Top Banner
Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved. 15 - 1
39

Statistik 1 11 15 edited_chi square

Jul 07, 2015

Download

Education

Selvin Hadi
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 1

Page 2: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 2

Understand the nature and role of chi-square distribution

Identify a wide variety of uses of the chi-square distribution

Conduct a test of hypothesis comparing an observed frequency distribution to an

expected frequency distribution

When you have completed this chapter, you will be able to:

Page 3: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 3

Conduct a hypothesis test to determine whether two attributes are

independent

Conduct a test of hypothesis for normality using the chi-square distribution

Page 4: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 4

Sebelumnya:• skala pengukuran data: interval atau rasio• pengujian rata-rata, proporsi, varians populasi• mengasumsikan populasi mengikuti distribusi normalMetode Parametrik

Sekarang, bagaimana dengan:• data skala pengukuran: nominal atau ordinal• asumsi tidak ada pada bentuk distribusi populasiMetode Non Parametrik

Salah satu Metode Non Parametrik: Chi Square (χ2)

Page 5: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 5Karakteristik Distribusi Chi-Square

Karakteristik Distribusi Chi-Square

… Menceng/menjulur ke kanan (positively skewed)

… non-negatif

… berdasarkan degrees of freedom

… Ketika degrees of freedom berubah maka distribusi baru akan terbentuk

…e.g.

Page 6: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 6

df = 3

df = 5

df = 10

χ2

Karakteristik Distribusi Chi-Square

Karakteristik Distribusi Chi-Square

Page 7: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

• Uji Chi-square digunakan untuk:1. Menguji apakah frekuensi aktual sama

dengan frekuensi yang diharapkan (berasal dari distribusi populasi yang dihipotesiskan).

2. Menentukan apakah sampel pengamatan berasal dari distribusi tertentu yang berdistribusi normal

3. Analisis tabel kontinjensi, digunakan untuk menguji apakah dua sifat atau karakteristik saling berkaitan (Uji Independensi)

Page 8: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 8Goodness-of-Fit Test:

Equal Expected Frequencies

Goodness-of-Fit Test: Equal Expected

Frequencies

f0: frekuensi observasi (observed)fe : frekuensi yang diharapkan (expected)

H0: Tidak ada perbedaan antara frekuensi observasi (observed) dan yang diharapkan (expected).

H1: Ada perbedaan antara frekuensi observasi (observed) dan yang diharapkan (expected).

Page 9: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 9

…Nilai kritis: nilai chi-square dengan degrees of freedom: k-1 ,

di mana k adalah jumlah kategori

Goodness-of-Fit Test: Equal Expected

Frequencies

Goodness-of-Fit Test: Equal Expected

Frequencies

( )∑

−=

e

eo

f

ff 22χ

… Uji statistik:

Page 10: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 10

Informasi berikut menunjukkan jumlah karyawan yang absen setiap hari dalam seminggu di sebuah pabrik.

Hari FrekuensiSenin 120Selasa 45Rabu 60Kamis 90Jumat 130

TOTAL 445

Hari FrekuensiSenin 120Selasa 45Rabu 60Kamis 90Jumat 130

TOTAL 445

Goodness-of-Fit Test: Equal Expected

Frequencies

Goodness-of-Fit Test: Equal Expected

Frequencies

Pada tingkat signifikansi 0.05, apakah ada perbedaan jumlah absensi per hari dalam

seminggu tersebut?

Page 11: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 11

Hypothesis Test Hypothesis Test

Step 1Step 1

Step 2Step 2

Step 3Step 3

Step 4Step 4

H0: Tidak ada perbedaan tingkat absensi per hari dalam seminggu tersebut…

H1: Tingkat absensi per hari tidak semua sama

α = 0.05

Use Chi-Square test

Tolak H0 jika χ2 >

(5-1) = 4Degrees of freedom

9.488 (dari tabel Chi-square)

(120+45+60+90+130)/5 = 89

Goodness-of-Fit Test: Equal Expected

Frequencies

Goodness-of-Fit Test: Equal Expected

Frequencies

Chi-Square

Page 12: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 12

Page 13: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 13

Right-Tail Area

α = 0.05

Degrees of Freedom 5 – 1

= 4

Reject H0 if χ2 > 9.488

Using the Table…Using the Table…

Page 14: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 14

= 1.98 = 1.98

Hari Frekuensi Expected (fo – fe)2/fe

Senin 120 89 10.80Selasa 45 89 21.75Rabu 60 89 9.45Kamis 90 89 0.01Jumat 130 89 18.89

Total 445 445 60.90

Hari Frekuensi Expected (fo – fe)2/fe

Senin 120 89 10.80Selasa 45 89 21.75Rabu 60 89 9.45Kamis 90 89 0.01Jumat 130 89 18.89

Total 445 445 60.90

( )∑

−=

e

eo

f

ff 22χ

χ2

Kesimpulan: Hypothesis Nol (H0) ditolak. Tingkat ansensi tidak sama pada tiap hari dalam seminggu

tersebut.

