Post on 09-Jun-2018
ANOVA Menguji apakah rerata 1 variabel berbeda
secara bermakna pada lebih dari 2 kategori (beda rerata kadar kolesterol antar 3 kategori usia) Buka SPSS: file – data –dietstudy Analyze – Compare means – One-way ANOVA:
Dependent List : data rasio (wg0) Factor : lebih 2 kategori (agegroup) Option:
Statistics: descriptive dan homogeneity-of-variance Post-hoc: Bonferroni dan Tukey
Continue OK
Tests of Normality
.156 16 .200* .938 16 .320CholesterolStatistic df Sig. Statistic df Sig.
Kolmogorov-Smirnova Shapiro-Wilk
This is a lower bound of the true significance.*.
Lilliefors Significance Correctiona.
Descriptives
Cholesterol
5 187.40 29.433 13.163 150.85 223.95 158 2336 215.50 37.212 15.192 176.45 254.55 151 2575 188.80 29.987 13.410 151.57 226.03 157 222
16 198.38 33.472 8.368 180.54 216.21 151 257
<5050-60>60Total
N Mean Std. Deviation Std. Error Lower Bound Upper Bound
95% Confidence Interval forMean
Minimum Maximum
Multiple Comparisons
Dependent Variable: Cholesterol
-28.10 19.861 .362 -80.54 24.34-1.40 20.744 .997 -56.17 53.3728.10 19.861 .362 -24.34 80.5426.70 19.861 .397 -25.74 79.141.40 20.744 .997 -53.37 56.17
-26.70 19.861 .397 -79.14 25.74-28.10 19.861 .542 -82.64 26.44-1.40 20.744 1.000 -58.36 55.5628.10 19.861 .542 -26.44 82.6426.70 19.861 .605 -27.84 81.241.40 20.744 1.000 -55.56 58.36
-26.70 19.861 .605 -81.24 27.84
(J) age grouping50-60>60<50>60<5050-6050-60>60<50>60<5050-60
(I) age grouping<50
50-60
>60
<50
50-60
>60
Tukey HSD
Bonferroni
MeanDifference
(I-J) Std. Error Sig. Lower Bound Upper Bound95% Confidence Interval
ANOVA
Cholesterol
2820.250 2 1410.125 1.311 .30313985.500 13 1075.80816805.750 15
Between GroupsWithin GroupsTotal
Sum ofSquares df Mean Square F Sig.
GLM - univariat
Menguji hubungan usia dengan kadar kolesterol:Buka SPSS: file – data –dietstudyAnalyze – General Linear Model – Univariate:
Dependent variable: masukkan variabel wgt0 (data rasio)
Covariate: masukkan variabel age (data rasio) sebagai variabel independen
OK
GLM - univariat
Tests of Between-Subjects Effects
Dependent Variable: Cholesterol
306.756a 1 306.756 .260 .61812956.153 1 12956.153 10.994 .005
306.756 1 306.756 .260 .61816498.994 14 1178.500
646448.000 1616805.750 15
SourceCorrected ModelInterceptAGEErrorTotalCorrected Total
Type III Sumof Squares df Mean Square F Sig.
R Squared = .018 (Adjusted R Squared = -.052)a.
GLM – univariat (options)
Kadar kolesterol darah = 234,135 – 0,654 usia
Parameter Estimates
Dependent Variable: Cholesterol
234.135 70.614 3.316 .005 82.682 385.587-.654 1.282 -.510 .618 -3.403 2.095
ParameterInterceptAGE
B Std. Error t Sig. Lower Bound Upper Bound95% Confidence Interval
Correlations
1 -.135. .618
16 16-.135 1.618 .
16 16
Pearson CorrelationSig. (2-tailed)NPearson CorrelationSig. (2-tailed)N
Age in years
Cholesterol
Age in years Cholesterol
Linear Regression
45 50 55 60
Age in years
150
175
200
225
250
Cho
lest
erol
Cholesterol = 234.13 + -0.65 * ageR-Square = 0.02
GLM - univariat
Menguji hubungan jenis kelamin dan kategori usia dengan kadar kolesterol darah: Buka SPSS: file – data –dietstudy Analyze – General Linear Model – Univariate:
Dependent variable: masukkan variabel wgt0 (data rasio) Fixed factors: masukkan variabel jenis kelamin dan kategori
usia (data nominal/ordinal) OK
GLM - univariat
Tests of Between-Subjects Effects
Dependent Variable: Cholesterol
13794.217a 5 2758.843 9.161 .002462728.955 1 462728.955 1536.523 .00010936.955 1 10936.955 36.317 .000
37.922 2 18.961 .063 .939491.654 2 245.827 .816 .469
3011.533 10 301.153646448.000 1616805.750 15
SourceCorrected ModelInterceptGENDERAGEGROUPGENDER * AGEGROUPErrorTotalCorrected Total
Type III Sumof Squares df Mean Square F Sig.
