Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | "Joseph Coveney" <stajc2@gmail.com> |
| To | <statalist@hsphsun2.harvard.edu> |
| Subject | st: Re: ANOVA |
| Date | Fri, 7 Mar 2014 09:56:49 +0900 |
Wendy Alfaro wrote:
I need to run and Anova with these data. How do I perform the test with Stata?
Dureza primer mordisco
Tratamiento (Sust. Huevo)% Muestra 1 Muestra 2 Muestra 3
1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3
100% 29,7562 35,0566 32,8892 19,9594 28,9730 22,3698 37,4985 31,2923 31,8955
75% 22,3752 27,6344 30,8096 22,0975 22,4783 28,2321 27,9858 25,2717 21,7764
50% 25,9648 29,3983 26.9672 30,4386 27,0703 31,4941 43,7090 62,9589 44,2992
0% 24,8714 24,0784 28,6552 33,1658 33,7440 33,9544 20,8814 24,6869 19,3974
I need to run and Anova with these data. How do I perform the test with Stata?
Dureza segundo mordisco
Tratamiento (Sust. Huevo)% Muestra 1 Muestra 2 Muestra 3
1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3
100% 20,0689 25,8021 21,6245 15,8371 21,7363 17,6238 29,5198 25,1263 25,4528
75% 15,3175 19,0080 22,3080 17,3005 19,1132 20,8489 23,4535 21,6940 19,2889
50% 18,8561 22,0628 19,1392 22,9024 21,7797 26,4107 34,7594 50,0161 36,8400
0% 19,5395 17,8472 23,7215 28,3482 27,5487 28,6693 18,2164 20,9888 16,1398
Dureza, pico negativo de elasticidad del primer pico
Tratamiento (Sust. Huevo)% Muestra 1 Muestra 2 Muestra 3
1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3
100% -0,0409 -0,0499 -0,1909 -0,0835 -0,1052 -0,0282 -0,1182 -0,1367 -0,0488
75% -0,0824 -0,1410 -0,0184 -0,0846 -0,0239 -0,0998 0,0033 -0,0759 0,0141
50% -0,0358 -0,0955 -0,0108 -0,0054 -0,0184 0,0011 -0,0098 -0,0749 -0,0098
0% -0,0228 -0,0152 -0,0011 -0,0033 -0,0521 0 -0,0054 0,0087 0
Dureza, pico negativo de elasticidad del segundo pico
Tratamiento (Sust. Huevo)% Muestra 1 Muestra 2 Muestra 3
1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3
100% -0,1258 -0,0911 -0,0949 -0,0879 -0,0868 -0,0358 -0,1345 -0,0868 -0,1009
75% -0,0683 -0,0087 -0,0477 -0,0792 -0,0184 -0,1041 -0,0759 -0,0694 -0,0629
50% -0,0250 -0,1041 -0,0065 0 -0,0022 -0,0597 -0,0477 -0,0239 -0,0184
0% -0,0011 -0,0119 -0,0466 -0,0553 0,0499 -0,0510 0,0065 -0,0304 -0,0108
--------------------------------------------------------------------------------
Try something like that below. I show the MANCOVA table beneath the do-file.
I have no idea what your study is all about (for example, for the sake of data
input, es. mordisco => Eng. bite), or whether the pair of paired datasets is
supposed to represent two outcome variables for common explanatory variables.
But MANCOVA is always more fun to me than ANOVA, so that's how I set it up.
Joseph Coveney
clear *
set more off
// Nota bene: "hard-coded" correction of data-entry error: 26.9672 -> 26,9672
quietly input str244 a1-a30
100% 29,7562 35,0566 32,8892 19,9594 28,9730 22,3698 37,4985 31,2923 31,8955
75% 22,3752 27,6344 30,8096 22,0975 22,4783 28,2321 27,9858 25,2717 21,7764
50% 25,9648 29,3983 26,9672 30,4386 27,0703 31,4941 43,7090 62,9589 44,2992
0% 24,8714 24,0784 28,6552 33,1658 33,7440 33,9544 20,8814 24,6869 19,3974
end
generate byte bite = 1
tempfile tmpfil0
quietly save `tmpfil0'
drop _all
quietly input str244 a1-a30
100% 20,0689 25,8021 21,6245 15,8371 21,7363 17,6238 29,5198 25,1263 25,4528
75% 15,3175 19,0080 22,3080 17,3005 19,1132 20,8489 23,4535 21,6940 19,2889
50% 18,8561 22,0628 19,1392 22,9024 21,7797 26,4107 34,7594 50,0161 36,8400
0% 19,5395 17,8472 23,7215 28,3482 27,5487 28,6693 18,2164 20,9888 16,1398
end
generate byte bite = 2
append using `tmpfil0'
generate long h11 = real(a3 + a5)
generate long h12 = real(a6 + a8)
generate long h13 = real(a9 + a11)
generate long h21 = real(a12 + a14)
generate long h22 = real(a15 + a17)
generate long h23 = real(a18 + a20)
generate long h31 = real(a21 + a23)
generate long h32 = real(a24 + a26)
generate long h33 = real(a27 + a29)
destring a1, generate(trt)
drop a*
quietly save `tmpfil0', replace
drop _all
