ASRB NET Sampling Theory II: Unequal Probability, Cluster, Multistage & Repeated Sampling: practice questions
ASRB NET - Agricultural Statistics (and Computer Application / Bioinformatics) · India · Agricultural Scientists Recruitment Board (ICAR)
Original practice questions written to the published syllabus · not past papers
10worked here, free
94written to this exam in this subject
12syllabus units
About ASRB NET
ASRB NET - Agricultural Statistics (and Computer Application / Bioinformatics) is a professional certification run by Agricultural Scientists Recruitment Board (ICAR), for candidates in India. NET: one objective paper of about 150 questions, 2 hours, negative marking, online, Master's level (pattern as reported for other disciplines).
We have not been able to confirm these details against the exam body's own specification, so fees, dates and pass marks are left out on purpose. Check the current format, eligibility, fees and dates on the official site.
Sampling Theory II: Unequal Probability, Cluster, Multistage & Repeated Sampling in ASRB NET
Applied design and estimation for agricultural surveys with these methods.
The designs real agricultural surveys use: PPS sampling with and without replacement and its estimators, cluster and multistage sampling, double sampling, sampling on successive occasions, randomised response, non-response and imputation.
Our 94 ASRB NET questions in this subject: 15 easy, 56 medium, 23 hard, across 12 units.
10 practice questions, worked
Choose an answer, then open the working. Each explanation says why the right answer is right and why the others are not.
These 10 were written to ASRB NET and are not shown on any other exam's page.
Question 1 · medium
In the Midzuno-Sen scheme, how are units selected?
All by systematic sampling
First PPS, rest by SRS
All by PPS with replacement
First by SRS, rest PPS
Answer and working
B. First PPS, rest by SRS
This makes the ratio estimator unbiased.
Question 2 · easy
Clusters of 10 units have intracluster correlation 0.1. What is the design effect?
2.0
1.9
0.9
1.1
Answer and working
B. 1.9
1 + (M - 1) rho = 1 + 9 x 0.1 = 1.9.
Question 3 · medium
Which estimator of the mean per unit suits unequal clusters best?
Median cluster total
Ratio estimator
Largest cluster mean in the whole sample of clusters
Mean of cluster totals
Answer and working
B. Ratio estimator
The ratio of sampled cluster totals to sampled sizes stabilises estimates when totals vary with size.
Question 4 · medium
What is the double-sampling regression estimator?
y-bar x-bar'/x-bar^2
y-bar + b(x-bar' - x-bar)
y-bar - b(x-bar' - x-bar)
x-bar' + b y-bar
Answer and working
B. y-bar + b(x-bar' - x-bar)
x-bar' is the first-phase mean.
Question 5 · medium
What is rotation group bias?
Bias of the ratio estimator in small samples
Variance from clusters
Responses changing with time in sample
Bias from PPS selection
Answer and working
C. Responses changing with time in sample
A non-sampling error of rotation designs.
Question 6 · medium
What is a 4-8-4 rotation pattern?
Matched 4 of every 8
4 strata, 8 clusters, 4 units
4 occasions only
In 4 periods, out 8, in 4
Answer and working
D. In 4 periods, out 8, in 4
As in the US Current Population Survey.
Question 7 · medium
500 mailed; 300 respond with mean 40; 50 of 200 non-respondents interviewed with mean 30. What is the Hansen-Hurwitz estimate?
36
38
35
40
Answer and working
A. 36
(300 x 40 + 200 x 30)/500 = 36.
Question 8 · hard
Which model estimates each unit's response probability for weighting?
A Latin square model
A response-propensity model
A path model
A Hardy-Weinberg model
Answer and working
B. A response-propensity model
Usually logistic regression on frame variables.
Question 9 · medium
What do replicate weights on public-use files allow?
Exact population totals for every domain of interest
Selection of new samples
Removal of non-response
Correct variances without design details
Answer and working
D. Correct variances without design details
Users recompute estimates with each replicate weight.
Question 10 · hard
Why are ordinary estimators biased in adaptive cluster sampling?
Units are drawn with equal probability
The initial sample is not random
Inclusion depends on observed values
Neighbours are never added
Answer and working
C. Inclusion depends on observed values
Modified HT/HH estimators based on networks are unbiased.
A study plan for Sampling Theory II: Unequal Probability, Cluster, Multistage & Repeated Sampling in ASRB NET
This plan follows where our questions actually fall, so the time goes where the marks are likely to be. It is our reading of the syllabus, not the exam body's weighting.
First, the heavy units (36% of the 94 questions): Two-stage and multistage sampling (9, 10%); Variance estimation in complex surveys: linearisation, jackknife and bootstrap (9, 10%); Cluster sampling with equal and unequal clusters (8, 9%); Imputation methods (8, 9%).
Read the manual chapters for these first and do every worked question you can find on them.
Then the middle band: Non-response: Hansen-Hurwitz call-backs and weighting adjustments (8, 9%); PPS sampling with replacement: Hansen-Hurwitz estimator (8, 9%); Sampling on successive occasions and rotation designs (8, 9%); Sampling without replacement with unequal probabilities: Horvitz-Thompson estimator (8, 9%).
Aim for one sitting per unit, then mixed questions across them.
Then the lighter units: Des Raj and Murthy's estimators; Midzuno-Sen and Rao-Hartley-Cochran schemes (7, 7%); Double (two-phase) sampling for stratification and for ratio and regression (7, 7%); Inverse and adaptive sampling (introduction) (7, 7%); Randomised response techniques (7, 7%).
Learn the definitions, classifications and standard numbers; these units are more often recalled than worked.
Leave time for the hard ones: Des Raj and Murthy's estimators; Midzuno-Sen and Rao-Hartley-Cochran schemes (2 hard of 7), Double (two-phase) sampling for stratification and for ratio and regression (2 hard of 7), Inverse and adaptive sampling (introduction) (2 hard of 7).
Come back to these after a first pass, untimed, and read every explanation, including for the options you rejected.
Bank the easy marks: Cluster sampling with equal and unequal clusters (2 easy of 8); Imputation methods (2 easy of 8).
Do these early in revision so they are fast on the day.
Finish timed. Sit the subject as a timed set without notes, mark it, and send each wrong answer back to its unit above.