R Squared = .821 (Adjusted R Squared = .731)a.
GLM – univariat (plots)Estimated Marginal Means of Cholesterol
age grouping
>6050-60<50
Estim
ated
Mar
gina
l Mea
ns240
220
200
180
160
140
Gender
Male
Female
MANOVA (GLM-multivariate)
Variabel dependen lebih dari 1 tapi kelompok1, semisal:
Bagaimana rata-rata kadar trigliserida dan kolesterol (2 variabel dependen) berbeda secara bermakna untuk tiap kelompok usia (1 kelompok)
MANOVA (GLM-multivariate)
Variabel dependen lebih dari 1 tapi kelompok1
Buka SPSS: file – data –dietstudy Analyze – General Linear Model – Multivariate:
Dependent variables: masukkan variabel kolesterol dan trigliserida darah
Fixed factor: masukkan variabel kategori usia Options - Display: aktifkan Homogenity test Continue dan OK
MANOVA (GLM-multivariate)
Box's Test of Equality of Covariance Matricesa
.864
.1116
3329.708.995
Box's MFdf1df2Sig.
Tests the null hypothesis that the observed covariancematrices of the dependent variables are equal across groups.
Design: Intercept+AGEGROUPa.
MANOVA (GLM-multivariate)
Levene's Test of Equality of Error Variancesa
.095 2 13 .910
.617 2 13 .555CholesterolTriglyceride
F df1 df2 Sig.
Tests the null hypothesis that the error variance of the dependentvariable is equal across groups.
Design: Intercept+AGEGROUPa.
MANOVA (GLM-multivariate)
Multivariate Testsc
.982 322.080a 2.000 12.000 .000
.018 322.080a 2.000 12.000 .00053.680 322.080a 2.000 12.000 .00053.680 322.080a 2.000 12.000 .000
.231 .849 4.000 26.000 .507
.770 .838a 4.000 24.000 .515
.298 .818 4.000 22.000 .527
.293 1.905b 2.000 13.000 .188
Pillai's TraceWilks' LambdaHotelling's TraceRoy's Largest RootPillai's TraceWilks' LambdaHotelling's TraceRoy's Largest Root
EffectIntercept
AGEGROUP
Value F Hypothesis df Error df Sig.
Exact statistica.
The statistic is an upper bound on F that yields a lower bound on the significance level.b.
Design: Intercept+AGEGROUPc.
MANOVA (GLM-multivariate)Tests of Between-Subjects Effects
2820.250a 2 1410.125 1.311 .303334.204b 2 167.102 .176 .840
617839.218 1 617839.218 574.303 .000305882.549 1 305882.549 322.877 .000
2820.250 2 1410.125 1.311 .303334.204 2 167.102 .176 .840
13985.500 13 1075.80812315.733 13 947.364
646448.000 16319289.000 1616805.750 1512649.937 15
Dependent VariableCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglyceride
SourceCorrected Model
Intercept
AGEGROUP
Error
Total
Corrected Total
Type III Sumof Squares df Mean Square F Sig.
R Squared = .168 (Adjusted R Squared = .040)a.
R Squared = .026 (Adjusted R Squared = -.123)b.
MANOVA (GLM-multivariate)
Hubungan gender dengan kadar trigliserida dan kolesterol darah:
Buka SPSS: file – data –dietstudy Analyze – General Linear Model – Multivariate:
Dependent variables: masukkan variabel kolesterol dan trigliserida darah
Fixed factor: masukkan variabel gender Options - Display: aktifkan Homogenity test Continue dan OK
MANOVA (GLM-multivariate)
Box's Test of Equality of Covariance Matricesa
.864
.1116
3329.708.995
Box's MFdf1df2Sig.
Tests the null hypothesis that the observed covariancematrices of the dependent variables are equal across groups.
Design: Intercept+AGEGROUPa.
MANOVA (GLM-multivariate)
Levene's Test of Equality of Error Variancesa
1.521 1 14 .238.630 1 14 .440
CholesterolTriglyceride
F df1 df2 Sig.
Tests the null hypothesis that the error variance of the dependentvariable is equal across groups.
Design: Intercept+GENDERa.
MANOVA (GLM-multivariate)
Multivariate Testsb
.996 1564.542a 2.000 13.000 .000
.004 1564.542a 2.000 13.000 .000240.699 1564.542a 2.000 13.000 .000240.699 1564.542a 2.000 13.000 .000
.818 29.256a 2.000 13.000 .000
.182 29.256a 2.000 13.000 .0004.501 29.256a 2.000 13.000 .0004.501 29.256a 2.000 13.000 .000
Pillai's TraceWilks' LambdaHotelling's TraceRoy's Largest RootPillai's TraceWilks' LambdaHotelling's TraceRoy's Largest Root
EffectIntercept
GENDER
Value F Hypothesis df Error df Sig.
Exact statistica.
Design: Intercept+GENDERb.
