percentage of variation explained

    by the extracted components, but that variation is now spread more evenly over

    the components. The small changes in the individual totals suggest that the rotated

    component matrix will be harder to interpret than the unrotated matrix. The final

    Cumulative % is not different.

    Table 4.10 Rotation Sum of Squared Loadings

    Rotation Sums of Squared Loadings

    Total % of Variance

    2.909 32.324

    2.374 26.373

    4.9 Interpreting Factor

    After decide number of factor retained, the next step is classify each

    variables into new factor. First of all, we computed the initial unrotated factor

    matrix to assist in obtaining a preliminary indication of the number of factor to

    extract. The factor matrix contain factor loading for each variable on each factor.Factor loading is correlation between the original variables and the factor and the

    key to understanding the nature of particular factor. The high loading factor means

    the variables are representative of the factor or have strong correlation. The

    unrotated factor matrix is shown below:

  • 7/29/2019 Final Exam AM(1)

    35/44

    31

    Universitas Indonesia

    Table 4.11 Component Matrix

    Component Matrixa

    Component

    1 2

    X1 - kepuasan anda terhadap BB .695 -.478

    X2 - kecepatan operasi/penggunaan program BB yang baik .648 -.426

    X4 - kesesuaian service (program yang dimasukkan) dengan

    kebutuhan anda

    .504 .571

    X5 - kesesuaian harga dengan produk .558 .478

    X6 - service pasca penjualan .709 -.222

    X7 - kemudahan dalam menggunakan untuk pemula .606 -.380

    X8 - kemudahan dalam maintenance .776 -.146

    X9 - promosi produk yang efektif .538 .559

    X10 - design yang menarik .404 .632

    Extraction Method: Principal Component Analysis.

    a. 2 components extracted.

    X1 factor loading for component (Factor) 1 has 0.695 values. It is

    obvious that this value is higher than factor loading for other component. So, wecan easily put X1 Variable into Factor 1. However, factor loading of X4 for factor

    1 (0.504) has similar factor loading for factor 2 (0.571). It is difficult to decide

    whether X4 is in factor 1 or 2.

    Based on facts above, we can conclude that unrotated factor solution may

    not provide a meaningful pattern of variable loading. If the unrotated factors are

    expected to be meaningful, we have to rotate the data.

    Rotation

    Rotation method can increase interpretation level by decreasing the

    occurrence of ambiguity that often happen on first extraction. Thats why in this

    factor analysis we also use rotation method. We conduct rotation for every score

    that doesnt show any significance between them. The term rotation means

    exactly what it implies. Specifically the reference axes of the factor are turned

    about the origin until some other position has been reached.

  • 7/29/2019 Final Exam AM(1)

    36/44

    32

    Universitas Indonesia

    The simplest case of rotation is an orthogonal rotation, in which the axes

    are maintained at 90 degrees. There are 3 methods of orthogonal rotation,

    Varimax, Quartimax, and Equimax. In this case, we chose Varimax. While

    quartimax rotation is to simplify the rows of factor matrix, the varimax criterion

    centers on simplifying the column of the factor matrix. With varimax rotational

    approach, the maximum possible simplification is reached if there are only 1s and

    0s in a column. So, with varimax, we will get clear variable classification.

    Table 4.12 Rotated Component Matrix

    Rotated Component Matrixa

    Component

    1 2

    X1 - kepuasan anda terhadap BB .843 .000

    X2 - kecepatan operasi/penggunaan program BB yang baik .775 .016

    X4 - kesesuaian service (program yang dimasukkan) dengan

    kebutuhan anda

    .091 .756

    X5 - kesesuaian harga dengan produk .188 .710

    X6 - service pasca penjualan .710 .219

    X7 - kemudahan dalam menggunakan untuk pemula .714 .030

    X8 - kemudahan dalam maintenance .722 .320

    X9 - promosi produk yang efektif .126 .765

    X10 - design yang menarik -.026 .749

    Extraction Method: Principal Component Analysis.

