ANALISIS DATA EPIDEMIOLOGI
MENORMALKAN DATA DAN ANALISIS REGRESI LOGISTIK
Nama : Meldi Nur Anisa
NIM : N1A118078
Dosen Pengampu: Dr. Ummi Kalsum, S.KM., M.KM.
PROGRAM STUDI ILMU KESEHATAN MASYARAKAT
FAKULTAS KEDOKTERAN DAN ILMU KESEHATAN
UNIVERSITAS JAMBI
2021
Data Usair
Interpretasi
. use "C:\Users\BC5720\Downloads\USAIR.DTA"
. des
Contains data from C:\Users\BC5720\Downloads\USAIR.DTA
obs: 41
vars: 8 28 Apr 2012 10:34
storage display value
variable name type format label variable label
town str9 %9s
so2 int %8.0g
temp float %9.0g
manuf int %8.0g
pop int %8.0g
wind float %9.0g
precip float %9.0g
days int %8.0g
Sorted by:
Variabel independen (temperatur, jumlah pabrik, jumlah populasi, angin, precip, dan hari.
Variabel dependen (populasi udara). Dengan total 8 variabel.
. sum
Variable Obs Mean Std. Dev. Min Max
town 0
so2 41 30.04878 23.47227 8 110
temp 41 55.76341 7.227716 43.5 75.5
manuf 41 463.0976 563.4739 35 3344
pop 41 608.6098 579.113 71 3369
wind 41 9.443902 1.428644 6 12.7
precip 41 36.76902 11.77155 7.05 59.8
days 41 113.9024 26.50642 36 166
. list
town so2 temp manuf pop wind precip days
1. Phoenix 10 70.3 213 582 6 7.05 36
2. Lrock 13 61 91 132 8.2 48.52 100
3. Sfran 12 56.7 453 716 8.7 20.66 67
4. Denver 17 51.9 454 515 9 12.95 86
5. Hartford 56 49.1 412 158 9 43.37 127
6. Wilming 36 54 80 80 9 40.25 114
7. Washing 29 57.3 434 757 9.3 38.89 111
8. Jackson 14 68.4 136 529 8.8 54.47 116
9. Miami 10 75.5 207 335 9 59.8 128
10. Atlanta 24 61.5 368 497 9.1 48.34 115
11. Chicago 110 50.6 3344 3369 10.4 34.44 122
12. Indian 28 52.3 361 746 9.7 38.74 121
13. DesM 17 49 104 201 11.2 30.85 103
14. Wichita 8 56.6 125 277 12.7 30.58 82
15. Louisv 30 55.6 291 593 8.3 43.11 123
16. NewO 9 68.3 204 361 8.4 56.77 113
17. Baltim 47 55 625 905 9.6 41.31 111
18. Detroit 35 49.9 1064 1513 10.1 30.96 129
19. Minn 29 43.5 699 744 10.6 25.94 137
20. Kansas 14 54.5 381 507 10 37 99
21. StLouis 56 55.9 775 622 9.5 35.89 105
22. Omaha 14 51.5 181 347 10.9 30.18 98
23. Alburq 11 56.8 46 244 8.9 7.77 58
24. Albany 46 47.6 44 116 8.8 33.36 135
25. Buffalo 11 47.1 391 463 12.4 36.11 166
26. Cincinn 23 54 462 453 7.1 39.04 132
27. Cleve 65 49.7 1007 751 10.9 34.99 155
28. Colum 26 51.5 266 540 8.6 37.01 134
29. Philad 69 54.6 1692 1950 9.6 39.93 115
30. Pittsb 61 50.4 347 520 9.4 36.22 147
31. Provid 94 50 343 179 10.6 42.75 125
32. Memphis 10 61.6 337 624 9.2 49.1 105
33. Nashville 18 59.4 275 448 7.9 46 119
34. Dallas 9 66.2 641 844 10.9 35.94 78
35. Houston 10 68.9 721 1233 10.8 48.19 103
36. SLC 28 51 137 176 8.7 15.17 89
37. Norfolk 31 59.3 96 308 10.6 44.68 116
38. Richmond 26 57.8 197 299 7.6 42.59 115
39. Seattle 29 51.1 379 531 9.4 38.79 164
40. Charlest 31 55.2 35 71 6.5 40.75 148
41. Milwak 16 45.7 569 717 11.8 29.07 123
Objek penelitian terdiri dari 41 kota.