MANOVA (GLM-multivariate)Tests of Between-Subjects Effects
13274.766a 1 13274.766 52.633 .0001627.937b 1 1627.937 2.068 .172
597334.766 1 597334.766 2368.373 .000296331.437 1 296331.437 376.396 .00013274.766 1 13274.766 52.633 .0001627.937 1 1627.937 2.068 .1723530.984 14 252.213
11022.000 14 787.286646448.000 16319289.000 1616805.750 1512649.937 15
Dependent VariableCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglyceride
SourceCorrected Model
Intercept
GENDER
Error
Total
Corrected Total
Type III Sumof Squares df Mean Square F Sig.
R Squared = .790 (Adjusted R Squared = .775)a.
R Squared = .129 (Adjusted R Squared = .066)b.
MANOVA (Factorial design)
Variabel dependen lebih dari 1 dan kelompok lebih dari 1, semisal:
Bagaimana rata-rata kadar trigliserida dan kolesterol (2 variabel dependen) berbeda secara bermakna untuk tiap kelompok usia dan jenis kelamin (2 kelompok)
MANOVA (Factorial design)
Buka SPSS: file – data –dietstudy Analyze – general linear model –
multivariate:Dependent variable: TG0 dan WTG0Fixed factor: gender dan AGEGROUPOption – homogeneity testContinue- OK
MANOVA (Factorial design)
Box's Test of Equality of Covariance Matricesa
3.502.339
6463.698
.916
Box's MFdf1df2Sig.
Tests the null hypothesis that the observed covariancematrices of the dependent variables are equal across groups.
Design: Intercept+GENDER+AGEGROUP+GENDER *AGEGROUP
a.
MANOVA (Factorial design)
Levene's Test of Equality of Error Variancesa
1.852 5 10 .1901.091 5 10 .422
CholesterolTriglyceride
F df1 df2 Sig.
Tests the null hypothesis that the error variance of the dependentvariable is equal across groups.
Design: Intercept+GENDER+AGEGROUP+GENDER *AGEGROUP
a.
MANOVA (Factorial design)Multivariate Testsc
.996 1010.769a 2.000 9.000 .000
.004 1010.769a 2.000 9.000 .000224.615 1010.769a 2.000 9.000 .000224.615 1010.769a 2.000 9.000 .000
.826 21.364a 2.000 9.000 .000
.174 21.364a 2.000 9.000 .0004.748 21.364a 2.000 9.000 .0004.748 21.364a 2.000 9.000 .000.164 .448 4.000 20.000 .773.837 .420a 4.000 18.000 .792.194 .388 4.000 16.000 .814.188 .938b 2.000 10.000 .423.199 .551 4.000 20.000 .701.808 .506a 4.000 18.000 .732.230 .459 4.000 16.000 .765.187 .933b 2.000 10.000 .425
Pillai's TraceWilks' LambdaHotelling's TraceRoy's Largest RootPillai's TraceWilks' LambdaHotelling's TraceRoy's Largest RootPillai's TraceWilks' LambdaHotelling's TraceRoy's Largest RootPillai's TraceWilks' LambdaHotelling's TraceRoy's Largest Root
EffectIntercept
GENDER
AGEGROUP
GENDER * AGEGROUP
Value F Hypothesis df Error df Sig.
Exact statistica.
The statistic is an upper bound on F that yields a lower bound on the significance level.b.
Design: Intercept+GENDER+AGEGROUP+GENDER * AGEGROUPc.
MANOVA (Factorial design)Tests of Between-Subjects Effects
13794.217a 5 2758.843 9.161 .0023458.471b 5 691.694 .753 .603
462728.955 1 462728.955 1536.523 .000234481.867 1 234481.867 255.108 .00010936.955 1 10936.955 36.317 .0002734.239 1 2734.239 2.975 .115
37.922 2 18.961 .063 .9391426.678 2 713.339 .776 .486491.654 2 245.827 .816 .469430.109 2 215.054 .234 .796
3011.533 10 301.1539191.467 10 919.147
646448.000 16319289.000 1616805.750 1512649.937 15
Dependent VariableCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglycerideCholesterolTriglyceride
SourceCorrected Model
Intercept
GENDER
AGEGROUP
GENDER * AGEGROUP
Error
Total
Corrected Total
Type III Sumof Squares df Mean Square F Sig.
R Squared = .821 (Adjusted R Squared = .731)a.
R Squared = .273 (Adjusted R Squared = -.090)b.
