You are on page 1of 7

Muthu Kumar.

S 17MBA0109

LAB ASSIGNMENT 4
Methodology
Do the ANOVA analysis we have taken a hypothetical situation of the
impact of shelf space and the level of merchandising on the consumer off-
take from the retail outlets. Shelf-space has been categorized in three types:

 POP-SHELF= 1(Point of Purchase)



 WINDOWS= 2

 SHELF RACKS=3
Merchandising and outlet decorations are considered in general to be quite
significant to the level of intake from the respective outlets in consideration
and the classification of the merchandising activities has been made on the
basis of the variety used in the process and the budget involved in it. So the
three types of merchandising that we have taken are as follows:

 Full set (danglers, suspenders, stickers, posters, and POP stand


set=1)

 Retail Pack (danglers, stickers and posters) =2

 No-display pack (posters and stickers) =3

We shall go about envisaging the impact of different types of shelf space


acquired and that of the merchandising packages on the volume of sales
from the respective retail outlets taken into consideration, through the
analysis of variance (ANOVA).

Input Data: Problem (1)


SNo Sales Shelf Space
1 550 1
2 500 2
3 450 3
4 570 1
5 530 2
Muthu Kumar.S 17MBA0109

6 490 3
7 560 1
8 510 2
9 500 3
10 530 1
11 500 2
12 470 3
13 590 1
14 530 2
15 480 3
16 600 1
17 540 2
18 480 3
19 580 1
20 520 2
21 500 3
22 530 1
23 500 2
24 460 3

OUTPUT OBTAINED:
ANOVA
SALES

Sum of Squares df Mean Square F Sig.

Between Groups 29033.333 2 14516.667 34.398 .000


Within Groups 8862.500 21 422.024
Total 37895.833 23

INTERPRETATION
 We use ANOVA to test the significance of the model and its efficiency.
 From the above table we can see that shelf space has a significant impact
as evident from the value of significance level from above table when we
do ANOVA and while by manual calculation it does not have significant
difference.
 This method is generally used when quantity of respondents is less.
Muthu Kumar.S 17MBA0109

Problem (2)
We shall proceed with the problem now adding another column to the
table, that of the merchandising. We shall test two hypotheses by doing an
ANOVA with a randomized bock design. We have used ‘scales’ as the
dependent variable and shelf space as the factor and ‘merchandising” as
the block. These are used to test the null hypothesis.

1. The first hypothesis is that the ales are the same for all the three
categories of ‘shelf space’.

2. The second null hypothesis is that the block (merchandising) used


has no effect on the sales.
The input data for problem (2) are in the following table.
Input data

SNo Sales Shelf Space Merchandising


1 550 1 1
2 500 2 1
3 450 3 1
4 570 1 2
5 530 2 2
6 490 3 2
7 560 1 3
8 510 2 3
9 500 3 3
10 530 1 1
11 500 2 1
12 470 3 1
13 590 1 2
14 530 2 2
15 480 3 2
16 600 1 3
17 540 2 3
18 480 3 3
19 580 1 1
20 520 2 1
21 500 3 1
22 530 1 2
23 500 2 2
24 460 3 2
Muthu Kumar.S 17MBA0109

OUTPUT OBTAINED:
Between-Subjects Factors

SALES 450.00 1

460.00 1

470.00 1

480.00 2

490.00 1

500.00 5

510.00 1

520.00 1

530.00 4

540.00 1

550.00 1

560.00 1

570.00 1

580.00 1

590.00 1

600.00 1
MERCHNADISING 1.00 9

2.00 9

3.00 6

Tests of Between-Subjects Effects


Dependent Variable: SHELFSPACE

Type III Sum


Source of Squares df Mean Square F Sig.

