Program Pascasarjana Ilmu Manajemen Semester II Metodologi Riset Bisnis

Pertemuan

Pertemuan ke

ke--9

9

Analisis

Analisis Univariate

Univariate,, Bivariate

Bivariate,, dan

dan

Multivariate

Pengujian Hipotesis

Pemberian

Alasan Deduktif

Pemberian

Prosedur Statistik

Statistik

Deskriptif

Statistik

Pendekatan-Pendekatan Terhadap

Pengujian Hipotesis

Statistik klasik

• Pandangan objektif probabilitas

• Hipotesis yang dibuat ditolak atau gagal

ditolak

• Analisis berdasarkan data sampel

Statistik bayesian

• Penambahan dari pendekatan klasik

• Analisis berdasarkan data sampel

Tipe-Tipe Hipotesis

• Nol

– H₀: = 50 mpg

– H₀: < 50 mpg – H₀: > 50 mpg

• Alternatif

– H_A: = 50 mpg

Aturan Keputusan

Exhibit 18-3

Faktor-Faktor yang mempengaruhi

Probabilitas Melakukan Kesalahan

Nilai sebenarnya parameter

Memilih level alpha

Uji One atau two-tailed

Deviasi standard sampel

Prosedur Uji Statistik

Memperoleh nilai uji kritis

Menginter-pretasikan

tesnya Tahapan

-Tahapan

Memilih uji statistik Menyatakan

hipotesis nol

Memilih tingkat signifikannya Menghitung

Uji Signifikansi

Asumsi-Asumsi untuk Penggunaan

Uji Parametrik

Pengamatan independen

Distribusi normal

Variansnya sama

Kelebihan-Kelebihan Uji Nonparametrik

Mudah dipahami dan digunakan

Dapat digunakan dengan data nominal

Cocok untuk data ordinal

Bagaimana untuk Memiliih

Sebuah Uji

Berapa jumlah sampel yang terlibat?

Bila dua atau lebih sampel yang terlibat, apakah kasus-kasus individual independen

atau saling berhubungan?

Apakah skala pengukurannya

Exhibit 18-7 Merekomendasikan

Teknik-Teknik Statistik

Two-Sample Tests

____________________________________________

k-Sample Tests

____________________________________________

Measurement

Scale One-Sample Case Related Samples

Independent

Samples Related Samples

Independent Samples

Nominal • Binomial

•x2_{one-sample test}

• McNemar • Fisher exact test •x2_two-samples

test

• Cochran Q •x2_forksamples

Ordinal • Kolmogorov-Smirnov one-sample test

• Runs test

• Sign test

•Wilcoxon matched-pairs test

• Median test

•Mann-Whitney U •Kolmogorov-Smirnov

•Wald-Wolfowitz

• Friedman two-way ANOVA

• Median extension •Kruskal-Wallis one-way ANOVA

Interval and Ratio

• Repeated-measures ANOVA

• One-way ANOVA

Pertanyaan yang dijawab oleh Uji

One-Sample

• Apakah ada perbedaan antara frekuensi

pengamatan dan frekuensi yang kita

harapkan?

• Apakah ada perbedaan antara proporsi

pengamatan dan proporsi yang

diharapkan?

Uji Parametrik

Contoh One-Sample

t

-Test

Null Ho: = 50 mpg

Statistical test t-test

Significance level .05, n=100 Calculated value 1.786

Critical test value 1.66

Contoh One Sample Chi-Square

Test

Living Arrangement

Intend to Join

Number Interviewed

Percent

(no. interviewed/200)

Expected Frequencies (percent x 60)

Dorm/fraternity 16 90 45 27

Apartment/rooming

house, nearby 13 40 20 12

Apartment/rooming

house, distant 16 40 20 12

Live at home 15

_____

Contoh One-Sample Chi-Square

Null Ho: 0 = E

Statistical test One-sample chi-square Significance level .05

Calculated value 9.89 Critical test value 7.82

Contoh Two-Sample t-Test

A Group B Group

Average hourly sales X₁= $1,500 X2= $1,300

Contoh Two-Sample t-Test

Null Ho: A sales = B sales

Statistical test t-test

Significance level .05 (one-tailed) Calculated value 1.97, d.f. = 20 Critical test value 1.725

Uji Nonparametrik Two-Sample :

Chi-Square

On-the-Job-Accident

Cell Designation Count

Expected Values Yes No Row Total

Smoker

Heavy Smoker

1,1

Contoh Two-Sample Chi-Square

Null There is no difference in

distribution channel for age categories.

