AOAC Gluten Quantitative Validation Guidance-Round 1(Nov 2023)

422

Figure 2. Data summary view in R, of data from Figure 1 423 424 The following is a list of the commands you will enter to perform the analysis of variance. Hit Enter after 425 typing in each command. 426 >library(car) 427 >Anova(lm(Result ~ Size + Time + Temp, data = data1) 428 [in this line of code, “lm” is telling it to run a linear model; “Result” is your result column, the 429 dependent variable; “Size”, “Time”, and “Temp” are the column titles for your independent 430 variables, and “data1” is the name of your dataset] 431 The results of the analysis are displayed as shown below: 432

433 434 435 436 437 438 439 440 441 442

Anova Table (Type II tests)

Response: Result

Sum Sq 555.02

Df F value

Pr(>F)

Size Time Temp

1 1 1

133.1173 9.1199 54.1139

1.191e-13 *** 0.004629 ** 1.126e-08 ***

38.02

225.63

Residuals

---

Signif. codes:

0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 443 444 445 The p-values for each factor are shown in the Pr(>F) column. Significance is determined as p<.05, but 446 you have done three comparisons here, so you will want to make a Bonferroni adjustment, and consider 447 a factor to be significant when the p-value is less than 0.017 (i.e., 0.05 divided by 3). In this example, 448 changes to each of the factors makes a significant impact on the results. This means that the method 449 instructions should warn the end user to avoid deviations in any of these steps. 450 If you were analyzing more than the three factors, the code and commands would be the same, you 451 would simply enter the additional column titles after the ~ sign using + signs in between. 452 If you were interested in seeing the interaction effects between each of the factors, you would use the 453 following code: 454 >Anova(lm(Result ~ Size*Time*Temp, data = data1)

455 456