Bonjour à tous,
j'ai rencontré un statisticien qui m'a expliqué comment m'y prendre... Voici ce à quoi je suis parvenu. J'ai finalement utilisé un paired t test, est-ce correct ? Comment interpréter ces résultats ?
Merci si quelqu'un à le courage de me donner quelques explications...
A bientôt
Vlad
AGE:
x1<-c(64,64,52,50,38,55,57,45,48,59)
y1<-c(64,64,52,50,38,55,57,45,48,59)
z1<-c(64,64,52,50,38,55,57,45,48,59)
a<-c(rep(1,length(x1)),rep(2,length(y1)),rep(3,length(z1)))
boxplot(c(x1,y1,z1)~a)
TAILLE:
x2<-c(154,155,157,177,171,160,156,163,157,185)
y2<-c(154,155,157,176,171,160,156,162,157,185)
z2<-c(154,155,156,176,171,160,156,163,157,185)
a<-c(rep(1,length(x2)),rep(2,length(y2)),rep(3,length(z2)))
boxplot(c(x2,y2,z2)~a)
POIDS:
x3<-c(81,82,98,112,87,76,81,103,83,105)
y3<-c(80,83,100,114,87,76,78,105,86,103)
z3<-c(77,82,95,114,87,76,73,101,87,104)
a<-c(rep(1,length(x3)),rep(2,length(y3)),rep(3,length(z3)))
boxplot(c(x3,y3,z3)~a)
x3<-c(81,82,98,112,87,76,81,103,83,105)
y3<-c(80,83,100,114,87,76,78,105,86,103)
z3<-c(77,82,95,114,87,76,73,101,87,104)
boxplot(c(x3,y3,z3)~a)
t.test(y3,z3,paired=T)
Paired t-test
data: x3 and y3
t = -0.647, df = 9, p-value = 0.5338
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval: -1.7985588 0.9985588
sample estimates:
mean of the differences -0.4
Paired t-test
data: y3 and z3
t = 2.0969, df = 9, p-value = 0.06545
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval: -0.1261048 3.3261048
sample estimates:
mean of the differences 1.6
TOUR DE TAILLE:
x4<-c(118,115,132,118,106,98,107,120,113,115)
y4<-c(116,114,130,119,106,98,105,120,109,114)
z4<-c(113,118,132,120,107,96,101,113,103,113)
a<-c(rep(1,length(x4)),rep(2,length(y4)),rep(3,length(z4)))
boxplot(c(x4,y4,z4)~a)
x4<-c(118,115,132,118,106,98,107,120,113,115)
y4<-c(116,114,130,119,106,98,105,120,109,114)
z4<-c(113,118,132,120,107,96,101,113,103,113)
boxplot(c(x4,y4,z4)~a)
t.test(y4,z4,paired=T)
Paired t-test
data: x4 and y4
t = 2.4004, df = 9, p-value = 0.03987 ---‡ POURQUOI ???
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval: 0.06334936 2.13665064
sample estimates:
mean of the differences 1.1
Paired t-test
data: y4 and z4
t = 1.3299, df = 9, p-value = 0.2163
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval: -1.051551 4.051551
sample estimates:
mean of the differences 1.5
BMI:
x5<-c(34.15,34.2,39.76,35.75,29.75,29.69,33.08,38.77,33.67,30.68)
y5<-c(33.73,34.33,40.57,36.8,29.58,29.69,32.05,40.01,34.89,30.09)
z5<-c(32.48,34.13,39.03,36.8,29.75,29.88,30,38.01,35.3,30.39)
a<-c(rep(1,length(x5)),rep(2,length(y5)),rep(3,length(z5)))
boxplot(c(x5,y5,z5)~a)
x5<-c(34.15,34.2,39.76,35.75,29.75,29.69,33.08,38.77,33.67,30.68)
y5<-c(33.73,34.33,40.57,36.8,29.58,29.69,32.05,40.01,34.89,30.09)
z5<-c(32.48,34.13,39.03,36.8,29.75,29.88,30,38.01,35.3,30.39)
boxplot(c(x5,y5,z5)~a)
t.test(y5,z5,paired=T)
Paired t-test
data: x5 and y5
t = -0.8738, df = 9, p-value = 0.4049
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:-0.8038874 0.3558874
sample estimates:
mean of the differences -0.224
Paired t-test
data: y5 and z5
t = 1.8947, df = 9, p-value = 0.09066
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:-0.1157933 1.3097933
sample estimates: mean of the differences 0.597
TEST 12 MINUTES:
x6<-c(1155,1170,930,1470,935,1215,1160,980,1065,1289)
y6<-c(1115,1140,975,1495,995,1165,1225,925,1145,1315)
z6<-c(1230,1180,1120,1480,970,1385,1290,1055,1250,1455)
a<-c(rep(1,length(x6)),rep(2,length(y6)),rep(3,length(z6)))
boxplot(c(x6,y6,z6)~a)
x6<-c(1155,1170,930,1470,935,1215,1160,980,1065,1289)