. ladder so2
Transformation formula chi2(2) P(chi2)
cubic so2^3 43.63 0.000
square so2^2 33.57 0.000
identity so2 16.19 0.000
square root sqrt(so2) 6.53 0.038
log log(so2) 2.82 0.244
1/(square root) 1/sqrt(so2) 6.27 0.043
inverse 1/so2 5.44 0.066
1/square 1/(so2^2) 7.72 0.021
1/cubic 1/(so2^3) 14.99 0.001
. g log_so2= log(so2)
. ladder log_so2
Transformation formula chi2(2) P(chi2)
cubic log_so2^3 7.67 0.022
square log_so2^2 3.84 0.146
identity log_so2 2.82 0.244
square root sqrt(log_so2) 3.90 0.142
log log(log_so2) 4.91 0.086
1/(square root) 1/sqrt(log_so2) 5.52 0.063
inverse 1/log_so2 5.53 0.063
1/square 1/(log_so2^2) 4.75 0.093
1/cubic 1/(log_so2^3) 4.99 0.083
Setelah di analisis kita dapat melihat kenormalannya. Diketahui variabel polusi udara tidak normal. Maka
di lakukan penormalan dengan menglogin kan data data menjadi normal. Sebelumnya p= 0,000 menjadi
p= 0,2444
. ladder temp
Transformation formula chi2(2) P(chi2)
cubic temp^3 12.45 0.002
square temp^2 8.86 0.012
identity temp 5.67 0.059
square root sqrt(temp) 4.29 0.117
log log(temp) 2.87 0.238
1/(square root) 1/sqrt(temp) 1.61 0.446
inverse 1/temp 0.71 0.703
1/square 1/(temp^2) 0.00 0.998
1/cubic 1/(temp^3) 1.16 0.560
P=0,059 (normal)
. ladder manuf
Transformation formula chi2(2) P(chi2)
cubic manuf^3 61.00 0.000
square manuf^2 57.71 0.000
identity manuf 41.74 0.000
square root sqrt(manuf) 19.73 0.000
log log(manuf) 0.66 0.718
1/(square root) 1/sqrt(manuf) 12.70 0.002
inverse 1/manuf 24.90 0.000
1/square 1/(manuf^2) 38.26 0.000
1/cubic 1/(manuf^3) 45.25 0.000
. g log_manuf = log(manuf)
. ladder log_manuf
Transformation formula chi2(2) P(chi2)
cubic log_ma~f^3 10.40 0.006
square log_ma~f^2 3.62 0.164
identity log_ma~f 0.66 0.718
square root sqrt(log_ma~f) 1.93 0.381
log log(log_ma~f) 4.34 0.114
1/(square root) 1/sqrt(log_ma~f) 6.83 0.033
inverse 1/log_ma~f 9.65 0.008
1/square 1/(log_ma~f^2) 15.62 0.000
1/cubic 1/(log_ma~f^3) 21.29 0.000
Variabel jumlah pabrik awalnya tidak normal deng p=0.000, kemudian data di normalkan dengan meng-
log-an data. Kemudian data menjadi p=0,728 (normal).
