In: Statistics and Probability
To study the effect of temperature on yield in a chemical process, five batches were produced at each of three temperature levels. The results follow.
Temperature |
||||
50°C | 60°C | 70°C | ||
32 | 34 | 29 | ||
22 | 35 | 34 | ||
34 | 38 | 34 | ||
37 | 27 | 36 | ||
30 | 31 | 37 |
a. Construct an analysis of variance table (to 2 decimals but p-value to 4 decimals, if necessary).
Source of Variation | Sum of Squares | Degrees of Freedom | Mean Square | F | p-value |
Treatments | |||||
Error | |||||
Total |
b. Use a level of significance to test whether the temperature level has an effect on the mean yield of the process.
Calculate the value of the test statistic (to 2 decimals).
Here, there are 3 levels of temperature namely 50°C, 60°C and 70°C. There are five batches which can be considered as the replicates of each treatment.
Yij: jth batch (response) receiving ith temperature (treatment);
i=1,2,3 and j=1,2,3,4,5.
k=levels of temperature=3, n=5
H0: Temperature (treatment) effect is Insignificant VS H1: at least effect of one of the temperature level is different from others
R-Software
commands and outputs:
T1=c(32,22,34,37,30)
T2=c(34,35,38,27,31)
T3=c(29,34,34,36,37)
y=c(T1,T2,T3) ##Response
yibar=c(mean(T1),mean(T2),mean(T3))
yibar
[1] 31 33 34
mean(y) ##Grand mean
[1] 32.66667
temp=rep(1:3,each=5)
temperature=factor(temp)
ANOVA=aov(y~temperature)
ANOVA
Call:
aov(formula = y ~ temperature)
Terms:
temperature Residuals
Sum of Squares 23.33333 236.00000
Deg. of Freedom 2 12
Residual standard error: 4.434712
Estimated effects may be unbalanced
summary(ANOVA)
Df Sum Sq Mean Sq F value Pr(>F)
temperature 2 23.33 11.67 0.593
0.568
Residuals 12 236.00 19.67
qf(1-0.05,2,12) ###Tabulated (or critical) value
[1] 3.885294
## Since, calculated F-value=0.593 is less than tabulated
F=3.885294, we Fail to Reject H0.
1-pf(0.593,2,12) ### p-value=P(F>0.593)
[1] 0.5680794
Final answer:
a. Construct an analysis of variance table (to 2 decimals but p-value to 4 decimals, if necessary).
Source of Variation | Sum of Squares | Degrees of Freedom | Mean Square | F | p-value |
Treatments | 23.33 | 2 | 11.67 | 0.593 | 0.568079 |
Error | 236 | 12 | 19.67 | ||
Total | 259.33 | 14 |
b. Use a level of significance to test whether the temperature level has an effect on the mean yield of the process. Calculate the value of the test statistic (to 2 decimals).
The value of test statistic (F-statistic)=0.593.
Since, p-value is greater than alpha (0.05) i.e. level of significance, we fail to reject the null hypothesis H0. This indicates that the temperature level has an NO significant effect on the mean yield of the process.