(120-89)2/89Step 5Step 5

Tes Statistik

Reject H0 if χ2 > 9.488

Page 15: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

Contoh : Penjualan Kaos Pemain Sepak Bola

PemainJumlah Terjual

(fo)

Jumlah yang diharapkan Terjual

(fe)

Owen 13 20

Ronaldo 33 20

Nesta 14 20

Dida 7 20

Becham 36 20

Zidane 17 20

TOTAL 120 120

Page 16: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

Cont..

Pemain fo fe (fo – fe) (fo – fe)2

Owen 13 20 -7 49 2,45

Ronaldo 33 20 13 169 8,45

Vieri 14 20 -6 36 1,80

Buffon 7 20 -13 169 8,45

Becham 36 20 16 256 12,80

Zidane 17 20 -3 9 0,45

0 34,40

2( )o e

e

f f

f

χ2

Page 17: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 17

Pada tingkat signifikansi 0.05, dapatkah disimpulkan bahwa area Philadelphia berbeda dengan Amerika

Serikat secara keseluruahan?

Married Widowed Divorced Single

63.9% 7.7% 6.9% 21.5%

Data Biro Sensus Amerika menunjukkan bahwa di Amerika Serikat:

Not re-married Never married

Sample 500 orang dewasa dari Philadelphia menunjukkan:

310 40 30 120

Page 18: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 18

Married 310 *319.5 ** .2825

Widowed 40 38.5 .0584

Divorced 30 34.5 .5870

Single 120 107.5 1.4535

Total 500 2.3814

… continued

χ2

2.3814

* Census figures would predict: i.e. 639*500 = 319.5

** Our sample: (310-319.5)2/319.5 = .2825

(fo – fe)2/feExpected

Page 19: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 19

Step 1Step 1

Step 2Step 2

Step 3Step 3

Step 4Step 4

H0: The distribution has not changed

… continued

H1: The distribution has changed.

H0 is rejected if

χ 2 >7.815, df = 3

α = 0.05

χ 2 = 2.3814

Reject the null hypothesis.The distribution regarding marital status in Philadelphia is different from the rest of the United States.

Page 20: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

Goodness-of-Fit Test: Equal Expected Frequencies

• Contoh :Dosen mengharapkan distribusi nilai ujian : A = 40%, B = 40%, dan C = 20%. Hasil ujian menunjukkan distribusi nilai sebagai berikut :

A : 30 orang B : 20 orang C : 10 orang

Uji dengan level of significance 10%, apakah distribusi nilai tersebut sesuai dengan harapan dosen tersebut ?

Page 21: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 21Goodness-of-Fit Test: Normality

Goodness-of-Fit Test: Normality

… tes ini menguji apakah frekuensi aktual (observed) pada sebuah distribusi frekuensi sesuai dengan distribusi normal.

(1) Hitung nilai-z untuk batas bawah kelas dan batas atas kelas pada tiap kelas(2) Tentukan fe untuk setiap kategori.

(3) Gunakan uji chi-square (goodness-of-fit test) untuk menentukan apakah fo sesuai dengan fe

…untuk menentukan rata-rata dan standar deviasi distribusi frekuensi

Page 22: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 22

Sampel uang saku 500 siswa dilaporkan dalam distribusi frekuensi berikut ini

Apakah dapat disimpulkan bahwa distribusinya normal dengan rata-rata $10 dan standar deviasi $2?

Gunakan tingkat signifikansi 0.05

Goodness-of-Fit Test: Normality

Goodness-of-Fit Test: Normality

Page 23: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 23

Uang saku fo Nilai Z Area-Z

fe

(fo- fe )

2/fe

<$6 20

$6 up to $8 60

$8 up to $10 140

$10 up to $12

120

$12 up to $14

90

>$14 70

Total 500

… continued

Page 24: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 24

To compute fe for the first class, first determine the z - value

To compute fe for the first class, first determine the z - value

… continued

Now… find the probability of a z - value less than –2.00

Now… find the probability of a z - value less than –2.00

00.22

106 −=−

=−

µXz

.02284772.5000.0)00.2( =−=−<zP

Page 25: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 25

Amount Spent fo Nilai -Z Area-Z

<$6 20 < -2 0.5-0.47= 0.02

$6 up to $8 60 <-2 to -1 (0.5-0.34)-0.02=0.14

$8 up to $10 140 -1 to 0 0.34

$10 up to $12 120 0 to 1 0.34

$12 up to $14 90 1 to 2 0.14

>$14 70 > 2 0.02

Total 500

… continued

Page 26: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 26

The expected frequency is the probability of a z-value less than –2.00 times the sample size

… continued

The other expected frequencies are computed similarly

40.11)500)(0228(. ==ef

Page 27: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 27

Amount Spent

fo Area

fe

(fo- fe )

2/fe

<$6 20 .02 11.40 6.49

$6 up to $8 60 .14 67.95 .93

$8 up to $10 140 .34 170.65 5.50

$10 up to $12 120 .34 170.65 15.03

$12 up to $14 90 .14 67.95 7.16

>$14 70 .02 11.40 301.22

Total 500 500 336.33

… continued

Page 28: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 28

… continued

Step 1Step 1

Step 2Step 2

Step 3Step 3

Step 4Step 4

H0: The observations follow the normal distribution

H0 is rejected if χ 2 >7.815, df = 6

α = 0.05

χ 2 = 336.33

H0: is rejected.