Intercept Hypothesis 73.997 1 73.997 296.983 .370

Error .059 .238 .249a


SALES Hypothesis 14.335 15 .956 3.769 .159
Error .727 2.866 .254b
MERCHNADISING Hypothesis .465 2 .232 .983 .519
Error .410 1.731 .237c
SALES * Hypothesis .483 2 .241 .724 .539
MERCHNADISING Error 1.333 4 .333d

a. 2.467 MS(MERCHNADISING) - 1.789 MS(SALES * MERCHNADISING) + .323 MS(Error)


b. .868 MS(SALES * MERCHNADISING) + .132 MS(Error)
c. 1.052 MS(SALES * MERCHNADISING) - .052 MS(Error)
Muthu Kumar.S 17MBA0109

d. MS(Error)

Expected Mean Squaresa,b

Variance Component

Var(SALES *
Var(MERCHNA MERCHNADISI
Source DISING) NG) Var(Error) Quadratic Term

Intercept Intercept,
6.482 1.028 1.000
SALES
SALES .000 1.107 1.000 SALES
MERCHNADISING 2.628 1.342 1.000
SALES * MERCHNADISING .000 1.276 1.000
Error .000 .000 1.000

a. For each source, the expected mean square equals the sum of the coefficients in the cells times
the variance components, plus a quadratic term involving effects in the Quadratic Term cell.
b. Expected Mean Squares are based on the Type III Sums of Squares.

INTERPRETATION:
 Here dependent variable is merchandising and fixed factor is sales.
 We can see that sales has least impact when analyse it single handed and
when we consider sales with merchandising it has highest impact and the
most preferred one to know the outcome. Hence the when we use both
sales and merchandising it is significant.
a. 2.467 MS(MERCHNADISING) - 1.789 MS(SALES *
MERCHNADISING) + .323 MS(Error)
b. .868 MS(SALES * MERCHNADISING) + .132 MS(Error)
c. 1.052 MS(SALES * MERCHNADISING) - .052 MS(Error)
 Above equations helps to analyse the outcome if we substitute the values
put in above equation.
Muthu Kumar.S 17MBA0109

Problem (3) for this, we have used a modified approach to the previous
problem. The same procedure for problem (2) is followed but in the model
setup we have chosen Full Factorial instead of custom. The error term is
taken to be (within +residual). The input data are in the following table.

OUTPUT OBTAINED:
Between-Subjects Factors

MERCHNADISING 1.00 9

2.00 9

3.00 6
SHELFSPACE 1.00 8

2.00 8

3.00 8

Tests of Between-Subjects Effects


Dependent Variable: SALES

Type III Sum of


Source Squares df Mean Square F Sig.

Corrected Model 30712.500a 8 3839.063 8.017 .000


Intercept 6280134.127 1 6280134.127 13113.969 .000
MERCHNADISING 1523.611 2 761.806 1.591 .236
SHELFSPACE 28392.063 2 14196.032 29.644 .000
MERCHNADISING *
155.556 4 38.889 .081 .987
SHELFSPACE
Error 7183.333 15 478.889
Total 6517100.000 24
Corrected Total 37895.833 23

a. R Squared = .810 (Adjusted R Squared = .709)


Muthu Kumar.S 17MBA0109

Tests of Between-Subjects Effects


Dependent Variable: SALES

Type III Sum of


Source Squares df Mean Square F Sig.

Corrected Model 30712.500a 8 3839.063 8.017 .000


Intercept 6280134.127 1 6280134.127 13113.969 .000
MERCHNADISING 1523.611 2 761.806 1.591 .236
SHELFSPACE 28392.063 2 14196.032 29.644 .000
MERCHNADISING *
155.556 4 38.889 .081 .987
SHELFSPACE
Error 7183.333 15 478.889
Total 6517100.000 24
Corrected Total 37895.833 23

a. R Squared = .810 (Adjusted R Squared = .709)

INTERPRETATION:
 Here we analysed the single impact of merchandising, shelf space and
shelf space + merchandising together.
 We can clearly see that shelf space alone is significant while
merchandising and shelf space + merchandising are insignificant as
evident from the above table.

You might also like