Samples of peanut butter produced by 2 different manufacturers are tested for af

Question

Samples of peanut butter produced by 2 different manufacturers are tested for aflatoxin content, with the following results: a. What statistical test would you perform to determine whether there is a difference in peanut butter brand performance? What would the experimental design be? b. Assume you want to test creamy verses crunchy and you want to add a third brand (C) to the experiment. What would your experimental design and statistical test be in that case? How many replications do you think you should have? Provide a "plan" for how you would do all of the testing. c. Is there a standard for testing Aflatoxin? How would you decide how to do the testing?

Explanation / Answer

a. We can perform 2-sample 2 sided t-test (with equal variances) with assumptions of i.i.d and Normality.

The result says there is no significant difference between both brands p-value=0.33>0.05

p-value=(from excel formula=T.TEST(A2:A8,B2:B8,2,2)=0.33

b. We can use ANOVA with 1 factor i.e. brand that has 3 levels. So it will be a 1-way ANOVA

H0: there is no difference between brands

Ha: H0 is not true

We should have at least 30 replication to expect Normality & homoskedasticty

c. We need to check for Normality of the samples. If samples are Normal then we can use parametric t-test else we have to resort to Non-parametric testing such as Wilcoxon rank sum test.

Related Questions

Navigate

• Browse (All)
• Browse S
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.