State statistical services: Statistical Officer / Assistant Statistical Officer (State PSCs and selection boards)
India · State Public Service Commissions and Subordinate Service Selection Boards · unverified source
681practice questions written to this syllabus
22subjects
What State statistical services is
State statistical services: Statistical Officer / Assistant Statistical Officer (State PSCs and selection boards) is a government recruitment exam run by State Public Service Commissions and Subordinate Service Selection Boards, for candidates in India.
Format
Varies by state: usually an objective screening test in statistics and general studies, some with a descriptive paper and interview.
Sections
Statistics (descriptive, probability, sampling, design of experiments, index numbers, official statistics)
Economics and general studies (varies)
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: mospi.gov.in.
How it has changed
Posts sit in each state's Directorate of Economics and Statistics, which runs the crop-cutting experiments, the agricultural census and land-use statistics at state level. Examples include the TNPSC Combined Statistical Subordinate Services exam and RPSC Statistical Officer recruitment; no single state's scheme was read.
How to prepare for State statistical services
Where our 681 questions for it fall
Our reading of the syllabus, from the questions we have written to this exam; the exam body's own weighting, where published, is under Format above.
Heaviest (66% of the questions): Statistical Methods for Agricultural Sciences (88); Statistical Quality Control & Reliability (84); Optimization Techniques for Statisticians (61); Statistical Inference II: Testing of Hypotheses & Sequential Analysis (46); Sampling Theory I: Simple, Stratified, Systematic, Ratio & Regression Estimation (44); Probability Distributions & Sampling Distributions (43); Demography & Vital Statistics (43); Probability Theory (40).
Start here: read these subjects' manuals first and work their questions until the reasoning is automatic.
Original questions written to this exam's published syllabus and pattern. Not past papers.
1. In a chi-square goodness-of-fit test of insect counts, several classes have expected frequencies below 5. What is the usual remedy?
Replace the chi-square test by a t-test on the class frequencies
Pool adjacent classes, then count df after pooling
Add 0.5 to every observed count and keep all of the classes
Drop those classes from the table and keep the original degrees of freedom
Answer and working
B. Pool adjacent classes, then count df after pooling
Small expected counts make the χ² approximation unreliable; adjacent classes are pooled and the df counted on the pooled classes. Dropping classes discards data and biases the test.
Nonparametric & Rank-Based Methods · Goodness of fit: chi-square and Kolmogorov-Smirnov tests
2. A 95% two-sided confidence interval for a mean difference is 0.1 to 0.5 t/ha. What does the corresponding 5% two-sided test conclude?
It rejects neither, because the interval does not include 0.6
It rejects both a difference of 0 and a difference of 0.3
It rejects a difference of 0 and does not reject a difference of 0.3
It rejects a difference of 0.3 and does not reject a difference of 0
Answer and working
C. It rejects a difference of 0 and does not reject a difference of 0.3
A two-sided level-α test rejects θ₀ exactly when θ₀ lies outside the 1 − α interval: 0 is outside, 0.3 inside.
Statistical Inference II: Testing of Hypotheses & Sequential Analysis · Relation between tests and confidence intervals
3. Why is there no UMP test of a one-sided hypothesis about the location θ of a Cauchy distribution based on one observation?
The Cauchy family has a parameter-dependent support (unlike the normal), which rules out all MP tests
The Cauchy distribution has no mean, so no test of its location can be constructed
The likelihood ratio f(x; θ₁)/f(x; θ₀) is not monotone in x, so the MP region depends on θ₁
Its Fisher information is zero, so every test has power equal to its size
Answer and working
C. The likelihood ratio f(x; θ₁)/f(x; θ₀) is not monotone in x, so the MP region depends on θ₁
f(x; θ₁)/f(x; θ₀) rises and then falls back to 1 as x → ±∞, so the MP region against θ₁ is a bounded interval whose location depends on θ₁; no single region is best for all θ₁ > θ₀. Tests do exist; the support is the whole line; information is positive.
Statistical Inference II: Testing of Hypotheses & Sequential Analysis · Monotone likelihood ratio and uniformly most powerful tests
4. A random sample of 16 is drawn from N(μ, 4). What is the variance of the sample variance s² (divisor n − 1)?
1.0667
2.1333
2.0000
30.0000
Answer and working
B. 2.1333
(n − 1)s²/σ² ~ χ²₁₅ with variance 30, so Var(s²) = σ⁴ × 2(n − 1)/(n − 1)² = 2σ⁴/(n − 1) = 32/15 = 2.1333. 30 is the variance of the chi-square itself; dividing by n gives 2.0000.
Probability Distributions & Sampling Distributions · Sampling distributions: chi-square, t and F and their properties
5. A CRD has 5 treatments each replicated 4 times. How many degrees of freedom does the error have?
15
12
19
16
Answer and working
A. 15
N = 20; total df 19, treatment df 4, error df = N − t = 15. 12 = (5 − 1)(4 − 1) is the RBD error df.
Statistical Methods for Agricultural Sciences · Basic principles of experimental design, ANOVA, CRD and RBD
6. The probability that a plant is infected is 0.3. In a random sample of 5 plants, what is the probability that exactly 2 are infected?