Statistical test Chi-square Significance level .05

Calculated value 6.86, d.f. = 2 Critical test value 5.99

Uji Two-Related-Samples

Exhibit 18-9 Data Penjualan untuk

Paired-Samples t-Test

Company

Sales

Year 2

Sales

Year 1 Difference D D2

Contoh Paired-Samples t-Test

Null Year 1 sales = Year 2 sales

Statistical test Paired sample t-test Significance level .01

Calculated value 6.28, d.f. = 9 Critical test value 3.25

Uji Nonparametrik Related-Samples:

McNemar Test

Before After

Do Not Favor

After Favor

Favor A B

Sebuah Contoh dari McNemar Test

Before After

Do Not Favor

After Favor

Favor A=10 B=90

k-Independent-Samples Tests:

ANOVA

• Uji hipotesis nol yang rerata dari tiga atau

lebih populasi adalah sama

• One-way: Menggunakan faktor tunggal,

Exhibit 18-12

Contoh ANOVA

__________________________________________Model Summary_________________________________________

Source d.f. Sum of Squares Mean Square FValue p Value

Model (airline) 2 11644.033 5822.017 28.304 0.0001

Residual (error) 57 11724.550 205.694

Total 59 23368.583

_______________________Means Table________________________

Count Mean Std. Dev. Std. Error

Delta 20 38.950 14.006 3.132

Lufthansa 20 58.900 15.089 3.374

KLM 20 72.900 13.902 3.108

Contoh ANOVA (Lanjutan)

Null A1 = A2 = A3

Statistical test ANOVA and F ratio Significance level .05

Calculated value 28.304, d.f. = 2, 57 Critical test value 3.16

Post Hoc:

Prosedur

Scheffe’s S

Multiple Comparison

Verses Diff

Crit.

Diff. _{p Value}

Delta Lufthansa 19,950 11.400 .0002

KLM 33.950 11.400 .0001

Exhibit 18-13 Prosedur Multiple

Bonferroni X X X

Tukey HSD X X X

Tukey-Kramer X X X

Games-Howell X X X

Tamhane T2 X X X

Scheffé S X X X X

Brown-Forsythe X X X X

Newman-Keuls X X

Duncan X X

Dunnet’s T3 X

Exhibit 18-15 Contoh Two-Way

ANOVA

__________________________________________Model Summary_________________________________________

Source d.f. Sum of Squares Mean Square FValue p Value

Airline 2 11644.033 5822.017 39.178 0.0001

Seat selection 1 3182.817 3182.817 21.418 0.0001

Airline by seat selection 2 517.033 258.517 1.740 0.1853

Residual 54 8024.700 148.606

All data are hypothetical

__________Means Table Effect: Airline by Seat Selection___________

Count Mean Std. Dev. Std. Error

Delta economy 10 35.600 12.140 3.839

Delta business 10 42.300 15.550 4.917

Lufthansa economy 10 48.500 12.501 3.953

Lufthansa business 10 69.300 9.166 2.898

KLM economy 10 64.800 13.037 4.123

k-Related-Samples Tests

Lebih dari dua level dalam

faktor pengelompokkan

Pengamatannya sesuai

Exhibit 18-17 Contoh

Repeated-Measures ANOVA

___________________________________Means Table by Airline _________________________________________________________________________

Count Mean Std. Dev. Std. Error

Rating 1, Delta 20 38.950 14.006 3.132

Rating 1, Lufthansa 20 58.900 15.089 3.374

Rating 1, KLM 20 72.900 13.902 3.108

Rating 2, Delta 20 32.400 8.268 1.849

Rating 2, Lufthansa 20 72.250 10.572 2.364

Rating 2, KLM 20 79.800 11.265 2.519

__________________________________________________________Model Summary_________________________________________________________ Source d.f. Sum of Squares Mean Square FValue p Value

Airline 2 3552735.50 17763.775 67.199 0.0001

Subject (group) 57 15067.650 264.345

Ratings 1 625.633 625.633 14.318 0.0004

Ratings by air... 2 2061.717 1030.858 23.592 0.0001

Ratings by subj... 57 2490.650 43.696

All data are hypothetical.

