Date:11-03-2022

Registration number:

ST. JOSEPH’S COLLEGE (AUTONOMOUS), BENGALURU-27

B.Sc. STATISTICS – V SEMESTER
SEMESTER EXAMINATION: OCTOBER 2021
(Examination conducted in March 2022)
ST - 5118: Sampling Theory and Designs of Experiment
Time: 2½ Hours Max: 70 Marks
This question paper contains Two printed pages and Three parts
PART A
I Answer any FIVE from the following 3 x 5 = 15

1. Define (i) Population (ii) Sampling frame (iii) Precision

2. Show that sample mean is an unbiased estimator of population if the sampling scheme
is systematic sampling.
3. Obtain the confidence interval for population proportion
4. What are the basic ideas behind stratified sampling?
5. Stating the assumptions, give the model for one way ANOVA
6. Write a short note on Tuckey’s test
7. What is confounding in factorial experiment? Identify the confounded effect for the
below given block
abc ab c 1
a b ac bc

PART B
II Answer any FIVE from the following 7 x 5 = 35
8. A) What are the Qualities of a Good Questionnaire? Explain
B) Compare sampling errors and nonsampling errors (4+3)
9. A) How do you select simple random sample using random numbers?
B) Show that the element selection probability in sample under simple random
sampling is 1/N (3+4)
10. A) What is stratified random sampling? How do you select sample using stratified
random sampling scheme?
B) Obtain an unbiased estimator for population mean if we use cluster sampling as our
sampling scheme. Assume that all clusters have equal size (4+3)
11. Explain the basic principles of designs of experiment? How these are implemented in
randomized block design (7)
12. Outline the analysis of Latin square design (7)
13. A) Define Main effect and interaction effect in 22 factorial experiments
B) Explain the Yates method of finding sum of squares in factorial experiments (2+5)
14. A) Identify the appropriate sampling scheme in following conditions and Justify

ST-5118-A-21
 i. A situation where a research team is seeking opinions about religion
amongst various age groups
ii. The police set up checkpoints at randomly selected road locations, then
inspected every driver at those locations.
iii. A school chooses 3 randomly selected athletes from each of its sports teams
to participate in a survey about athletics at the school.
B) What is the need for ANOVA (5+2)

PART C
III Answer any TWO from the following 10 x 2 = 20

15. A) Show that E(s2) = S2 under simple random sampling without replacement
B) Distinguish between simple random sampling with replacement (SRSWR) and
simple random sampling without replacement (SRSWOR). Show that SRSWOR is
more efficient than SRSWR (5+5)
16. A) Write a short note on different types of allocations in stratified random sampling
B) Show that 𝑉𝑎𝑟(𝑦̅)𝑠𝑟𝑠 ≥ 𝑉𝑎𝑟(𝑦̅)𝑃𝑟𝑜𝑝 ≥ 𝑉𝑎𝑟(𝑦̅)𝑁𝑒𝑦 (3+7)
17. A) Derive the least square estimators of parameters of completely randomized test
B) Explain the analysis of two way ANOVA (4+6)

ST-5118-A-2021
 Register Number:

DATE:

ST. JOSEPH’S COLLEGE (AUTONOMOUS), BENGALURU-27

B.Sc. STATISTICS - V SEMESTER
SEMESTER EXAMINATION - OCTOBER 2019
ST: 5117 – SAMPLING THEORY AND DESIGN OF EXPERIMENTS

Time: 2½ hrs Max: 70 Marks

This question paper has TWO printed pages and THREE parts

SECTION – A

I Answer any FIVE of the following: 5 x 3 = 15

1. Distinguish between simple random sampling with and without replacement

2. Systematic sampling is a particular case of stratified sampling. Justify
3. Write a note on equal allocation with reference to stratified random sampling
4. Give any three assumptions of one way classification model
5. Define following with examples (i) Yield (ii) Block (iii) Treatment
6. i) In SRSWOR and SRSWR, equality of both variances occurs when
(a) n = 1 (b) n = 2 (c) n ≠1 (d) n > 2

ii) The cluster sampling is more efficient if

(a) Within cluster variation is less (b) Within cluster variation is more
(c) Between cluster variation is more (d) None of the above

iii) In cluster sampling, after selecting clusters, further sampling is done in each cluster.
State True or False

7. i) In a Latin Square design, numbers of rows, columns, and treatments are

(a) All different (b) Not necessarily equal (c) Always equal (d) None

ii) In 22 factorial experiment with two treatments A and B. The symbol a stands for
(a) A and B both at higher level (b) A at higher level and B at lower level
(c) A and B both at lower level (d) A at lower level and B at higher level

iii) When the same interaction is confounded in all the replicates, it is called partial
confounding. True / False

ST-5117-C-19
 SECTION – B

II Answer any FIVE of the following: 5 x 7 = 35

8. A) Explain the procedure of selecting simple random sampling using random number table
B) Give the an expression for 100(1 – 𝛼)% confidence interval for population mean under
simple random sampling with replacement (5+2)

9. A) Derive the expression for strata sample size under Neymann allocation.
B) Derive the expression for cluster mean when clusters are selected from SRSWOR and
clusters with equal size. (4+3)

10. A) Briefly explain the Least Significant Difference along with limitations
B) Why 2X2 Latin square is not used? (5+2)

11. A) How Stratified random Sampling is different from Cluster sampling (3+4)
B) Give the ANOVA table for two way classified data with m number of observations per cell

12. Outline analysis of one way classified data. (7)

13. A) Explain Yates method of computing factorial effect in 23 factorial experiments.

B) Define factorial experiments. Explain the terms (i) Main effects (ii) Interaction effects.
(4+3)
14. A) Describe the procedure of analyzing a Latin Square Design when single observation is
missing.
B) What do you understand by uniformity trial? (5+2)

SECTION – C

III Answer any TWO of the following: 2 x 10 = 20

15. A) Explain the procedure of selection of systematic sample with an example. Derive an
unbiased estimator of population mean under the systematic sampling scheme
B) Distinguish between
i) Population & Sample ii) Census & Sampling survey (6+4)

16. A) Describe the principles of design of experiments.

B) What is meant by confounding in an experiment? Explain the term complete
confounding. (6+4)
17. A) Give the ANOVA table for 22 factorial experiments in r randomized blocks.
B) Show that optimal allocation is better sampling scheme than proportional allocation and
Simple Random sampling (3+7)

ST-5117-C-19