ASSIGNMENT – 2 (STAT-703)
Q. 1:- (a) Construct the layout plan for four treatments T1, T2, T3 with replications 3, 4, 5 respectively.
(b) An experiment was conducted to test the effects of four transplanting methods on yield of tomato.
Because the field site was fairly uniform, it was decided to use five replicate plots in a completely
randomized design. Unfortunately, two of the plots on the edge of the field were damaged by browsing deer
and had to be discarded, as shown by highlighting below. The transplanting methods are labeled A through
D.
C B B A C
A D C D B
C D C B A
B A D A D
Yield data in t/ha collected from the remaining plots were as follows:
A B C D
37 27 19 30
31 14 15 24
25 22 23 28
38 19 17 27
32 36
1. Discuss how can you handle the missing data shown above?
2. What would the null hypothesis and alternative hypothesis be for this experiment?
3. Compute the analysis of variance of the data?
4. Once you have completed the analysis manually, use the "ANOVA: single factor" program in Excel
to check your results
5. What is your conclusion regarding the null hypothesis for this experiment? Justify your answer. Also
apply the mean comparison test if needed.
6. What is the CV for this experiment and how would you interpret it?
7. Assume that method A is the standard method, but that the newer methods (B, C, and D) are less
expensive. On the basis of these results, what advice would you offer to the researcher? (what do you
think he should do next?). Use statistics to support your answer.
Q. 2:- An irrigation experiment was conducted in a randomized complete block design in an orange grove. Five
irrigation treatments were used in six blocks of trees. The data that follow are the weight in pounds of
harvested fruit from each plot.
Block
Method 1 2 3 4 5 6
Trickle 450 469 299 280 352 251
Basin 358 512 281 352 293 283
Spray 339 402 183 258 281 219
Sprinkler 317 423 379 289 239 269
Flood 245 380 263 336 282 171
1. Compute the analysis of variance manually and do verify your answers by using the Excel Analysis
'Two Factor ANOVA without replication'.
2. Were there significant differences among the irrigation methods in this experiment? How did you
reach your conclusion?
3. Were the blocks significantly different in this experiment? What is your proof?
4. Use mean separation test if needed.
5. Compute relative efficiency of the RCBD and interpret it. How many reps would you need to obtain
the same level of precision in a CRD?
6. Suppose the observation 402 in Block-2 under method Spray is missing, estimate this missing
observation. What changes will the estimation of missing observation bring to the ANOVA table?
Q. 3:- An experimenter wishes to study the effect on the yield of lima beans of different forms of fertilizer. He has
five treatments of interest:
C Control – no application of fertilizer
M Farmyard manure – as an organic fertilizer
F1 Fertilizer F1 – an organic fertilizer at 200 kg/ha
F2 Fertilizer F2 – an organic fertilizer at 200 kg/ha
F Combined F1 and F2 – both at 100 kg/ha
Note that F1 and F2 are different types of inorganic fertilizers.
He plants five plots of lima beans for each treatment, in land believed C M F1 F2 F
to be homogeneous, and obtains the yields (kg) for each plot. The 66.7 69.9 76.2 74.1 82.4
ASSIGNMENT – 2 (STAT-703)
ANOVA table and treatment means are as follows:
Source of Variation Df SS MS F ratio
Treatments 4 727.06 181.77 5.72
Residual 20 635.20 31.76
Total 24 1362.26
(a) The experimenter expresses interest in the following comparisons:-
(i) C vs the rest (ii) M vs F1&F2 (iii) F1 vs F2 (iv) F1&F2 vs F
Test the hypothesis above and explain the steps in the argument.
(b) Calculate C.V and interpret it.
Q. 4:- The following table shows some of the results of an experiment on the effects of applications of sulphur in
reducing scab disease of potatoes. The object in applying sulphur is to increase the acidity of the soil since
scab does not thrive in very acid soil. In addition to untreated plots which serve as a control, 3 amounts of
dressing were compared – 300, 600, and 900 lb. per acre. Both a fall and a spring application of each
amount was tested, so that in all there were seven distinct treatments. The sulphur was spread by hand on
the surface of the soil, and then diced into a depth of about 4 inches. The quantity to be analyzed is the
“scab index”. That is roughly speaking, the percentage of the surface area of potato that is infected with
scab. It is obtained by examining 100 potatoes at random from each plot, grading each potato on a scale
from 0 to 100% infected, and taking the average.