MANOVA (Factorial design)
Estimated Marginal Means of Cholesterol
age grouping
>6050-60<50
Estim
ated
Mar
gina
l Mea
ns
240
220
200
180
160
140
Gender
Male
Female
Estimated Marginal Means of Triglyceride
age grouping
>6050-60<50
Estim
ated
Mar
gina
l Mea
ns
180
160
140
120
100
Gender
Male
Female
PENGUKURAN BERULANG
Mengetahui apakah ada perbedaan yang bermakna pada suatu variabel yang diukur secara berulang
PENGUKURAN BERULANG Buka SPSS: file – data –dietstudy Analyze – General Linear Model – Repeated Measures:
Within subject factor name: ketik kolest Number of levels: ketik 5 (wgt0 – wgt4) Klik: Add – define:
Within subject factor: pindahkan wgt0 s/d wgt4 Between subject factor: pindahkan gender Klik: plot:
Horizontal axis: gender Separate line: kolest
Add dan Continue OK
PENGUKURAN BERULANG
Within-Subjects Factors
Measure: MEASURE_1
WGT0WGT1WGT2WGT3WGT4
KOLES12345
DependentVariable
Between-Subjects Factors
Male 9Female 7
01
GenderValue Label N
PENGUKURAN BERULANG
Multivariate Testsb
.908 27.123a 4.000 11.000 .000
.092 27.123a 4.000 11.000 .0009.863 27.123a 4.000 11.000 .0009.863 27.123a 4.000 11.000 .000.139 .444a 4.000 11.000 .775.861 .444a 4.000 11.000 .775.162 .444a 4.000 11.000 .775.162 .444a 4.000 11.000 .775
Pillai's TraceWilks' LambdaHotelling's TraceRoy's Largest RootPillai's TraceWilks' LambdaHotelling's TraceRoy's Largest Root
EffectKOLES
KOLES * GENDER
Value F Hypothesis df Error df Sig.
Exact statistica.
Design: Intercept+GENDER Within Subjects Design: KOLES
b.
PENGUKURAN BERULANG
Mauchly's Test of Sphericityb
Measure: MEASURE_1
.399 11.423 9 .252 .763 1.000 .250Within Subjects EffectKOLES
Mauchly's WApprox.
Chi-Square df Sig.Greenhouse-Geisser Huynh-Feldt Lower-bound
Epsilona
Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed dependent variables isproportional to an identity matrix.
May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in theTests of Within-Subjects Effects table.
a.
Design: Intercept+GENDER Within Subjects Design: KOLES
b.
PENGUKURAN BERULANG
Tests of Within-Subjects Effects
Measure: MEASURE_1
639.892 4 159.973 57.534 .000639.892 3.052 209.668 57.534 .000639.892 4.000 159.973 57.534 .000639.892 1.000 639.892 57.534 .000
2.142 4 .536 .193 .9412.142 3.052 .702 .193 .9042.142 4.000 .536 .193 .9412.142 1.000 2.142 .193 .667
155.708 56 2.780155.708 42.727 3.644155.708 56.000 2.780155.708 14.000 11.122
Sphericity AssumedGreenhouse-GeisserHuynh-FeldtLower-boundSphericity AssumedGreenhouse-GeisserHuynh-FeldtLower-boundSphericity AssumedGreenhouse-GeisserHuynh-FeldtLower-bound
SourceKOLES
KOLES * GENDER
Error(KOLES)
Type III Sumof Squares df Mean Square F Sig.
PENGUKURAN BERULANGTests of Within-Subjects Contrasts
Measure: MEASURE_1
639.032 1 639.032 133.643 .000.737 1 .737 .479 .500
9.921E-05 1 9.921E-05 .000 .996.123 1 .123 .089 .770.032 1 .032 .007 .936.309 1 .309 .201 .661.500 1 .500 .146 .708
1.301 1 1.301 .947 .34766.943 14 4.78221.531 14 1.53847.994 14 3.42819.241 14 1.374
KOLESLinearQuadraticCubicOrder 4LinearQuadraticCubicOrder 4LinearQuadraticCubicOrder 4
SourceKOLES
KOLES * GENDER
Error(KOLES)
Type III Sumof Squares df Mean Square F Sig.
PENGUKURAN BERULANG
Tests of Between-Subjects Effects
Measure: MEASURE_1Transformed Variable: Average
2860620.105 1 2860620.105 2162.867 .00066062.105 1 66062.105 49.948 .00018516.483 14 1322.606
SourceInterceptGENDERError
Type III Sumof Squares df Mean Square F Sig.
PENGUKURAN BERULANGEstimated Marginal Means of MEASURE_1
KOLES
54321
Estim
ated
Mar
gina
l Mea
ns240
220
200
180
160
140
Gender
Male
Female
REGRESI BERGANDA
Memprediksi besar variabel dependen dengan menggunakan data variabel bebas yang sudah diketahui besarnya
REGRESI BERGANDA
Analyze – regression – linear:Dependent : WGT4 Independent(s): WGT0, TG0, AGECase labels: genderMethod: enterOK
REGRESI BERGANDA
Variables Entered/Removedb
Cholesterol, Age inyears,Triglyceride
a
. Enter
Model1
VariablesEntered
VariablesRemoved Method
All requested variables entered.a.
Dependent Variable: Final cholesterolb.
REGRESI BERGANDA
Model Summary
.997a .994 .992 2.953Model1
R R SquareAdjustedR Square
Std. Error ofthe Estimate
Predictors: (Constant), Cholesterol, Age in years,Triglyceride
a.
