0% found this document useful (0 votes)
46 views9 pages

Hasil Permodelan Regresi Linier Multivariat: Coefficients

This document summarizes the results of a multiple linear regression analysis with birth weight as the dependent variable and mother's weight, smoking status, and presence of uterine irritability as the independent variables. The analysis found that all three independent variables were statistically significant predictors of birth weight, with presence of uterine irritability having the strongest influence. Specifically: 1) The regression model explained 11.5% of the variance in birth weight. 2) Higher mother's weight, nonsmoking status, and absence of uterine irritability were associated with higher birth weight. 3) Presence of uterine irritability had the largest standardized effect size and was the most dominant predictor of lower birth weight.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
46 views9 pages

Hasil Permodelan Regresi Linier Multivariat: Coefficients

This document summarizes the results of a multiple linear regression analysis with birth weight as the dependent variable and mother's weight, smoking status, and presence of uterine irritability as the independent variables. The analysis found that all three independent variables were statistically significant predictors of birth weight, with presence of uterine irritability having the strongest influence. Specifically: 1) The regression model explained 11.5% of the variance in birth weight. 2) Higher mother's weight, nonsmoking status, and absence of uterine irritability were associated with higher birth weight. 3) Presence of uterine irritability had the largest standardized effect size and was the most dominant predictor of lower birth weight.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 9

1.

HASIL PERMODELAN REGRESI LINIER


MULTIVARIAT

Model R R Square Adjusted R Std. Error of the


Square Estimate

1 .359a .129 .115 685.879

a. Predictors: (Constant), Presence of uterine irritability, Smoking


status, Weight of mother (pounds)

Coefficientsa

Model Unstandardized Coefficients Standardized t Sig.


Coefficients

B Std. Error Beta

(Constant) 2685.424 229.454 11.704 .000

Weight of mother (pounds) 3.336 1.656 .140 2.014 .045


1
Smoking status -249.198 102.476 -.167 -2.432 .016

Presence of uterine irritability -515.141 142.331 -.252 -3.619 .000

a. Dependent Variable: Birth weight (gram)


2. UJI ASUMSI DAN KOLENIARITAS
a. HOMOSCEDASTICITY
Hasil uji asumsi sbb:

Hasil diagram tebar terlihat bahwa tebaran residual merata diatas dan dibawah garis diagonal 0.
Artinya asumsi homoscedasticity terpenuhi.
b. EKSISTENSI
Hasil uji asumsi sbb:

Residuals Statisticsa

Minimum Maximum Mean Std. Deviation N

Predicted Value 2188.00 3489.50 2944.66 261.823 189


Std. Predicted Value -2.890 2.081 .000 1.000 189
Standard Error of Predicted
66.543 214.952 95.012 30.558 189
Value
Adjusted Predicted Value 2172.04 3464.10 2945.13 262.820 189
Residual -1925.140 1894.195 .000 680.384 189
Std. Residual -2.807 2.762 .000 .992 189
Stud. Residual -2.855 2.776 .000 1.003 189
Deleted Residual -1991.846 1913.226 -.474 694.965 189
Stud. Deleted Residual -2.912 2.828 -.001 1.008 189
Mahal. Distance .775 17.470 2.984 2.899 189
Cook's Distance .000 .071 .005 .010 189
Centered Leverage Value .004 .093 .016 .015 189

a. Dependent Variable: Birth weight (gram)

Nilai residual menunjukkan hasil nilai mean=0 dan standar deviasi =680,384
Berarti asumsi eksistensi terpenuhi.

c. INDEPENDENSI
Hasil uji asumsi sbb:

Model Summaryb

Model R R Square Adjusted R Std. Error of the Durbin-Watson


Square Estimate

1 .359a .129 .115 685.879 .259

a. Predictors: (Constant), Presence of uterine irritability, Smoking status, Weight of


mother (pounds)
b. Dependent Variable: Birth weight (gram)

Hasil uji didapatkan nilai Durbin-Watson=0,259


Berarti asumsi independensi terpenuhi
d. LINEARITAS
Hasil uji sbb:

ANOVAa

Model Sum of Squares df Mean Square F Sig.

Regression 12887607.267 3 4295869.089 9.132 .000b

1 Residual 87029445.378 185 470429.434

Total 99917052.646 188

a. Dependent Variable: Birth weight (gram)


b. Predictors: (Constant), Presence of uterine irritability, Smoking status, Weight of mother
(pounds)

Hasil uji Anova regresi menunjukkan nilai p value=0,005 < 0,05


Berarti asumsi linearitas terpenuhi
e. GAUSS/NORMALITAS
Hasil uji sbb:

Hasil uji menunjukkan :grafik normal p-p plot residual, titik tebarannya menyebar sekitar garis
diagonal.
Berarti asumsi normalitas terpenuhi.
f. KOLENIERITAS
Ketentuannya: bilaa antar independen terjadi hub yg sangat kuat : r ≥ 0,8. Atau nilai VIF
> 10
Bila terjaadi kolenieritas maka salah satu variable yang berkolenieritas harus dikeluarkan
dari model

Coefficientsa

Model Unstandardized Coefficients Standardized t Sig. Collinearity Statistics


Coefficients

B Std. Error Beta Tolerance VIF

(Constant) 2685.424 229.454 11.704 .000


Weight of mother (pounds) 3.336 1.656 .140 2.014 .045 .975 1.025
1
Smoking status -249.198 102.476 -.167 -2.432 .016 .995 1.005

Presence of uterine irritability -515.141 142.331 -.252 -3.619 .000 .974 1.027

a. Dependent Variable: Birth weight (gram)

Hasil dari penilaian kolinearitas ternyata didapatkan nilai VIF < 10 berarti antar variable
independen didalam model ini tidak terjadi kolinearitas.