. ladder pop
Transformation formula chi2(2) P(chi2)
cubic pop^3 59.51 0.000
square pop^2 54.09 0.000
identity pop 35.81 0.000
square root sqrt(pop) 15.82 0.000
log log(pop) 1.13 0.569
1/(square root) 1/sqrt(pop) 11.87 0.003
inverse 1/pop 23.51 0.000
1/square 1/(pop^2) 38.03 0.000
1/cubic 1/(pop^3) 44.70 0.000
. g log_pop = log(pop)
. ladder log_pop
Transformation formula chi2(2) P(chi2)
cubic log_pop^3 7.70 0.021
square log_pop^2 2.66 0.265
identity log_pop 1.13 0.569
square root sqrt(log_pop) 2.24 0.326
log log(log_pop) 4.15 0.126
1/(square root) 1/sqrt(log_pop) 6.06 0.048
inverse 1/log_pop 8.21 0.017
1/square 1/(log_pop^2) 12.90 0.002
1/cubic 1/(log_pop^3) 17.67 0.000
Variabel populasi awalnya tidak normal deng p=0.000, kemudian data di normalkan dengan meng-log-an
data. Kemudian data menjadi p=0,569 (normal).
. ladder wind
Transformation formula chi2(2) P(chi2)
cubic wind^3 7.20 0.027
square wind^2 2.87 0.239
identity wind 0.60 0.742
square root sqrt(wind) 1.65 0.439
log log(wind) 4.31 0.116
1/(square root) 1/sqrt(wind) 7.40 0.025
inverse 1/wind 11.07 0.004
1/square 1/(wind^2) 19.04 0.000
1/cubic 1/(wind^3) 26.56 0.000
P=0,742 (normal)
. ladder precip
Transformation formula chi2(2) P(chi2)
cubic precip^3 11.34 0.003
square precip^2 2.35 0.309
identity precip 5.41 0.067
square root sqrt(precip) 12.98 0.002
log log(precip) 21.58 0.000
1/(square root) 1/sqrt(precip) 29.27 0.000
inverse 1/precip 35.24 0.000
1/square 1/(precip^2) 42.42 0.000
1/cubic 1/(precip^3) 45.80 0.000
P=0,067 (normal)
. ladder days
Transformation formula chi2(2) P(chi2)
cubic days^3 8.02 0.018
square days^2 1.72 0.424
identity days 4.88 0.087
square root sqrt(days) 11.69 0.003
log log(days) 20.75 0.000
1/(square root) 1/sqrt(days) 30.37 0.000
inverse 1/days 39.15 0.000
1/square 1/(days^2) 51.68 0.000
1/cubic 1/(days^3) 57.88 0.000
p=0,087 (normal)
Data Sebaran
. scatter log_so2 temp
Hubungan Antara SO2 dengan temperatur
5
4
log_so2
3
2
40 50 60 70 80
temp
. scatter log_so2 log_manuf
Hubungan antara SO2 dengan jumlah pabrik
5
4
log_so2
3
2
3 4 5 6 7 8
log_manuf
. scatter log_so2 log_pop
Hubungan antara SO2 dengan populasi
5
4
log_so2
3
2
4 5 6 7 8
log_pop
Hubungan antara SO2 dengan angin
5
4
log_so2
3
2
6 8 10 12 14
wind
. scatter log_so2 precip
. scatter log_so2 days
Hubungan antara SO2 dengan precip
5
4
log_so2
3
2
10 20 30 40 50 60
precip
Hubungan antara SO2 dengan hari
5
4
log_so2
3
2
0 50 100 150 200
days
. twoway (lfitci so2 temp) (scatter log_so2 temp)
. twoway (lfitci log_so2 temp) (scatter log_so2 temp)
. twoway (lfitci log_so2 log_manuf) (scatter log_so2 log_manuf)
. twoway (lfitci log_so2 log_pop) (scatter log_so2 log_pop)
. twoway (lfitci log_so2 wind) (scatter log_so2 wind)
. twoway (lfitci log_so2 wind) (scatter log_so2 wind)
. twoway (lfitci log_so2 precip) (scatter log_so2 precip)