REGRESI BERGANDA
ANOVAb
16736.790 3 5578.930 639.737 .000a
104.648 12 8.72116841.438 15
RegressionResidualTotal
Model1
Sum ofSquares df Mean Square F Sig.
Predictors: (Constant), Cholesterol, Age in years, Triglyceridea.
Dependent Variable: Final cholesterolb.
REGRESI BERGANDACoefficientsa
3.375 8.574 .394 .701-.164 .111 -.034 -1.477 .165-.010 .027 -.009 -.373 .716.995 .024 .994 42.243 .000
(Constant)Age in yearsTriglycerideCholesterol
Model1
B Std. Error
UnstandardizedCoefficients
Beta
StandardizedCoefficients
t Sig.
Dependent Variable: Final cholesterola.
Persamaan regresi:
Kadar kolesterol akhir = 3,375 – 0,164 usia – 0,10 kadar trigliserida awal + 0,995 kadar kolesterol awal
REGRESI BERGANDA
Residuals Statisticsa
142.66 249.33 190.31 33.403 16-5.05 4.88 .00 2.641 16
-1.426 1.767 .000 1.000 16-1.712 1.652 .000 .894 16
Predicted ValueResidualStd. Predicted ValueStd. Residual
Minimum Maximum Mean Std. Deviation N
Dependent Variable: Final cholesterola.
REGRESI BERGANDAVariables Entered/Removedb
Cholesterola . Enter
Model1
VariablesEntered
VariablesRemoved Method
All requested variables entered.a.
Dependent Variable: Final cholesterolb.
Model Summaryb
.996a .993 .992 2.986Model1
R R SquareAdjustedR Square
Std. Error ofthe Estimate
Predictors: (Constant), Cholesterola.
Dependent Variable: Final cholesterolb.
REGRESI BERGANDA
ANOVAb
16716.618 1 16716.618 1874.976 .000a
124.819 14 8.91616841.438 15
RegressionResidualTotal
Model1
Sum ofSquares df Mean Square F Sig.
Predictors: (Constant), Cholesterola.
Dependent Variable: Final cholesterolb.
REGRESI BERGANDACoefficientsa
-7.536 4.630 -1.628 .126.997 .023 .996 43.301 .000
(Constant)Cholesterol
Model1
B Std. Error
UnstandardizedCoefficients
Beta
StandardizedCoefficients
t Sig.
Dependent Variable: Final cholesterola.
Persamaan regresi:
Kadar kolesterol akhir = -7,536 + 0,997 kadar kolesterol awal
Correlations
1 .996**. .000
16 16.996** 1.000 .
16 16
Pearson CorrelationSig. (2-tailed)NPearson CorrelationSig. (2-tailed)N
Cholesterol
Final cholesterol
CholesterolFinal
cholesterol
Correlation is significant at the 0.01 level (2-tailed).**.
Linear Regression
150 175 200 225 250
Cholesterol
150
175
200
225
250
Fina
l cho
lest
erol
Final cholesterol = -7.54 + 1.00 * w gt0R-Square = 0.99
Uji regresi logistik binari
Ingin memprediksi variabel dependen yang berskala binari (ya=1 dan tidak=0) dengan menggunakan data variabel independen yang sudah diketahui besarnya
Uji regresi logistik binari
Buka SPSS: file – data –dietstudy Analyze – Regression – Binary logistic:
Dependent: cholst0 (status kadar kolesterol awal, 1=tinggi, 0=normal)
Covariates: age dan TG0Options: Homer-Lemeshow goodness of fitOK
Uji regresi logistik binari
Case Processing Summary
16 100.00 .0
16 100.00 .0
16 100.0
Unweighted Cases a
Included in AnalysisMissing CasesTotal
Selected Cases
Unselected CasesTotal
N Percent
If weight is in effect, see classification table for the totalnumber of cases.
a.
Uji regresi logistik binari
Omnibus Tests of Model Coefficients
1.902 2 .3861.902 2 .3861.902 2 .386
StepBlockModel
Step 1Chi-square df Sig.
Uji regresi logistik binariModel Summary
20.028 .112 .150Step1
-2 Loglikelihood
Cox & SnellR Square
NagelkerkeR Square
Hosmer and Lemeshow Test
9.129 6 .166Step1
Chi-square df Sig.
Uji regresi logistik binari
Contingency Table for Hosmer and Lemeshow Test
2 1.570 0 .430 22 1.516 0 .484 21 1.453 1 .547 20 1.118 2 .882 20 1.024 2 .976 22 .920 0 1.080 21 .787 1 1.213 21 .613 1 1.387 2
12345678
Step1
Observed Expected
cholesterol status =normal
Observed Expected
cholesterol status =high
Total
Uji regresi logistik binari
Classification Tablea
5 4 55.64 3 42.9
50.0
Observednormalhigh
cholesterol status
Overall Percentage
Step 1normal highcholesterol status Percentage
Correct
Predicted
The cut value is .500a.
