Upload
rcoolrahul0039341
View
215
Download
0
Embed Size (px)
Citation preview
7/29/2019 Experimental DesignTotorial
1/4
Experimental Designs ( Tutorials)
I. To test effect of 2 factors, pack design & price on sales of a toilet soap brand.a sampleof 18 retail outlets was selected. They were exposed to three levels of pack designs blue,
green and red and 3 price levels Rs8, Rs11 & Rs 14. The data collected is shown in
Table 1
Table 1 Input Data for Experimental Design
S.No. Sales Pack design Price
1 500 1 1
2 440 2 1
3 360 3 1
4 300 1 2
5 280 2 2
6 250 3 2
7 200 1 3
8 150 2 3
9 250 3 3
10 600 1 1
11 450 2 1
12 510 3 1
13 400 1 2
14 350 2 2
15 300 3 2
16 250 1 3
17 275 2 3
18 220 3 3
The output of SPSS is shown in Table 2
Table 2
Source of Variation Sum of Squares DF Mean Square F Sig of F
Main effects 209305.556 4 52326.389 13.645 .001
Pack design 12536.11 2 6238.056 1.635 .248
Price 196769.444 2 98384.722 25.656 .000
2 Way interactions 9838.889 4 2459.722 .641 .646
Pack Designx price 9838.889 4 2459.722 .641 .646
Residual 34512.500 9 3834.722
Total 253656.944 17 14920.997
Q1 Identify dependent and independent variables
Q2 Identify the type of Design
Q3 Formulate appropriate hypothesis
Q4 Interpret the results?
Solution
Ans 1
1. Dependent variable = Sales
2. Independent variables = prices , pack designs
Ans 2
Factorial Design
Ans 3
Hypothesis 1: The mean level of sales remains the same for all three levels of pack design (main effect 1)
Hypothesis 2: The mean level of sales remain the same for all 3 levels of prices (main effect 2)
Hypothesis 3: The mean level of sales remain the same for all combination of pack designs and price (interaction effect)
Ans 4
We find significance of F values (P values) are as follows
Pack design = .248 > 0.05,
Price = .000 < .05Pack design by price = .646 >.05
1
7/29/2019 Experimental DesignTotorial
2/4
Accordingly price effect is statistically significant at 95% confidence level. Since P value is < .05, this means hypothesis No. 2 is
rejected
2. Hypothesis 1 & 3 cannot be rejected as corresponding p values are greater than .05
3. We conclude that price alone has an impact as sales. Neither pack design nor combination of pack design with price has any
significant impact on sales
II. The Taj group hotel chain wants to test preference of the target customer (high incomegroup) for 4 non vegetarian dishes. 28 customers were asked to give their opinion about
the taste of the 4 common non veg dishes served to them on 10 point rating scale.
The dishes were coded as follows
CODE DISH NAME
1 CHICKEN PLATTER
2 HONEY CHICKEN
3 CHICKEN SPINACH
4 TANDOORI CHICKEN
The data on rating given by 28 respondents is shown in Table 1
S. No. Dish Type Rating
1 1 6.00
2 1 7.003 1 8.00
4 1 5.00
5 1 9.00
6 1 8.00
7 1 7.00
8 2 8.00
9 2 8.00
10 2 9.00
11 2 8.00
12 2 7.00
13 2 9.00
14 2 8.00
15 3 7.00
16 3 6.00
17 3 6.00
18 3 5.0019 3 7.00
20 3 7.00
21 3 5.00
22 4 6.00
23 4 6.00
24 4 7.00
25 4 6.00
26 4 8.00
27 4 7.00
28 4 6.00
The output of SPSS is shown in Table 2
Source of Variation Sum of Squares DF Mean Square F Sig of F
Dish Type 15.714 3 5.238 5.641 .005
Error 22.286 24 .929Total 38.000 27 1.407
Q1 Identify dependent and independent variable
Q2 Describe the type of DesignQ3 Formulate the hypothesis
Q4 Interpret the results
2
7/29/2019 Experimental DesignTotorial
3/4
III.
In the above example the location of the Taj Group Hotel was considered as an extraneous variable. Seven locations were used and coded as
follows
CODE LOCATION
1 CHENNAI
2 BANGALORE3 MUMBAI
4 NEW DELHI
5 GOA
6 KODAIKANAL
7 JAIPUR
The data on rating, dishes & cities is given in Table 1
S.No Dish Type Rating Location
1 1 6.00 1
2 1 7.00 2
3 1 8.00 3
4 1 5.00 4
5 1 9.00 5
6 1 8.00 6
7 1 7.00 7
8 2 8.00 1
9 2 8.00 2
10 2 9.00 3
11 2 8.00 4
12 2 7.00 5
13 2 9.00 6
14 2 8.00 7
15 3 7.00 1
16 3 6.00 2
17 3 6.00 3
18 3 5.00 4
19 3 7.00 5
20 3 7.00 6
21 3 5.00 7
22 4 6.00 1
23 4 6.00 2
24 4 7.00 3
25 4 6.00 4
26 4 8.00 5
27 4 7.00 6
28 4 6.00 7
The output of SPSS is given in Table 3
3
7/29/2019 Experimental DesignTotorial
4/4
Table 3
Source of Variation SS DF MS F Sig
Dish Type 15.71 3 5.24 8.35 .001
Location 11.00 6 1.83 2.92 .036
Error 11.29 18 .63
(Total) 38.00 27 1.41
Q1 Identify the dependent, independent & extraneous variable
Q2 Identify the type of designQ3 Formulate the hypothesisQ4 Interpret the results
Ans 1
Dependent variable = rating , Independent variable = dish type, extraneous variable = location of the hotel
Ans 2
Type of design randomized block design
Ans 3
Hypothesis no 1 = The mean rating of the dish type is same for all 4 dishesHypothesis no 2 = The mean rating of the dish type is same for all 7 blocks
Ans4
Since P values of dish type(.001) as well as location(.036) are both less than0.05, we reject both the hypotheses & Conclude that
i) Mean ratings given to 4 types of dishes are significantly different.i i) The locations have an impact on ratings.
IV) Analyse and interpret the following data concerning output of wheat for four varieties A, B, C & D obtained as a result of experiment
conducted using a latin square design. Columns indicate four levels of moisture contents. Rows indicate four levels of soil fertility
C 25 B 23 A 20 D 20
A 19 D 19 C 21 B 18
B 19 A 14 D 17 C 20
D 17 C 20 B 21 A 15
The output of SPSS is shown in Table 1
Table 1
Source of Variation Sum of Squares DF Mean Square F Table
value of F
at 5%
Between Columns 7.5 3 2.50 1.43 4.76
Between Rows 46.5 3 15.50 8.85 4.76
Between Varieties 48.5 3 16.17 9.24 4.76
Residual or Error 10.5 6 1.75
Total 113.0 15
Q1 Identify the dependent, independent & extraneous variable
Q2 Identify the type of design
Q3 Formulate the hypothesis
Q4 Interpret the results
4