Hasil uji asumsi pada model ini ternyata tidak ada yang dilanggar (semua memenuhi
asumsi regresi linear ganda/multivariate)

3. INTERPRETASI MODEL AKHIR REGRESI LINIER MULTIVARIATE

Model R R Square Adjusted R Std. Error of the


Square Estimate

1 .359a .129 .115 685.879

a. Predictors: (Constant), Presence of uterine irritability, Smoking


status, Weight of mother (pounds)

Coefficientsa
Model Unstandardized Coefficients Standardized t Sig.
Coefficients

B Std. Error Beta

(Constant) 2685.424 229.454 11.704 .000

Weight of mother (pounds) 3.336 1.656 .140 2.014 .045


1
Smoking status -249.198 102.476 -.167 -2.432 .016

Presence of uterine irritability -515.141 142.331 -.252 -3.619 .000

a. Dependent Variable: Birth weight (gram)

Hasil analisis multivariate regresi linier didapatkan ada tiga variable yang berhubungan
dengan berat badan bayi yaituberat ibu, merokok, dan kelainan uterus. Dari ketiga variable
tersebut, variable yang paling dominan berhubungan berat badan bayi adalah variable
kelainan uterus. Hasil analisis didapatkan R Square = 0,115 artinya ketiga variable yaitu
berat ibu, merokok dan kelainan uterus berpengaruh terhadap berat badan bayi sebesar
11,5% sedangkan sisanya dipengaruhi oleh factor lain.

Untuk mencari dominan dilihat dari kolom Beta. Dari kolom Beta mana yang paling besar
nilainya. Nilai yang tertinggi adalah nilai yang paling nominan. Pada hasil diatas nilai yang
paling dominan 0,252 ada pada Presence of uterine irritability ( tanpa melihat tanda minus )

Persamaan garis regresi linier adalah sbb:


Berat bayi = 2685,4+3,4berat ibu -249,2 merokok -515,1 lelainan uterus dilihat pada kolom
B

Koefisien b1 untuk berat ibu=3,3


Artinya setiap berat badan ibu bertambah 1 pound maka berat baayi akan bertambah sebesar
3,3 gram

Koefisien b2 untuk merokok=-249,2


Artinya ibu yang merokok berat badan bayi akan lebih rendah sebesar 249,2 gram
dibandingkan ibu yang tidak merokok

Koefisien b3 untuk uterus=-515,1


Artinya ibu yang mempunyai kelainan uterus berat badan bayi akan lebih rendah sebesar
515,5 gram dibandingkan ibu yang tidak ada kelainan uterus
INTERPRETASI REGRESI LOGISTIC MULTIVARIATE
Model Summary

Step -2 Log likelihood Cox & Snell R Nagelkerke R


Square Square
a
1 208.340 .130 .183

a. Estimation terminated at iteration number 4 because


parameter estimates changed by less than .001.

Variables in the Equation

B S.E. Wald df Sig. Exp(B)

merokok .937 .386 5.881 1 .015 2.551

ras .526 .206 6.500 1 .011 1.693

Step 1a hipertensi 1.288 .638 4.075 1 .044 3.624

premature 1.296 .435 8.878 1 .003 3.654

Constant -2.519 .540 21.804 1 .000 .081

a. Variable(s) entered on step 1: merokok, ras, hipertensi, premature.

Haasil permodelan regresi logistic didaptkan bahwa ada 4 variabel yang berhubungan dengan
kejadian BBLR yaitu merokok, Ras, riwayat hipertensi, riwayat premature. Dari ke 4 variabel
tersebut, variable yg paling dominan berhubungan dengan kejadian BBLR adalah riwayat
premature, dengan OR=3,654 artinya ibu yg mempunyai riwayat premature punya resiko
melahirkan bayi BBLR sebesar 3,654 lebih tinggi dibandingkan ibu yg tidak ada riwayat
premature.
Hasil analisis didaptkan R Square 0,183 artinya variable merokok, ras, riwayat hipertensi dan
riwayat premature berpengaruh terhadap kejadian BBLR sebesar 18,3% sedangkan sisanya
dipengaruhi oleh fakto-faktor lain.
Persamaan garis regresi logistic:
P (BBLR)= -2,5+ 0,9 merokok +0,5 ras +1,28 hipertensi +1,3 prematur

You might also like