Uji regresi logistik binari
Variables in the Equation
.042 .079 .277 1 .598 1.043
.025 .020 1.527 1 .217 1.025-5.970 5.416 1.215 1 .270 .003
AGETG0Constant
Step1
a
B S.E. Wald df Sig. Exp(B)
Variable(s) entered on step 1: AGE, TG0.a.
Penafsiran dan prediksi:
Kadar kolesterol tinggi = -5,970 + 0,42 usia + 0,025 kadar trigliserida
REGRESI BERGANDA – variabel dummy Memprediksi besar variabel dependen
dengan menggunakan data variabel bebas dimana satu atau lebih variabel bebas adalah variabel dummy (ya=1 dan tidak=0)
REGRESI BERGANDA – variabel dummy Buka SPSS: file – data –dietstudy Analyze – Regression – Linier:
Dependent: wgt4 Independent: gender, cholst0, trigst0Methods: enterOK
REGRESI BERGANDA – variabel dummy
Variables Entered/Removedb
triglyceridestatus,cholesterolstatus,Gender
a
. Enter
Model1
VariablesEntered
VariablesRemoved Method
All requested variables entered.a.
Dependent Variable: Final cholesterolb.
REGRESI BERGANDA – variabel dummy
Model Summary
.932a .869 .836 13.551Model1
R R SquareAdjustedR Square
Std. Error ofthe Estimate
Predictors: (Constant), triglyceride status, cholesterolstatus, Gender
a.
REGRESI BERGANDA – variabel dummy
ANOVAb
14637.729 3 4879.243 26.569 .000a
2203.709 12 183.64216841.438 15
RegressionResidualTotal
Model1
Sum ofSquares df Mean Square F Sig.
Predictors: (Constant), triglyceride status, cholesterol status, Gendera.
Dependent Variable: Final cholesterolb.
REGRESI BERGANDA – variabel dummy
Coefficientsa
194.020 10.219 18.987 .000-35.437 10.971 -.542 -3.230 .00730.003 10.877 .459 2.758 .017-3.039 7.100 -.046 -.428 .676
(Constant)Gendercholesterol statustriglyceride status
Model1
B Std. Error
UnstandardizedCoefficients
Beta
StandardizedCoefficients
t Sig.
Dependent Variable: Final cholesterola.
REGRESI BERGANDA – variabel dummy
Coefficientsa
192.500 9.276 20.752 .000-34.786 10.518 -.532 -3.307 .00629.786 10.518 .455 2.832 .014
(Constant)Gendercholesterol status
Model1
B Std. Error
UnstandardizedCoefficients
Beta
StandardizedCoefficients
t Sig.
Dependent Variable: Final cholesterola.
Penafsiran dan prediksi:
Kadar kolesterol akhir = 192,500 – 34,786 gender + 29,786 kadar kolesterol awal
Uji analisis faktor
Ingin diketahui kadar kolesteol yang mana yang menentukan kadar kolesterol akhir dari beberapa data kadar kolesterol yang ada?
Uji analisis faktor
Buka SPSS: file – data –dietstudy Analyze – Data reduction - Factor:
Wgt0 s/d wgt4Descriptives:
Correlation matrix: KMO and Bartletts’s test of spherity Anti image
Continue dan OK
Uji analisis faktor
KMO and Bartlett's Test
.894
294.19810
.000
Kaiser-Meyer-Olkin Measure of SamplingAdequacy.
Approx. Chi-SquaredfSig.
Bartlett's Test ofSphericity
Uji analisis faktor
Anti-image Matrices
.003 .000 -.001 -.001 .001
.000 .003 -.001 .000 -1.847E-05-.001 -.001 .001 .000 -.001-.001 .000 .000 .003 -.001.001 -1.847E-05 -.001 -.001 .003.899a -.092 -.536 -.248 .301
-.092 .935a -.516 .065 -.007-.536 -.516 .820a -.246 -.512-.248 .065 -.246 .937a -.378.301 -.007 -.512 -.378 .889a
Cholesterol1st interim cholesterol2nd interim cholesterol3rd interim cholesterolFinal cholesterolCholesterol1st interim cholesterol2nd interim cholesterol3rd interim cholesterolFinal cholesterol
Anti-image Covariance
Anti-image Correlation
Cholesterol1st interimcholesterol
2nd interimcholesterol
3rd interimcholesterol
Finalcholesterol
Measures of Sampling Adequacy(MSA)a.
Uji analisis faktor
Communalities
1.000 .9981.000 .9981.000 .9991.000 .9981.000 .998
Cholesterol1st interim cholesterol2nd interim cholesterol3rd interim cholesterolFinal cholesterol
Initial Extraction
Extraction Method: Principal Component Analysis.