O F3 S3 F6 S6 F9 S9
30 9 30 16 18 10 17
18 9 7 10 24 4 7
32 16 21 18 12 4 16
26 4 9 18 19 5 17
Total 106 38 67 62 73 23 57 426
Mean 26. 9.5 16.75 15. 18.25 5.75 14.25
5 5
(a) Split the treatment sum of squares into following four contrasts:-
(i) Control vs Treatment with 1 df (ii) Fall vs Spring with 1 df (iii) Among fall applications with 2 df
(iv) Among Spring applications with two df
(b) Fit response curve of appropriate degree within fall application and within spring application separately.
Q. 5:- An agronomist conducted an experiment in which he used the following treatments:
1. T1=Fertilizer 2. T2=Herbicide 3. T3=Fertilizer + Herbicide
4. T4=Insecticide 5. T5=Fertilizer + Insecticide 6. T6=Herbicide + Insecticide
Write down the contrast coefficients for the following five contrasts:-
1. Fertilizer alone against herbicide alone
2. Treatments with fertilizer against treatments without fertilizers
3. Treatments with insecticides against treatments without insecticide
4. Fertilizer + herbicide against fertilizer with insecticide
5. Fertilizer + insecticides against herbicide + insecticide
Q. 6:- Four catalysts that may affect the concentration of one component in a three component liquid mixture are
being investigated. The following concentrations are obtained:
Catalyst
TI ri=5 TII rj=4 TIII TIV
58.2 56.3 50.1 52.9
57.2 54.5 54.2 49.9
58.4 57.0 55.4 50.0
55.8 55.3 51.7
54.9
1. Use graphical method to compare mean concentration for different catalyst.
2. Do the four catalysts have the same effect on the concentration?
3. Use appropriate procedure for inference after ANOVA if needed.
Q. 7:- A chemist wishes to test the effect of four chemical agents on the strength of a particular type of cloth.
Because there might be variability from one bolt to another, the chemist decides to use a randomized block
design, with the bolts of cloth considered as blocks. She selects five bolts and applies all four chemicals in
random order to each bolt. The resulting tensile strengths are as follows:-
ASSIGNMENT – 2 (STAT-703)
Bolt
Chemical I II III IV V
I 73 68 74 71 67
II 74 67 76 72 71
III 75 69 78 74 68
IV 65 63 69 70 69
1. Analyze the data from this experiment using α=0.05 and draw appropriate conclusions.
2. Use graphical method to compare mean strength for four chemical.
3. Do the four chemicals have the same effect on the strength? If no, which chemicals have significantly
different effect?
Q. 8:- An article in the ACI Materials Journal (Vol. 84, 1987, pp.213-216) describes several experiments
investigating the rodding of concrete to remove entrapped air. A 3-inch×6-inch cylinder was used, and the
number of times this rod was used is the design variable. The resulting compressive strength of the concrete
specimen is the response. The data are show in the following table:
Rodding Levels Compressive Strength
10 1530 1530 1440
15 1610 1650 1500
20 1560 1730 1530
25 1500 1490 1510
1. Use graphical method to compare mean compressive strength at different rodding levels.
2. Is there any difference in compressive strength due to the rodding level?
3. Calculate CV and interpret it.
4. Fit response curve of appropriate degree.
Q. 9:- Consider the problem of determining whether or not different brands of tires exhibit different amount of
tread loss after 20,000 miles of driving. A fleet manager wishes to consider four brands A,B,C and D that
are available and make some decision about which brand might show the least amount of tread wear after
20,000 miles. The variable to be measured (i.e. Response variable or study variable) is the difference in
maximum tread thickness on a tire between the time it is mounted on the wheel of a car and after it has
completed 20,000 miles on this car. Because of the performance of the cars four cars were used and each
car got one tire of each brand and the following data about difference in thickness in mils (0.001 inch) were
recorded.
