Test equality of efficiency of filling machines A, B, C, D and E using the sampled tin weights.
machine_data <- list(
A = c(14.85, 15.00, 15.25, 15.10, 14.80),
B = c(14.28, 14.42),
C = c(14.16, 14.15, 14.19, 14.50),
D = c(15.25, 15.30, 15.10, 15.35, 15.00),
E = c(14.60, 14.84, 14.82, 14.74)
)
df <- stack(machine_data)
names(df) <- c("Weight", "Machine")
df$Machine <- factor(df$Machine)
df
## Weight Machine
## 1 14.85 A
## 2 15.00 A
## 3 15.25 A
## 4 15.10 A
## 5 14.80 A
## 6 14.28 B
## 7 14.42 B
## 8 14.16 C
## 9 14.15 C
## 10 14.19 C
## 11 14.50 C
## 12 15.25 D
## 13 15.30 D
## 14 15.10 D
## 15 15.35 D
## 16 15.00 D
## 17 14.60 E
## 18 14.84 E
## 19 14.82 E
## 20 14.74 E
aggregate(Weight ~ Machine, data = df, FUN = mean)
## Machine Weight
## 1 A 15.00
## 2 B 14.35
## 3 C 14.25
## 4 D 15.20
## 5 E 14.75
fit <- aov(Weight ~ Machine, data = df)
anova_table <- summary(fit)
anova_table
## Df Sum Sq Mean Sq F value Pr(>F)
## Machine 4 2.6205 0.6551 28.11 8.18e-07 ***
## Residuals 15 0.3496 0.0233
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
shapiro.test(residuals(fit))
##
## Shapiro-Wilk normality test
##
## data: residuals(fit)
## W = 0.95131, p-value = 0.3874
bartlett.test(Weight ~ Machine, data = df)
##
## Bartlett test of homogeneity of variances
##
## data: Weight by Machine
## Bartlett's K-squared = 1.0404, df = 4, p-value = 0.9036
TukeyHSD(fit)
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Weight ~ Machine, data = df)
##
## $Machine
## diff lwr upr p adj
## B-A -0.65 -1.044417135 -0.25558287 0.0010765
## C-A -0.75 -1.066237240 -0.43376276 0.0000216
## D-A 0.20 -0.098151329 0.49815133 0.2816168
## E-A -0.25 -0.566237240 0.06623724 0.1576262
## C-B -0.10 -0.508260521 0.30826052 0.9391613
## D-B 0.85 0.455582865 1.24441713 0.0000651
## E-B 0.40 -0.008260521 0.80826052 0.0561479
## D-C 0.95 0.633762760 1.26623724 0.0000012
## E-C 0.50 0.166656681 0.83334332 0.0025661
## E-D -0.45 -0.766237240 -0.13376276 0.0040554
p_machine <- anova_table[[1]]["Machine", "Pr(>F)"]
if (p_machine < alpha) {
cat("Decision: Reject H0 at 5% significance level.\n")
cat("Conclusion: Machine means are not all equal; efficiency differs significantly.")
} else {
cat("Decision: Fail to reject H0 at 5% significance level.\n")
cat("Conclusion: No significant difference in mean efficiency is detected across machines.")
}
## Decision: Reject H0 at 5% significance level.
## Conclusion: Machine means are not all equal; efficiency differs significantly.