Uji analisis faktor
Total Variance Explained
4.991 99.816 99.816 4.991 99.816 99.816.004 .080 99.896.003 .059 99.954.002 .033 99.988.001 .012 100.000
Component12345
Total % of Variance Cumulative % Total % of Variance Cumulative %Initial Eigenvalues Extraction Sums of Squared Loadings
Extraction Method: Principal Component Analysis.
Uji analisis faktor
Component Matrixa
.999
.9991.000.999.999
Cholesterol1st interim cholesterol2nd interim cholesterol3rd interim cholesterolFinal cholesterol
1
Component
Extraction Method: Principal Component Analysis.1 components extracted.a.
ANALISIS DISKRIMINAN
Ingin membuat model yang bisa secara jelas menunjukkan perbedaan antar isi variabel dependen, misal:Kadar kolesterol dan trigliserida pada
kelompok laki-laki (=0) dan perempuan (=1)
ANALISIS DISKRIMINAN
Buka SPSS: file – data –dietstudy Analyze – Clasify - Discriminant:
Grouping variable: gender Define range: 0 dan 1 Independent: age, wgt0, tg0, wgt4 dan tg4 Statistics:
Descriptives: Means Function coefficients: Fisher’s dam Unstandardized
ANALISIS DISKRIMINAN
Use stepwise method Method: Mahalanobis distance Criteria: use probability of F
Clasify: Display: Casewise results, Leave-one-out-
classificationContinue dan OK
ANALISIS DISKRIMINANGroup Statistics
54.00 7.036 9 9.000147.33 26.847 9 9.000223.78 18.754 9 9.000117.11 28.790 9 9.000215.67 18.076 9 9.00055.57 7.208 7 7.000
127.00 29.597 7 7.000165.71 10.935 7 7.000133.71 29.607 7 7.000157.71 12.932 7 7.00054.69 6.916 16 16.000
138.44 29.040 16 16.000198.38 33.472 16 16.000124.38 29.412 16 16.000190.31 33.508 16 16.000
Age in yearsTriglycerideCholesterolFinal triglycerideFinal cholesterolAge in yearsTriglycerideCholesterolFinal triglycerideFinal cholesterolAge in yearsTriglycerideCholesterolFinal triglycerideFinal cholesterol
GenderMale
Female
Total
Mean Std. Deviation Unweighted WeightedValid N (listwise)
ANALISIS DISKRIMINAN
Variables Entered/Removeda,b,c,d
Cholesterol 13.367 Male and
Female 52.633 1 14.000 4.192E-06
Step1
Entered StatisticBetweenGroups Statistic df1 df2 Sig.
Exact F
Min. D Squared
At each step, the variable that maximizes the Mahalanobis distance between the two closestgroups is entered.
Maximum number of steps is 10.a.
Maximum significance of F to enter is .05.b.
Minimum significance of F to remove is .10.c.
F level, tolerance, or VIN insufficient for further computation.d.
ANALISIS DISKRIMINAN
Variables in the Analysis
1.000 .000CholesterolStep1
ToleranceSig. of F to
Remove
ANALISIS DISKRIMINANVariables Not in the Analysis
1.000 1.000 .668 .049 Male andFemale
1.000 1.000 .172 .525 Male andFemale
1.000 1.000 .000 13.367 Male andFemale
1.000 1.000 .277 .325 Male andFemale
1.000 1.000 .000 12.998 Male andFemale
.995 .995 .977 13.368 Male andFemale
.945 .945 .178 16.003 Male andFemale
.957 .957 .869 13.404 Male andFemale
.034 .034 .953 13.372 Male andFemale
Age in years
Triglyceride
Cholesterol
Final triglyceride
Final cholesterol
Age in years
Triglyceride
Final triglyceride
Final cholesterol
Step0
1
ToleranceMin.
ToleranceSig. of Fto Enter
Min. DSquared
BetweenGroups
ANALISIS DISKRIMINANWilks' Lambda
1 .210 1 1 14 52.633 1 14.000 .000Step1
Number ofVariables Lambda df1 df2 df3 Statistic df1 df2 Sig.
Exact F
Eigenvalues
3.760a 100.0 100.0 .889Function1
Eigenvalue % of Variance Cumulative %CanonicalCorrelation
First 1 canonical discriminant functions were used in theanalysis.
a.
Wilks' Lambda
.210 21.062 1 .000Test of Function(s)1
Wilks'Lambda Chi-square df Sig.
ANALISIS DISKRIMINAN
Structure Matrix
1.000.983
-.234-.207-.069
CholesterolFinal cholesterola
Triglyceridea
Final triglyceridea
Age in yearsa
1Function
Pooled within-groups correlations between discriminatingvariables and standardized canonical discriminant functions Variables ordered by absolute size of correlation within function.
This variable not used in the analysis.a.