Car
I II III IV
B (14) D (11) A (13) C (9)
C (12) C (12) B (13) D (9)
A (17) B (14) D (11) B (8)
D (13) A (14) C (10) A (13)
1. Test the hypothesis that the performance of four brands of tires are same.
2. Calculate C.V and interpret it.
3. Calculate Standard error for brand means.
4. Compare the mean of A&C with B&D.
5. What information is provided by F-ratio for cars.
6. Estimate the gain in precision by introducing different cars as a valid source of variation.
Q. 10:- With reference to Q. 9, experience shows that rear tires get different wear than front tires and even different
sides of the same car may show different amounts of tread wear. So positions can impose another restriction
on the randomization in such a way that each brand is not only used once on each car but also only once in
each of the four possible positions: (I) left front, (II) left rear, (III) right front, and (IV) right rear and the
data is given below:
Car
Position I II III IV
I C (12) D (11) A (13) B (8)
II B (14) C (12) D (11) A (13)
III A (17) B (14) C (10) D (9)
IV D (13) A (14) B (13) C (9)
1. Perform analysis of variance.
2. Calculate CV and compare it with CV calculated in Q.9 and comment on the difference between two
ASSIGNMENT – 2 (STAT-703)
values.
3. Apply Tukey’s Test.
4. Estimate whether position is effective in reducing experimental error.
5. Estimate gain in efficiency due to position as well as cars.
6. Suppose that right front tire of care III blew out and was ruined before completing the 20,000 miles.
Estimate the observation of response variable for missing value and them complete the ANOVA table
and compare it with ANOVA table in part 1.
Q. 11:- (a) Prepare the layout plan for 5 treatments in Latin square design.
An agronomist would like to compare five cassava varieties for yield and quality of processed gari. She
decides to use blocks in the field to control variation due to soil texture. She also suspects that the women
who process cassava may have slightly different processing techniques that could affect the yield of the
final end-product. She hires five women for the study and uses a Latin Square Design to account for
processing differences among the women. Harvesting and processing are time-consuming and take several
days to complete, so the work is carried out on all plots in a block before moving to the next block. Data
below are dry weights of gari in kg/plot.
Block
Processor 1 2 3 4 5
1 V5 24.2 V1 12.3 V2 20.1 V4 26.3 V3 31
2 V4 26.1 V3 24.6 V1 17.4 V2 21.2 V5 33.4
3 V1 17.8 V4 28 V5 24.3 V3 34.2 V2 30.3
4 V3 21.7 V2 23.2 V4 24.8 V5 29 V1 25.4
5 V2 14.3 V5 16.3 V3 27.8 V1 18.3 V4 30
1. Test the hypothesis that variety means are equal.
2. Were both rows (processors) and columns (field blocks) effective in reducing the error variance of this
trial? What factors might have contributed to the variation among blocks?
3. What is the relative efficiency of this design when compared to: (i) RBD with rows as blocks? (ii)
RBD with columns as blocks? (iii) CRD?
4. How many reps would you need to obtain the same level of precision in a CRD?
5. Suppose that the value for Block-3 / Processor-2 is missing. Estimate the missing value and then
complete the ANOVA table.
Q. 12:- An oil company tested four different blends of gasoline for fuel efficiency according to a Latin Square
design in order to control for the variability of four different drivers and four different models of cars. Fuel
efficiency was measured in miles per gallon (mpg) after driving cars over a standard course. Blends are
indicated by letters A-D.
Car model
Driver I II III IV
I D (15.5) B (33.9) C (13.2) A (29.1)
II B (16.3) C (26.6) A (19.4) D (22.8)
III C (10.8) A (31.1) D (17.1) B (30.3)
IV A (14.7) D (34.0) B (19.7) C (21.6)
Analyze the data.
Suppose that the mpg value for Driver III/Model II is missing. Estimate the missing value and then
complete the ANOVA table.
Q.13-factor factorial experiment was carried out in RCBD with 3 blocks. Complete the following.
S.O.V DF SS MSS Fcal. Ftab
Blocks 18.50
A 36.75
B 0.26
AB 1
ASSIGNMENT – 2 (STAT-703)
Error 17.50
TOTAL n-1=11 74.25
Best of Luck!