The observations do NOT follow the normal distribution

H0: The observations do NOT follow the normal distribution

Page 29: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 29

Tabel kontinjensi digunakan untuk analisis apakah dua sifat atau karakteristik

saling berkaitan

Tabel kontinjensi digunakan untuk analisis apakah dua sifat atau karakteristik

saling berkaitan

… Frekuensi yang diharapkan (expected) dihitung dengan cara:

Expected Frequency = (Total baris)(Total kolom)/grand total

… masing-masing observasi dikelompokkan berdasarkan dua kriteria

…Menggunakan prosedur uji hipothesis biasa

… degrees of freedom :

(jumlah baris -1)(jumlah kolom -1)

Page 30: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 30

Apakah ada hubungan antara lokasi kecelakaan dan jenis kelamin orang yang terlibat dalam kecelakaan itu?

Sampel 150 kecelakaan yang dilaporkan ke polisi dikelompokkan berdasarkan

lokasi dan jenis kelamin.

Pada tingkat signifikansi 0.05, dapatkah disimpulkan bahwa jenis kelamin dan

lokasi kecelakaan saling berhubungan?

Page 31: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 31

… continued

SexWork Home Other

Total

Male 60 20 10 90

Female 20 30 10 60

Total 80 50 20 150

Location

Frekuensi yang diharapkan (expected) untuk kombinasi work-male dihitung dengan cara: (90)

(80)/150 =48

Expected frequencies untuk yang lain bisa dihitung dengan cara yang sama.

Page 32: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 32

Sex Work Home Other Total Fe work

Fe Home

Fe Other

Male 60 20 10 9048 30 12 90

Female 20 30 10 6032 20 8 60

Total 80 50 20 150

Page 33: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 33

Categories fo

fe

(fo- fe )2 (fo- fe )2/fe

Male-Work 60 48 144 3.00

Male-Home 20 30 100 3.33

Male-Other 10 12 4 0.33

Female-Work 20 32 144 4.50

Female-Home 30 20 100 5.00

Female-Other 10 8 4 0.50

Total 150 150 16.667

Page 34: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 34

Step 1Step 1

Step 2Step 2

Step 3Step 3

Step 4Step 4

H0: The Gender and Location are NOT related

H0 is rejected if χ 2 >5.991, df = 2

α = 0.05

H0: is rejected.Gender and Location are related!

H0: The Gender and Location are related

(…there are (3- 1)(2-1) = 2 degrees of freedom)

Find the value of χ 2

( ) ( )8

810...

48

4860 222 −++−=χ

… continued

667.16=

Page 35: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

Latihan 1: Equal Expected Frequencies

Dosen mengharapkan distribusi nilai ujian : A = 40%, B = 40%, dan C = 20%. Hasil ujian menunjukkan distribusi nilai sebagai berikut :

A : 30 orang B : 20 orang C : 10 orang

Uji dengan level of significance 10%, apakah distribusi nilai tersebut sesuai dengan harapan dosen tersebut ?

Page 36: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

Latihan 2:

Manajer produksi meneliti tingkat kerusakan pada mesin produksi. Hasilnya pengamatan terhadap barang yang diproduksi sebagai berikut

Apakah kerusakan tersebut disebabkan mesin atau kebetulan saja ? Uji dengan α = 0,05

Kondisi Mesin 1 Mesin 2 Mesin 3

Rusak 12 15 6

Baik 88 105 74

Page 37: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

Latihan 3:Lembaga riset meneliti apakah ada hubungan antara jenis surat kabar yang dibaca dengan kelompok masyarakat. Hasilnya sebagai berikut :

Uji dengan α = 0,1

Surat Kabar

Kelompok A B C

Atas 170 124 90

Menengah 120 112 100

Bawah 130 90 88

Page 38: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 38

Test your learning…Test your learning…

www.mcgrawhill.ca/college/lindClick on…Click on…

Online Learning Centrefor quizzes

extra contentdata setssearchable glossaryaccess to Statistics Canada’s E-Stat data…and much more!

Page 39: Statistik 1 11 15 edited_chi square

Copyright © 2004 McGraw-Hill Ryerson Limited. All rights reserved.

15 - 39

This completes Chapter 15This completes Chapter 15