______________________________________Means Table Effect: Ratings_________________________________________________________________

Count Mean Std. Dev. Std. Error

Rating 1 60 56.917 19.902 2.569

Terminologi Kunci

• a priori contrasts • Hipotesis alternatif • Analysis of variance

(ANOVA)

• Statistik bayesian • Uji Chi-square

• Classical statistics • Nilai kritis

• F ratio

• Statistik inferensial

• Uji K-independent-samples

• Uji K-related-samples • Level signifikansi

• Mean square

• Uji Multiple comparison (range tests)

• Uji Nonparametrik

Terminologi Kunci

• Hipotesis nol

• Observed significance level

• Uji One-sample • Uji One-tailed • p value

• Uji Parametrik

• Power of the test

• Practical significance

• Region of acceptance • Region of rejection

• Signifikansi statistik • Distribusi t

• Trials • t-test

Terminologi Kunci

• Uji Two-related-samples

• Uji Two-tailed • Kesalahan tipe I

• Kesalahan tipe II • Distribusi Z

Ukuran

Exhibit 19-1 Ukuran Asosiasi:

Interval/Rasio

Koefisien korelasi pearson Untuk variabel kontinyu yang

berhubungan secara linier

Correlation ratio (eta)

Untuk data nonlinear atau

menghubungkan efek utama ke variabel dependen kontinyu

Biserial

Satu variabel kontinyu dan satu

variabel dichotomous dengan distribusi normal

Partial correlation Tiga variabel; berkaitan dua dengan

efek ketiga diambil

Multiple correlation Tiga variabel; menghubungkan satu

variabel dengan dua variabel lainnya

Bivariate linear regression Memprediksi satu variabel dari nilai

Exhibit 19-1 Ukuran Asosiasi :

Ordinal

Gamma

Berdasarkan concordant-discordant

pairs; interpretasi proportional reduction in error (PRE)

Kendall’s tau b P-Q based; adjustment for tied ranks

Kendall’s tau c P-Q based; adjustment for table

dimensions

Somers’s d P-Q based; asymmetrical extension of

gamma

Spearman’s rho Korelasi product moment untuk data

Exhibit 19-1 Ukuran Asosiasi :

Nominal

Phi Chi-square based for 2*2 tables

Cramer’s V CS based; adjustment when one table

dimension >2

Contingency coefficient C CS based; flexible data and distribution

assumptions

Lambda PRE based interpretation

Goodman & Kruskal’s tau PRE based with table marginals

emphasis

Uncertainty coefficient Berguna untuk tabel multidimensional

Hubungan

“Untuk benar-benar memahami motif dan tindakan konsumen, anda harus mencari hubungan antara apa yang mereka pikirkan dan rasakan serta apa yang sebenarnya mereka lakukan.”

Pearson’s Product Moment

Correlation

r

Apakah ada hubungan antara X dan Y?

Seberapa besar hubungannya?

Interpretasi Korelasi

X menyebabkan Y

Y menyebabkan X

X dan Y diaktifkan oleh satu atau lebih variabel

Exhibit 19-8

Interpretasi Koefisien

Suatu koefisien tidak benar-benar luar

biasa karena ia signifkan secara

statistik! Pada kenyataannya ia harus

Aplikasi Konsep

X

Suhu Rata-Rata (Celsius)

Y

Harga per Kotak (FF)

12 2,000

16 3,000

20 4,000

24 5,000

Menguji Goodness of Fit

Y tidak berhubungan terhadap X dan tidak ada pola sistematik adalah bukti

Ada nilai konstan dari Y untuk setiap nilai X

Datanya berhubungan tapi

Exhibit 19-19

Koefisien Determinasi:

r

2

Total proporsi varians dalam Y

dijelaskan oleh X

Exhibit 19-23 Perhitungan Concordant (P), Discordant (Q), Tied (Tx,Ty), and Total Paired

Exhibit 19-24 Data KDL untuk

Spearman’s Rho

_______ _____ Rank By_____ _____ _____

Applicant Panelx Psychologisty d d2

Terminologi Kunci

• Korelasi artefak • Analisis korelasi

Bivariat

• Distribusi bivariat normal

• Chi-square-based measures

• Koefisien kontigensi C • Cramer’s V

• Phi

• Koefisien determinasi (r2) • Concordant

• Matriks korelasi • Discordant

• Error term

Terminologi Kunci

• Linieritas

• Metode least squares • Ukuran ordinal

• Gamma

• Somers’s d

• Spearman’s rho • tau b

• tau c

• Koefisien korelasi pearson • Prediction and confidence

bands

• Proportional reduction in error (PRE)