    Rotation Method: Varimax with Kaiser Normalization.

    a. Rotation converged in 3 iterations.

    We can see on the table above, each factor loading has significant score.

    So, it makes us to classify variable into factor. Usually, factor loading score + 0.4

    is considered minimum bound, + 0.5 considered better and + 0.6 considered

    significant. Since all factor loading scores is over 0.6, we may say that classifying

    variable using factor analysis is suitable.

  • 7/29/2019 Final Exam AM(1)

    37/44

    33

    Universitas Indonesia

    From 9 variables, 5 variables are put into factor 1, and 4 variables

    into factor 2. The next step is giving name for each factor. The name of the new

    factors can be seen in table below:

    Table 4.13 Classified Factor from Variables

    Factor Name Variables

    1

    Kemudahan

    Penggunaan

    dan Perbaikan

    BB

    kepuasan anda terhadap BB

    kecepatan operasi/penggunaan program BB yang baik

    service pasca penjualan

    kemudahan dalam menggunakan untuk pemula

    kemudahan dalam maintenance

    2Harga dan

    Desain

    kesesuaian service (program yang dimasukkan) dengan

    kebutuhan anda

    kesesuaian harga dengan produk

    promosi produk yang efektif

    design yang menarik

    Besides classification variables into factor, the other important result of

    factor analysis is factor score. The factor score is composite measures created for

    each observation on each factor extracted in factor analysis. The factor score then

    can be used to represents the factor in subsequent analysis. The factor score of our

    project is shown in table below:

  • 7/29/2019 Final Exam AM(1)

    38/44

    34

    Universitas Indonesia

    Table 4.14 Component Score Coefficient Matrix

    Component Score Coefficient Matrix

    Component

    1 2

    X1 - kepuasan anda terhadap BB .312 -.092

    X2 - kecepatan operasi/penggunaan program BB yang baik .285 -.077

    X4 - kesesuaian service (program yang dimasukkan) dengan

    kebutuhan anda

    -.049 .333

    X5 - kesesuaian harga dengan produk -.008 .302

    X6 - service pasca penjualan .239 .022

    X7 - kemudahan dalam menggunakan untuk pemula .261 -.064

    X8 - kemudahan dalam maintenance .232 .066

    X9 - promosi produk yang efektif -.037 .333

    X10 - design yang menarik -.091 .343

    Extraction Method: Principal Component Analysis.

    Rotation Method: Varimax with Kaiser Normalization.

    Component Scores.

    For each case and each component, the factor score is computed by

    multiplying the cases standardized variable values (computed using listwise

    deletion) by the components score coefficients. The resulting four component

    score variables are representative of, and can be used in place of, the 9 original

    variables.

  • 7/29/2019 Final Exam AM(1)

    39/44

    35

    Universitas Indonesia

    Figure 4.2 Component Plot in Rotated Space

    4.10Component Transformation Matrix

    The factor transformation matrix describes the specific rotation applied to

    your factor solution. This matrix is used to compute the rotated factor matrix

    (Component Score Coefficient matrix) from the original (unrotated) factor matrix.

    Table 4.15 Component Transformation Matrix

    Component Transformation Matrix

    Component 1 2

    1 .824 .567

    2 -.567 .824

    Extraction Method: Principal Component Analysis.

    Rotation Method: Varimax with Kaiser Normalization.

  • 7/29/2019 Final Exam AM(1)

    40/44

    36

    Universitas Indonesia

    CHAPTER 5

    CONCLUSION

    From the research that we conducted, we take some conclusions, such as:

    1. Customer satisfaction is an important for the company sustainabilitybecause the customer is the reason for the company to create their

    products and deliver their services.

    2. The company must hear their voice of customer by conducting surveysand improve their subsequent product for the higher level of customer

    satisfaction. The company not only should conduct the survey but they

    also should processing the data with the appropriate method.

    3. In this research we use factor analysis to process the data that we hadgotten from the questionaire, which all questions are about aspects that

    influencing customer satisfaction.