ANALISIS DISKRIMINANCanonical Discriminant Function Coefficients
.063-12.491
Cholesterol(Constant)
1Function
Unstandardized coefficients
Skor Z = -12,491 + 0,063 kadar kolesterol awal
ANALISIS DISKRIMINAN
Functions at Group Centroids
1.600-2.057
GenderMaleFemale
1Function
Unstandardized canonical discriminantfunctions evaluated at group means
ANALISIS DISKRIMINAN
Prior Probabilities for Groups
.500 9 9.000
.500 7 7.0001.000 16 16.000
GenderMaleFemaleTotal
Prior Unweighted WeightedCases Used in Analysis
ANALISIS DISKRIMINANClassification Function Coefficients
.887 .657-99.967 -55.134
Cholesterol(Constant)
Male FemaleGender
Fisher's linear discriminant functions
Skor kolesterol pada laki-laki = -99,967 + 0,887 kolesterol awal
Skor kolesterol pada perempuan = -55,134 + 0,657 kolesterol awal
Selisih antar keduanya (skor Z) = -44,833 + 0,23 kolesterol awal
Skor Z sebelumnya = -12,491 + 0,063 kolesterol awal
ANALISIS DISKRIMINANCasewise Statistics
0 0 .105 1 .679 2.635 1 .321 4.133 -.0240 0 .405 1 1.000 .693 1 .000 20.148 2.4320 0 .561 1 1.000 .337 1 .000 17.951 2.1801 1 .403 1 .974 .700 0 .026 7.950 -1.2200 0 .764 1 .996 .091 1 .004 11.258 1.2991 1 .836 1 .997 .043 0 .003 11.897 -1.8500 0 .911 1 .998 .013 1 .002 12.561 1.4881 1 .935 1 .998 .007 0 .002 12.782 -1.9760 0 .119 1 .727 2.434 1 .273 4.393 .0390 0 .561 1 1.000 .337 1 .000 17.951 2.1801 1 .403 1 .974 .700 0 .026 7.950 -1.2201 1 .627 1 1.000 .236 0 .000 17.155 -2.5421 1 .583 1 1.000 .301 0 .000 17.681 -2.6050 0 .624 1 .993 .240 1 .007 10.026 1.1100 0 .036 1 1.000 4.376 1 .000 33.040 3.6911 1 .354 1 1.000 .858 0 .000 21.001 -2.9830 0 .047 1 .615 3.928 2 .385 4.8680 0 .353 1 1.000 .863 2 .000 19.8130 0 .523 1 1.000 .407 2 .000 17.1331 1 .332 1 .969 .939 1 .031 7.8390 0 .743 1 .995 .107 2 .005 10.5301 1 .816 1 .996 .054 1 .004 11.0870 0 .903 1 .997 .015 2 .003 11.6761 1 .927 1 .997 .008 1 .003 11.8750 0 .059 1 .681 3.556 2 .319 5.0710 0 .523 1 1.000 .407 2 .000 17.1331 1 .332 1 .969 .939 1 .031 7.8391 1 .581 1 1.000 .304 1 .000 16.2491 1 .532 1 1.000 .390 1 .000 16.8400 0 .592 1 .990 .287 2 .010 9.4930 0 .005 1 1.000 7.932 2 .000 47.3201 1 .280 1 1.000 1.169 1 .000 21.003
Case Number1234567891011121314151612345678910111213141516
Original
Cross-validated a
Actual GroupPredicted
Group p dfP(D>d | G=g)
P(G=g | D=d)
SquaredMahalanobisDistance to
Centroid
Highest Group
Group P(G=g | D=d)
SquaredMahalanobisDistance to
Centroid
Second Highest Group
Function 1
DiscriminantScores
For the original data, squared Mahalanobis distance is based on canonical functions.For the cross-validated data, squared Mahalanobis distance is based on observations.
Cross validation is done only for those cases in the analysis. In cross validation, each case is classified by the functions derived from all cases other than that case.a.
ANALISIS DISKRIMINANClassification Resultsb,c
9 0 90 7 7
100.0 .0 100.0.0 100.0 100.09 0 90 7 7
100.0 .0 100.0.0 100.0 100.0
GenderMaleFemaleMaleFemaleMaleFemaleMaleFemale
Count
%
Count
%
Original
Cross-validated a
Male Female
Predicted GroupMembership
Total
Cross validation is done only for those cases in the analysis. Incross validation, each case is classified by the functions derivedfrom all cases other than that case.
a.
100.0% of original grouped cases correctly classified.b.
100.0% of cross-validated grouped cases correctly classified.c.
ANALISIS DISKRIMINAN Kesimpulan:
Analisis Wilk’s Lambda (sig <0.001) Variable in analysis (Variabel yang membedakan
gender laki-laki dan perempuan adalah kadar kolesterol awal)
Model diskriminannya:Skor Z = -12,491 + 0,063 kadar kolesterol awal Model di atas mempunyai ketepatan
mengklasifikasikan gender sebesar 100% (ketepatan sangat tinggi), dan model dapat digunakan untuk mengklasifikasikan gender dari data kolesterol awal