// Nota bene: "hard-coded" correction of data format inconsistency
quietly input str244 a1-a30
100% -0,0409 -0,0499 -0,1909 -0,0835 -0,1052 -0,0282 -0,1182 -0,1367 -0,0488
75% -0,0824 -0,1410 -0,0184 -0,0846 -0,0239 -0,0998 0,0033 -0,0759 0,0141
50% -0,0358 -0,0955 -0,0108 -0,0054 -0,0184 0,0011 -0,0098 -0,0749 -0,0098
0% -0,0228 -0,0152 -0,0011 -0,0033 -0,0521 0,0000 -0,0054 0,0087 0,0000
end
generate byte bite = 1
tempfile tmpfil1
quietly save `tmpfil1'
drop _all
// Nota bene: "hard-coded" correction of data format inconsistency
quietly input str244 a1-a30
100% -0,1258 -0,0911 -0,0949 -0,0879 -0,0868 -0,0358 -0,1345 -0,0868 -0,1009
75% -0,0683 -0,0087 -0,0477 -0,0792 -0,0184 -0,1041 -0,0759 -0,0694 -0,0629
50% -0,0250 -0,1041 -0,0065 0,0000 -0,0022 -0,0597 -0,0477 -0,0239 -0,0184
0% -0,0011 -0,0119 -0,0466 -0,0553 0,0499 -0,0510 0,0065 -0,0304 -0,0108
end
generate byte bite = 2
append using `tmpfil1'
generate long enp11 = real(a3 + a5)
generate long enp12 = real(a6 + a8)
generate long enp13 = real(a9 + a11)
generate long enp21 = real(a12 + a14)
generate long enp22 = real(a15 + a17)
generate long enp23 = real(a18 + a20)
generate long enp31 = real(a21 + a23)
generate long enp32 = real(a24 + a26)
generate long enp33 = real(a27 + a29)
destring a1, generate(trt)
drop a*
merge 1:1 trt bite using `tmpfil0', assert(match) nogenerate noreport
quietly reshape long enp1 enp2 enp3 h1 h2 h3, i(trt bite) j(reading)
quietly reshape long enp h, i(trt bite reading) j(sample)
rename h hard
manova enp hard = trt bite trt#bite / sample|trt#bite ///
reading trt#reading bite#reading trt#bite#reading
exit
MANCOVA table, copied and pasted from Results Window:
Number of obs = 72
W = Wilks' lambda L = Lawley-Hotelling trace
P = Pillai's trace R = Roy's largest root
Source | Statistic df F(df1, df2) = F Prob>F
-----------------+--------------------------------------------------
Model | W 0.0402 39 78.0 62.0 3.17 0.0000 e
| P 1.5274 78.0 64.0 2.65 0.0000 a
| L 9.7576 78.0 60.0 3.75 0.0000 a
| R 7.9899 39.0 32.0 6.56 0.0000 u
|--------------------------------------------------
Residual | 32
-----------------+--------------------------------------------------
trt | W 0.1393 3 6.0 30.0 8.40 0.0000 e
| P 1.0668 6.0 32.0 6.10 0.0002 a
| L 4.7005 6.0 28.0 10.97 0.0000 a
| R 4.3612 3.0 16.0 23.26 0.0000 u
|--------------------------------------------------
bite | W 0.7214 1 2.0 15.0 2.90 0.0864 e
| P 0.2786 2.0 15.0 2.90 0.0864 e
| L 0.3861 2.0 15.0 2.90 0.0864 e
| R 0.3861 2.0 15.0 2.90 0.0864 e
|--------------------------------------------------
trt#bite | W 0.9862 3 6.0 30.0 0.03 0.9998 e
| P 0.0139 6.0 32.0 0.04 0.9997 a
| L 0.0140 6.0 28.0 0.03 0.9998 a
| R 0.0121 3.0 16.0 0.06 0.9779 u
|--------------------------------------------------
sample|trt#bite | 16
-----------------+--------------------------------------------------
reading | W 0.8531 2 4.0 62.0 1.28 0.2869 e
| P 0.1493 4.0 64.0 1.29 0.2831 a
| L 0.1694 4.0 60.0 1.27 0.2915 a
| R 0.1510 2.0 32.0 2.42 0.1053 u
|--------------------------------------------------
trt#reading | W 0.8122 6 12.0 62.0 0.57 0.8607 e
| P 0.1917 12.0 64.0 0.57 0.8616 a
| L 0.2264 12.0 60.0 0.57 0.8606 a
| R 0.2025 6.0 32.0 1.08 0.3950 u
|--------------------------------------------------
bite#reading | W 0.8767 2 4.0 62.0 1.05 0.3868 e
| P 0.1233 4.0 64.0 1.05 0.3880 a
| L 0.1406 4.0 60.0 1.05 0.3868 a
| R 0.1406 2.0 32.0 2.25 0.1218 u
|--------------------------------------------------
trt#bite#reading | W 0.9159 6 12.0 62.0 0.23 0.9960 e
| P 0.0846 12.0 64.0 0.24 0.9957 a
| L 0.0914 12.0 60.0 0.23 0.9962 a
| R 0.0859 6.0 32.0 0.46 0.8339 u
|--------------------------------------------------
Residual | 32
-----------------+--------------------------------------------------
Total | 71
--------------------------------------------------------------------
e = exact, a = approximate, u = upper bound on F
.
. exit
end of do-file
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/