    4. From the questionaire, we know that 9 from 10 factors has greatinfluence on customer satisfaction. This aspect is grouped into two, and

    the details shows in the table below.

    Factor Name Variables

    1Easy to use and easy

    to maintenance

    Satisfaction for BB

    Operation /program speed excellence

    After sales service

    Easy to use for beginner

    Easyto maintenance

    2 Price and design

    Suitability betwee available service to

    customer needsValue for money

    Effective product promotion

    Attractive design

    8. Factor number three about Blackberry reliability is not significantlyaffected customer preference for blackberry.

  • 7/29/2019 Final Exam AM(1)

    41/44

    37

    Universitas Indonesia

    REFERENCES

    DeCoster, J. (1998). Overview of Factor Analysis. Retrieved from http://www.stat-help.com/notes.html

    Hair, Joseph F. Jr., Anderson, Rolph E.Multivariate Data Analysis. Fifth Edition.

    1984. New Jersey: Prentice Hall

    http://www.stat-help.com/notes.htmlhttp://www.stat-help.com/notes.html
  • 7/29/2019 Final Exam AM(1)

    42/44

    38

    Universitas Indonesia

    THE ATTACHMENTS CHAPTER

    Kuesioner Pengguna Blackberry

    Kami mahasiswa Teknik Industri Universitas Indonesia akan mengadakan

    penelitian tentang kepuasaaan konsumen terhadap penggunaan Blackberry untuk

    kebutuhan tugas perkuliahaan. Diharapkan saudara/i membantu kami dengan

    mengisi kuisioner ini.

    Jenis Kelamin :

    Usia :

    Jenis Blackberry :

    Untuk menjawab pertanyaang dibawah ini. Silahkan lingkari jawaban yang sesuai

    dengan pilihan anda.

    1. Apakah anda puas dengan Blackberry yang anda miliki? Ya Tidak

    2. Apakah yang paling anda perhatikan ketika membeli Blackberry? Apakah haltersebut telah terpenuhi? (sebutkan dengan singkat)

    3. Berikan skala pada aspek penilaian dibawah ini (1 = sangat tidak puas dan 10= sangat puas) dengan melingkari pilihan yang ada.

  • 7/29/2019 Final Exam AM(1)

    43/44

    39

    Universitas Indonesia

    Kepuasan anda terhadap BB1 2 3 4 5 6 7 8 9 10

    Kecepatan operasi/penggunaan program Blackberry yang baik.1 2 3 4 5 6 7 8 9 10

    Kehandalan Blackberry dalam menangani program dan gangguaneksternal seperti terjatuh.

    1 2 3 4 5 6 7 8 9 10

    Kesesuaian service (program yang dimasukkan) dengan kebutuhan anda.1 2 3 4 5 6 7 8 9 10

    Kesesuaian harga dengan produk.1 2 3 4 5 6 7 8 9 10

    Service pasca penjualan.1 2 3 4 5 6 7 8 9 10

    Kemudahan dalam menggunakan untuk pemula.1 2 3 4 5 6 7 8 9 10

    Kemudahan dalam maintenance.1 2 3 4 5 6 7 8 9 10

  • 7/29/2019 Final Exam AM(1)

    44/44

    40

    Promosi produk yang efektif.1 2 3 4 5 6 7 8 9 10

    Design yang menarik.1 2 3 4 5 6 7 8 9 10

    4. Jika ada produk Blackberry baru, apa yang menjadi keinginan anda terkaitfaktor-faktor diatas?

    ..

    5. Jika ada produk Blackberry baru, dengan service yang sama pada saat ini,apakah anda akan membelinya?

    Ya Mungkin akan membeli Ragu-ragu Mungkin tidak akan membeli Tidak

    6. Jika ada produk Blackberry yang mengalami perbaikan dalam faktor diatas(dalam tabel), apakah anda akan membeli? Misalnya pengurangan jumlah

    program yang belum dibutuhkan oleh mahasiswa.

    Ya Mungkin akan membeli Ragu-ragu Mungkin tidak akan membeli