Anda di halaman 1dari 21

Nonparametric Methods :

Chi-Square Applications
Characteristics of
The Chi-Square Distribution
 It is never negative
 There is a family of chi-square
distributions
 The shape of the chi-square distribution does
not depend on the size of the sample, but
the number of categories used (k)
 It is positively skewed
 As the number of both d.f. increases, the
distribution begins to approximate the
normal distribution
2-2
CHI-SQUARE
CHI-SQUARE DISTRIBUTION
DISTRIBUTION

df = 3

df = 5
df = 10


Chi-Square Test
 Compare several proportion (Multinomial
Test)
 One of nonparametric or distribution-free
tests of hypothesis
 Data : nominal-scale or ordinal-scale
 The test statistic is :

x  
2  f0  fe  
2


 fe 
cont..
 Chi-square test is used to :
 Test whether an observed set of frequencies
could have come from a hypothesized
population distribution
 Determine whether the sample observations
come from a particular distribution such as the
normal distribution
 Contingency table analysis, is used to test
whether two traits or characteristics are
related (Test of Independency)
Goodness-of-Fit Test:
Equal Expected Frequencies
 The purpose of Goodness-of-Fit Test is to
compare an observed set of frequencies (fo)
to an expected set of frequencies (fe).
 Ho : no difference between fo and fe
 H1 : there is a difference between fo & fe
 The critical value is a chi-square value with
(k - 1) degrees of freedom, where k is the
number of categories
Contoh : Penjualan Kaos Pemain Sepak Bola
Jumlah Jumlah yang
Pemain Terjual Diharapkan Terjual
(fo) (fe)
Owen 13 20
Ronaldo 33 20
Nesta 14 20
Buffon 7 20
Beckham 36 20
Zidane 17 20
TOTAL 120 120
Cont..
Pemain fo fe (fo – fe) (fo – fe)2 ( fo  fe )2
fe
Owen 13 20 -7 49 2,45
Ronaldo 33 20 13 169 8,45
Nesta 14 20 -6 36 1,80
Buffon 7 20 -13 169 8,45
Beckham 36 20 16 256 12,80
Zidane 17 20 -3 9 0,45
0 34,40
2
X
Goodness-of-Fit Test:
Unequal 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 ?
Limitations of Chi-Square
 If there are only two cells, the expected
frequency in each cell should be 5 or more
 For more than two cells, Chi-Square should
not be used if more than 20% of the
expected frequency cells have expected
frequency less than 5.
Example
Level of Management fo fe
Foreman 30 32
Supervisor 110 113
Manager 86 87
Middle Manager 23 24
Assistant vice president 5 2
Vice president 5 4
Senior vice president 4 1
TOTAL 263 263
Level of Management fo fe
Foreman 30 32
Supervisor 110 113
Manager 86 87
Middle Manager 23 24
Vice president 14 7
TOTAL 263 263
Goodness-of-Fit Test for Normality
 Purpose: To test whether the observed
frequencies in a frequency distribution
match the theoretical normal distribution.
 Procedure:
 Determine the mean and standard deviation of
the frequency distribution.
 Compute the z-value for the lower class limit
and the upper class limit for each class.
 Determine fe for each category
 Use the chi-square goodness-of-fit test to
determine if fo coincides with fe.
EXAMPLE : Distribution of Salary
Salary ($ 000) frequency   54.03
20 – 30 4   13.76
30 – 40 20
40 – 50 41
50 – 60 44
60 – 70 29
70 – 80 16
80 – 90 2
90 – 100 4
TOTAL 160
Salary (S 000) Z Value Area fe
Under 30 Under –1.75 0.0401 6.416
30 – 40 -1.75 to -1.02 0.1138 18.208
40 – 50 -1.02 to -0.29 0.2320 37.120
50 – 60 -0.29 to 0.43 0.2805 44.880
60 – 70 0.43 to 1.16 0.2106 33.696
70 – 80 1.16 to 1.89 0.0936 14.976
80 or more over 1.89 0.0294 1.704
1 160
x
Z 

Calculation for Chi-Square
( fo  fe )2
Salary (S 000) fo fe (fo – fe) (fo – fe)2 fe
Under 30 4 6.416 -2.416 5.837 0.910
30 – 40 20 18.20 1.792 3.211 0.176
8
40 – 50 41 37.12 3.880 15.054 0.406
0
50 – 60 44 44.88 -0.880 0.774 0.017
0
60 – 70 29 33.69 -4.696 22.052 0.654
6
70 – 80 16 14.97 1.024 1.049 0.070
2
6
X
 Suppose we knew the mean and standard
deviation of population but wished to find
whether some sample data conform to the
normal distribution,
d.f. = k - 1
 On the other hand, if we don’t know the
mean and standard deviation of population
but we wish to test whether some sample
data follow the normal distribution,
d.f. = k – p – 1
(where p is the number of population parameter
being estimated from the sample data)
Contingency Table Analysis
 Contingency table analysis is used to test whether
two traits or variables are related.
 Two-way classification table
 Each observation is classified according to two
variables.
 d.f. : (number of rows-1)(number of columns-1).
 The expected frequency (fe) is computed as:
fe 
 Row _ total  Coloumn _ total 
Grand _ total 2
X
 Coefficient of Contingency : C
X2N
Contoh
Manajer produksi meneliti tingkat kerusakan
pada mesin produksi. Hasilnya pengamatan
terhadap barang yang diproduksi sebagai berikut

Kondisi Mesin 1 Mesin 2 Mesin 3


Rusak 12 15 6
Baik 88 105 74

Apakah kerusakan tersebut disebabkan mesin


atau kebetulan saja ? Uji dengan  = 0,05
Contoh
Lembaga riset meneliti apakah ada hubungan
antara jenis surat kabar yang dibaca dengan
kelompok masyarakat. Hasilnya sebagai
berikut :
Surat Kabar
Kelompok A B C
Atas 170 124 90
Menengah 120 112 100
Bawah 130 90 88

Uji dengan  = 0,1


Chi-square Contingency Table Test for Independence
A B C
Atas 170 124 90 384
Menengah 120 112 100 332
Bawah 130 90 88 308
420 326 278 1024

expected values:
A B C
Atas 157.50 122.25 104.25 384.00
Menengah 136.17 105.70 90.13 332.00
Bawah 126.33 98.05 83.62 308.00
420.00 326.00 278.00 1024.00

7.34 chi-square
4 df
.1190 p-value

Anda mungkin juga menyukai