. twoway (lfitci log_so2 days) (scatter log_so2 days)
twoway (lfitci so2 temp) (scatter log_so2 temp)
5
4
3
2
1
40 50 60 70 80
temp
95% CI Fitted values
log_so2
twoway (lfitci log_so2 log_manuf) (scatter log_so2 log_manuf)
5
4
3
2
3 4 5 6 7 8
log_manuf
95% CI Fitted values
log_so2
Log_pop
5
4
3
2
4 5 6 7 8
log_pop
95% CI Fitted values
log_so2
Wind
5
4
3
2
6 8 10 12 14
wind
95% CI Fitted values
log_so2
Precip
5
4
3
2
10 20 30 40 50 60
precip
95% CI Fitted values
log_so2
Days
5
4
3
2
1
0 50 100 150 200
days
95% CI Fitted values
log_so2
REGRESI LINIER
INTERPRETASI
. reg log_so2 temp
Source SS df MS Number of obs = 41
F(1, 39) = 16.60
Model 5.89142105 1 5.89142105 Prob > F = 0.0002
Residual 13.8375076 39 .354807887 R-squared = 0.2986
Adj R-squared = 0.2806
Total 19.7289286 40 .493223216 Root MSE = .59566
log_so2 Coef. Std. Err. t P>|t| [95% Conf. Interval]
temp -.0530981 .0130306 -4.07 0.000 -.079455 -.0267412
_cons 6.113935 .7325632 8.35 0.000 4.632186 7.595684
Dari hasil analisis, di temukan bahwa adanya pengaruh temperatur terhadap polusi udara. Variabel
temperatur dapat memprediksi variabel polusi udara sebesar 29,86%. Persamaan garis So2 = 6,114 + ( –
0,053) x temp.
. sum
Variable Obs Mean Std. Dev. Min Max
town 0
so2 41 30.04878 23.47227 8 110
temp 41 55.76341 7.227716 43.5 75.5
manuf 41 463.0976 563.4739 35 3344
pop 41 608.6098 579.113 71 3369
wind 41 9.443902 1.428644 6 12.7
precip 41 36.76902 11.77155 7.05 59.8
days 41 113.9024 26.50642 36 166
log_so2 41 3.153004 .7022985 2.079442 4.70048
log_manuf 41 5.692094 .9634348 3.555348 8.114923
log_pop 41 6.101705 .8044345 4.26268 8.122372
. reg log_so2 log_manuf
Source SS df MS Number of obs = 41
F(1, 39) = 5.04
Model 2.25607411 1 2.25607411 Prob > F = 0.0306
Residual 17.4728545 39 .448021911 R-squared = 0.1144
Adj R-squared = 0.0916
Total 19.7289286 40 .493223216 Root MSE = .66934
log_so2 Coef. Std. Err. t P>|t| [95% Conf. Interval]
log_manuf .2465042 .1098493 2.24 0.031 .024313 .4686954
_cons 1.749878 .6339505 2.76 0.009 .4675924 3.032164
Dari hasil analisis, di temukan bahwa ada pengaruh jumlah pabrik terhadap polusi udara. Variabel jumlah
pabrik dapat memprediksi variabel polusi udara sebesar 11,44%. Persamaan garis So2 = 1,750 + ( 0,247)
x manuf.
. reg log_so2 log_pop
Source SS df MS Number of obs = 41
F(1, 39) = 0.40
Model .199203126 1 .199203126 Prob > F = 0.5319
Residual 19.5297255 39 .500762193 R-squared = 0.0101
Adj R-squared = -0.0153
Total 19.7289286 40 .493223216 Root MSE = .70765
log_so2 Coef. Std. Err. t P>|t| [95% Conf. Interval]
log_pop .0877258 .1390897 0.63 0.532 -.1936097 .3690614
_cons 2.617727 .85585 3.06 0.004 .8866066 4.348847
Dari hasil analisis, di temukan bahwa tidak ada pengaruh jumlah populasi terhadap polusi udara.
Variabel jumlah populasi dapat memprediksi variabel polusi udara sebesar 1,01%. Persamaan garis So2 =
2,612 + ( 0,088) x pop.
