**Question 1**

Based on the values in Cell A51: A55, what formula can you copy and paste in to Cells B51: B55 to return the values shown?

If the formula in Cell D49 is copied to Cells E49:F49, what sequence of values would be generated in Cells D49:F49?

Question 2

In cell B28, calculate the number of workdays to deliver the order based on the lookup table in the range A14:C17 of the Deliver Calculator worksheet and using the value of cell B27 as the lookup value. If the formula returns an error value, display no text in the cell.

Question 3

a) Discuss the impact of departure from normality on the one-way analysis of variance.

b) Discuss the impact of departure from homoscedasticity (constant variance) on the one-way analysis of variance.

c) What is a conservative test of hypothesis?

d) Which multiple comparisons method (s) should be used if the residuals show a clear departure from normality?

Answer to question 1

Correct Option is =IF(A51=”Red”,”Yes”,”No”)

Explanation : If content of cell A51 is Red, print Yes in B51 else print No

Correct Option is C,C,C

The cell content is absolutely reffered and hence content of cell D49 is copied as it is

Answer to question 2

Put following formula in Cell B28

=IFERROR(VLOOKUP(B28,’Deliver Calculator’!A14:C17, **3**),””)

Above will put the value of Column C for the matching value of Cell B28 in A14 to A17

For the highlighted 3 in above formula, replace it according to your need

- If you want value from Column A then put 1 here
- If you want value from Column B then put 2 here
- If you want value from Column C then put 3 here

Answer to question 3

1. If the given data depart from the normal distribution in One-way ANOVA then the ANOVA is invalid. ANOVA is a parametric test that is based on the assumptions that the data follows a normal distribution, So it is necessary to test the normality.

2. If the homoscedasticity of one-way Analysis of variance is violated then the Analysis of variance is not valid. In such a case two test that will work is the Welch test or Brown & Forsythe test. Alternatively,the Kruskal-Wallis test will also be run.

3. A test is said to be conservative if, it is constructed for a nominal significance level. The true probability of incorrectly rejecting the null hypothesis can never be greater than the nominallevel.

4. The multiple comparisons method that should be used if the residuals show a clear departure from normality is Shapiro-Wilk Normality test