Question

1. A study was conducted to investigate the effect of different types of brands of drugs for the treatment of cold. Part of the analysis is available given below.

Source of variation	SS	df	MS	F ratio	p value
Between groups	12.5	2	6.25	2.63	E
Residual	A	B	C
Total	D	11

1. What type of statistical analysis was carried out?

2. What are the assumptions required for the above test?

3. What are the values for A, B, C, D and E?

4. How many brands of medicines are compared here?

5. If p = 0.130, what conclusion can we make about the results?

Answer 1

a) Type of statistical analysis:

One Way Analysis of Variance

b) Assumptions:

1. The cases are independent of each other.

2. The residuals are normally distributed.

3. The variances of each brand are equal.

c) Missing values:

By the formula:

F ratio = MS(Groups) / MS(Error)

So,

C = MS(Groups)/ F ratio

C = 6.25/2.63

C = 2.38

B = df(Total) - df(Groups) = 11 - 2 = 9

A = B*C = 9*2.38 = 21.39

D = A + 12.5 = 21.39 + 12.5 = 33.89

E = p value = P(F(2, 9) > 2.63) = 0.1261

d) Number of brands of medicines = df(Groups) + 1 = 2 + 1 = 3

e) Since p - value is very high, we do not have sufficient evidence to reject Ho. Hence,

We do not have sufficient evidence to conclude that there is a significant difference among the mean effect of different types of brands given.