. reg log_so2 wind
Source SS df MS Number of obs = 41
F(1, 39) = 0.01
Model .003610137 1 .003610137 Prob > F = 0.9331
Residual 19.7253185 39 .505777398 R-squared = 0.0002
Adj R-squared = -0.0255
Total 19.7289286 40 .493223216 Root MSE = .71118
log_so2 Coef. Std. Err. t P>|t| [95% Conf. Interval]
wind -.0066498 .0787092 -0.08 0.933 -.1658542 .1525546
_cons 3.215804 .7515743 4.28 0.000 1.695601 4.736006
Dari hasil analisis, di temukan bahwa tidak ada pengaruh angin terhadap polusi udara. Variabel
angin dapat memprediksi variabel polusi udara hanya sebesar 0,02%. Persamaan garis So2 = 3,216+ ( -
0,007) x wind.
. reg log_so2 precip
Source SS df MS Number of obs = 41
F(1, 39) = 0.09
Model .04557428 1 .04557428 Prob > F = 0.7654
Residual 19.6833544 39 .504701394 R-squared = 0.0023
Adj R-squared = -0.0233
Total 19.7289286 40 .493223216 Root MSE = .71042
log_so2 Coef. Std. Err. t P>|t| [95% Conf. Interval]
precip .0028675 .0095423 0.30 0.765 -.0164337 .0221686
_cons 3.04757 .3679858 8.28 0.000 2.303249 3.791892
Dari hasil analisis, di temukan bahwa tidak ada pengaruh presipitasi terhadap polusi udara.
Variabel presipitasi dapat memprediksi variabel polusi udara hanya sebesar 0,23%. Persamaan garis So2
= 3,047 + ( 0,003) x precip.
. reg log_so2 days
Source SS df MS Number of obs = 41
F(1, 39) = 11.59
Model 4.51844453 1 4.51844453 Prob > F = 0.0016
Residual 15.2104841 39 .390012413 R-squared = 0.2290
Adj R-squared = 0.2093
Total 19.7289286 40 .493223216 Root MSE = .62451
log_so2 Coef. Std. Err. t P>|t| [95% Conf. Interval]
days .0126798 .0037253 3.40 0.002 .0051448 .0202149
_cons 1.70874 .4353826 3.92 0.000 .8280956 2.589385
Dari hasil analisis, di temukan bahwa ada pengaruh hari terhadap polusi udara. Variabel jumlah
hari dapat memprediksi variabel polusi udara sebesar 22,90%, sisanya diprediksi oleh variabel lain.
Persamaan garis So2 = =1,709 + ( 0,013) x days.
Regresi Linier Multivariat
. reg log_so2 temp log_manuf days
Source SS df MS Number of obs = 41
F(3, 37) = 9.07
Model 8.36229392 3 2.78743131 Prob > F = 0.0001
Residual 11.3666347 37 .307206344 R-squared = 0.4239
Adj R-squared = 0.3771
Total 19.7289286 40 .493223216 Root MSE = .55426
log_so2 Coef. Std. Err. t P>|t| [95% Conf. Interval]
temp -.0382498 .0134931 -2.83 0.007 -.0655894 -.0109103
log_manuf .1700487 .0926046 1.84 0.074 -.017586 .3576834
days .0071896 .0036838 1.95 0.059 -.0002746 .0146538
_cons 3.499097 1.148032 3.05 0.004 1.172963 5.825231
Berdasarkan hasil regresi multivariate. Dapat diketahui bahwa adanya pengaruh antara
temperatur, manuf dan hari terhadap polusi udara dengan p value 0,0001 maka dapat di
simpulkan bahwa terdapat hubungan antara temperatur, manuf dan hari terhadap polusi udara.
[Koefisien x1 – 0,038; 0,17; 0,007 dan konstanta 3,499]
Persamaan garis So2 = [3,499 + (- 0,038) x temp] + [3,499 + 0,170 x manuf] + [3,499 +
(- 0,007) x days].
Artinya temp, manuf, dan hari hanya bisa menjelaskan variasi So2 sebesar 42,39%. F=
9,07 dan P= 0,0001 membuktikan bahwa koefisien β tidak sama dengan 0 atau ada pengaruh
temp, manuf dan hari terhadap So2.