Terminologi Kunci

• Intersep • Slope

• Residual

• Scatterplot

20-88

20-89

Mengklasifikasikan

Teknik-Teknik Multivariat

20-90

Exhibit 20-1

20-91

Exhibit 20-1

20-92

Exhibit 20-1

20-93

Teknik-Teknik Dependensi

Regresi Berganda

Analisis Diskriminan

MANOVA

Structural Equation Modeling (SEM)

20-94

Kegunaan-Kegunaan Regresi

Berganda

nilai DV

20-95

Generalisasi

20-96

20-97

Metode Pemilihan

Forward

Backward

20-98

Mengevaluasi dan Berhadapan

dengan Multikollinieritas

Pilih satu variabel

Dan hapus yang lainnya

Buat variabel baru

Yang merupakan gabungan dari yang lainnya

20-99

Analisis Diskriminan

Predicted Success

Actual Group

Number Note: Percent of “grouped” cases correctly classified: 83.33%

Unstandardized Standardized

20-100

20-101

Exhibit 20-5

20-102

20-103

20-104

Multivariate

20-105

Univariate

20-106

Structural Equation Modeling

(SEM)

Spesifikasi Model

Estimasi

Evaluation of Fit

Respesifikasi Model

20-107

20-108

Analisis Konjoin

Brand Bolle Hobbies Oakley Ski Optiks

Style* A

B

Flotation Yes

No

Yes Yes Yes

Price $100

$72

* A = multiple color choices for frames, lenses, and temples.

20-109

20-110

20-111

Exhibit 20-13

20-112

Teknik-Teknik Interdependensi

Analisis Faktor

Analisis Kluster

20-113

Exhibit 20-14

Analisis Faktor

20-114

Exhibit 20-15 Factor Matrices

A

_____Unrotated Factors_____

B

__Rotated Factors__

Variable I II h2 I II

Percent of variance Cumulative percent

20-115

20-116

Exhibit 20-17 Koefisien Korelasi, Studi

MBA Metro U

Variable Course V1 V2 V3 V10

V1

Financial Accounting Managerial Accounting Finance

Marketing

Human Behavior Organization Design Production

Probability

Statistical Inference Quantitative Analysis

20-117

Exhibit 20-18 Matriks Faktor, Studi

MBA Metro U

Variable Course Factor 1 Factor 2 Factor 3 Communality

V1

Percent of variance Cumulative percent

Financial Accounting Managerial Accounting Finance

Marketing

Human Behavior Organization Design Production

Probability

Statistical Inference Quantitative Analysis

20-118

Exhibit 20-19 Varimax Rotated

Matriks Faktor

Variable Course Factor 1 Factor 2 Factor 3

V1

Financial Accounting Managerial Accounting Finance

Marketing

Human Behavior Organization Design Production

Probability

Statistical Inference Quantitative Analysis

20-119

Analisis Kluster

Pilih sampel untuk dikluster

Definisikan variabel

Hitung persamaan

Pilih kluster mutually exclusive

20-120

20-121

Exhibit 20-21 Keanggotaan Kluster

________Number of Clusters ________

Film Country Genre Case 5 4 3 2

Cyrano de Bergerac Il y a des Jours Nikita

Les Noces de Papier Leningrad Cowboys . . . Storia de Ragazzi . . . Conte de Printemps Tatie Danielle

Crimes and Misdem . . . Driving Miss Daisy La Voce della Luna Che Hora E

Attache-Moi

White Hunter Black . . . Music Box

20-122

20-123

20-124

20-125

Terminologi Kunci

• Metode average linkage

• Backward elimination • Beta weights

• Centroid

• Analisis kluster • Kollinieritas

• Kommunalitas

• Confirmatory factor analysis

• Analisis Konjoin

• Teknik-teknik Dependensi • Analisis diskriminan

• Variabel dummy • Eigenvalue

20-126

Terminologi Kunci

• Faktor

• Forward selection • Holdout sample • Teknik-teknik

Interdependensi • Loadings

• Metric measures • Multikollinieritas

• Multidimensional scaling (MDS)

• Regresi berganda • Analisis multivariat • Multivaria analysis of

20-127

Terminologi Kunci

• Path diagram

• Analisis komponen-komponen prinsipal • Rotasi

• Kesalahan spesifikasi • Koefisien

terstandardisasi

• Stepwise selection • Stress index

• Structural equation modeling