y6<-c(1115,1140,975,1495,995,1165,1225,925,1145,1315)
z6<-c(1230,1180,1120,1480,970,1385,1290,1055,1250,1455)
boxplot(c(x6,y6,z6)~a)
t.test(y6,z6,paired=T)
Paired t-test
data: x6 and y6
t = -0.7718, df = 9, p-value = 0.46
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval: -49.53194 24.33194
sample estimates: mean of the differences -12.6
Paired t-test
data: y6 and z6
t = -3.8197, df = 9, p-value = 0.004092
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval: -146.48518 -37.51482
sample estimates:
mean of the differences -92
EVALUATION DE BORG:
x7<-c(15,15,16,15,13,13,13,15,14,11)
y7<-c(14,15,13,16,13,11,13,15,15,11)
z7<-c(15,13,13,16,11,12,14,11,13,11)
a<-c(rep(1,length(x7)),rep(2,length(y7)),rep(3,length(z7)))
boxplot(c(x7,y7,z7)~a)
x7<-c(15,15,16,15,13,13,13,15,14,11)
y7<-c(14,15,13,16,13,11,13,15,15,11)
z7<-c(15,13,13,16,11,12,14,11,13,11)
boxplot(c(x7,y7,z7)~a)
t.test(y7,z7,paired=T)
Paired t-test
data: x7 and y7
t = 1, df = 9, p-value = 0.3434
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval: -0.5048629 1.3048629
sample estimates:
mean of the differences 0.4
Paired t-test
data: y7 and z7
t = 1.2999, df = 9, p-value = 0.2259
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval: -0.5182089 1.9182089
sample estimates: mean of the differences 0.7
PODOMETRIE:
x8<-c(14121,11054,7003,6860,6257,5475,10309,6928,8401,6667)
y8<-c(6841,12098,9039,7375,4792,9556,6389,12061,6990,6675)
z8<-c(8468,10746,7032,8237,4966,9523,9847,12588,10120,8000)
a<-c(rep(1,length(x8)),rep(2,length(y8)),rep(3,length(z8)))
boxplot(c(x8,y8,z8)~a)
x8<-c(14121,11054,7003,6860,6257,5475,10309,6928,8401,6667)
y8<-c(6841,12098,9039,7375,4792,9556,6389,12061,6990,6675)
z8<-c(8468,10746,7032,8237,4966,9523,9847,12588,10120,8000)
boxplot(c(x8,y8,z8)~a)
t.test(y8,z8,paired=T)
Paired t-test
data: x8 and y8
t = 0.1087, df = 9, p-value = 0.9159
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval: -2495.408 2747.208
sample estimates:
mean of the differences 125.9
Paired t-test
data: y8 and z8
t = -1.4047, df = 9, p-value = 0.1937
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval: -2012.8747 470.6747
sample estimates:
mean of the differences -771.1
CORRELATIONS:
- Borg/test de 12 min.
Call:
lm(formula = w7 ~ w6)
Residuals:
Min 1Q Median 3Q Max
-2.5310 -0.5075 -0.4754 1.5019 2.5273
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.337e+01 2.194e+00 6.094 1.42e-06 ***
w6 1.112e-04 1.847e-03 0.060 0.952
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.685 on 28 degrees of freedom
Multiple R-Squared: 0.0001293, Adjusted R-squared: -0.03558
F-statistic: 0.003622 on 1 and 28 DF, p-value: 0.9524
- Podomètre/test 12 min
Call:
lm(formula = w8 ~ w6)
Residuals:
Min 1Q Median 3Q Max
-3730.2 -1639.6 -295.1 1590.5 5635.6
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 8750.7869 3144.9709 2.782 0.00955 **
w6 -0.2298 2.6479 -0.087 0.93147
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 2415 on 28 degrees of freedom
Multiple R-Squared: 0.0002688, Adjusted R-squared: -0.03544
F-statistic: 0.007529 on 1 and 28 DF, p-value: 0.9315
- Borg/podométrie
Call:
lm(formula = w8 ~ w7)
Residuals:
Min 1Q Median 3Q Max
-3585.1 -1847.6 -542.1 1705.4 5329.9
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 5685.8 3645.6 1.560 0.130
w7 207.0 268.1 0.772 0.446
Residual standard error: 2390 on 28 degrees of freedom
Multiple R-Squared: 0.02085, Adjusted R-squared: -0.01412
F-statistic: 0.5963 on 1 and 28 DF, p-value: 0.4465
-> Pas de corrélations observées !!!