LAISHA
Open Laisha

SSC Combined Graduate Level Statistical Inference II: Testing of Hypotheses & Sequential Analysis: practice questions

SSC Combined Graduate Level - Junior Statistical Officer / Statistical Investigator · India · Staff Selection Commission

Original practice questions written to the published syllabus · not past papers
10worked here, free
44written to this exam in this subject
8syllabus units

About SSC Combined Graduate Level

SSC Combined Graduate Level - Junior Statistical Officer / Statistical Investigator is a government recruitment exam run by Staff Selection Commission, for candidates in India. Multi-tier computer-based exam; candidates for Junior Statistical Officer and Statistical Investigator posts take an additional statistics paper.

Official site: ssc.gov.in.

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.

Statistical Inference II: Testing of Hypotheses & Sequential Analysis in SSC Combined Graduate Level

The logic and optimality of tests: errors and power, the Neyman-Pearson lemma, uniformly most powerful and unbiased tests, likelihood-ratio tests and their large-sample forms, and Wald's sequential probability ratio test.

Our 44 SSC Combined Graduate Level questions in this subject: 26 easy, 16 medium, 2 hard, across 8 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 SSC Combined Graduate Level and are not shown on any other exam's page.

Question 1 · easy

A survey officer tests H0: the mean household size of a district is 4.5. Rejecting H0 when it is in fact true is called:

  1. A Type II error
  2. A Type I error
  3. The power of the test
  4. The p-value
Answer and working

B. A Type I error

A Type I error is rejecting a true null hypothesis; its probability is the size (level) of the test, alpha. A Type II error is accepting a false null. Power is 1 minus the Type II error probability, and the p-value is a statistic computed from the sample, not a kind of error.

Question 2 · medium

For a fixed sample size, if the level of significance of a test is reduced from 5% to 1%, the probability of a Type II error will:

  1. Decrease
  2. Stay the same
  3. Increase
  4. Become zero
Answer and working

C. Increase

With n fixed, a smaller alpha shrinks the rejection region, so a false null is accepted more often: beta rises and power falls. The only way to lower both errors together is to increase the sample size.

Question 3 · easy

A sample of 16 crop-cutting plots gives a mean yield of 32 q/ha with sample standard deviation 4 q/ha. To test whether the population mean is 30 q/ha, the test statistic is:

  1. t = 2.0 with 15 degrees of freedom
  2. Z = 0.5
  3. t = 2.0 with 16 degrees of freedom
  4. Chi-square = 2.0 with 15 degrees of freedom
Answer and working

A. t = 2.0 with 15 degrees of freedom

With sigma unknown and a small sample, t = (x̄ − μ0)/(s/√n) = (32 − 30)/(4/4) = 2.0, on n − 1 = 15 degrees of freedom. A Z test would need the population standard deviation; 16 degrees of freedom forgets that one is used up by the sample mean.

Question 4 · medium

In a 3 × 4 contingency table tested for independence with the chi-square test, the degrees of freedom are:

  1. 12
  2. 11
  3. 7
  4. 6
Answer and working

D. 6

Degrees of freedom for independence in an r × c table are (r − 1)(c − 1) = 2 × 3 = 6. Twelve is the number of cells and eleven is cells minus one, the figure for a goodness-of-fit test with no estimated parameters.

Question 5 · medium

The F test is used to compare two independent sample variances. The ratio is formed so that:

  1. The smaller variance is in the numerator
  2. The larger sample variance is in the numerator
  3. The sample with fewer observations is in the numerator
  4. Both variances are first standardised to 1
Answer and working

B. The larger sample variance is in the numerator

By convention F = larger s² / smaller s², so the observed F is at least 1 and is compared with the upper tail of the F table, with numerator degrees of freedom from the larger-variance sample. Sample size does not decide which variance goes on top.

Question 6 · hard

Two large samples of 400 and 600 workers give proportions of literacy 0.40 and 0.45. Using the pooled proportion, the standard error of the difference is closest to:

  1. 0.032
  2. 0.025
  3. 0.045
  4. 0.010
Answer and working

A. 0.032

Pooled p = (160 + 270)/1000 = 0.43, q = 0.57. SE = √[pq(1/400 + 1/600)] = √[0.2451 × 0.004167] = √0.001021 ≈ 0.032. The observed difference 0.05 gives Z ≈ 1.56, not significant at 5% two-sided.

Question 7 · easy

A 95% confidence interval for a population mean is (48, 56). At the 5% level, a two-sided test of H0: μ = 50 will:

  1. Reject H0
  2. Need the sample size before deciding
  3. Not reject H0
  4. Reject H0 only if the sample is large
Answer and working

C. Not reject H0

A two-sided level-α test rejects exactly those values of μ0 that lie outside the (1 − α) confidence interval. 50 lies inside (48, 56), so H0 is not rejected. A value such as 58 would be rejected.

Question 8 · medium

Other things equal, which change increases the power of a test of a mean?

  1. Reducing the sample size
  2. Using a smaller level of significance
  3. Increasing the population variance
  4. Increasing the sample size
Answer and working

D. Increasing the sample size

A larger n shrinks the standard error, so a given true difference lies further into the rejection region and is detected more often. A smaller alpha, a smaller n or a larger variance all reduce power.

Question 9 · easy

A test result is reported as 'p = 0.03'. At the 5% level this means:

  1. H0 is rejected
  2. The probability that H0 is true is 3%
  3. H0 is accepted
  4. The power of the test is 97%
Answer and working

A. H0 is rejected

Reject H0 when p ≤ α; 0.03 < 0.05. The p-value is the probability, computed assuming H0 true, of a result at least as extreme as the one observed. It is not the probability that H0 is true, and it says nothing directly about power.

Question 10 · hard

Ten students are tested before and after a coaching programme. The mean of the ten differences is 3 marks and the standard deviation of the differences (divisor n − 1) is 3. The paired t statistic is:

  1. 1.00
  2. 3.16
  3. 9.49
  4. 0.95
Answer and working

B. 3.16

Paired t = d̄ / (s_d / √n) = 3 / (3/√10) = √10 ≈ 3.16, with 9 degrees of freedom. Treating the before and after scores as two independent samples would ignore the pairing and usually lose power.

A study plan for Statistical Inference II: Testing of Hypotheses & Sequential Analysis in SSC Combined Graduate Level

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.

  1. First, the heavy units (75% of the 44 questions): Tests based on Z, t, chi-square and F distributions (17, 39%); Hypotheses, the two types of error, size and power (9, 20%); Power and sample-size calculation for common tests (7, 16%).

    Read the manual chapters for these first and do every worked question you can find on them.

  2. Then the middle band: Likelihood ratio tests and Wilks' theorem (3, 7%); Multiple testing: family-wise error rate and false discovery rate (3, 7%); Neyman-Pearson lemma and most powerful tests (2, 5%).

    Aim for one sitting per unit, then mixed questions across them.

  3. Then the lighter units: Relation between tests and confidence intervals (2, 5%); Wald and score (Rao) tests (1, 2%).

    Learn the definitions, classifications and standard numbers; these units are more often recalled than worked.

  4. Leave time for the hard ones: Tests based on Z, t, chi-square and F distributions (2 hard of 17).

    Come back to these after a first pass, untimed, and read every explanation, including for the options you rejected.

  5. Bank the easy marks: Likelihood ratio tests and Wilks' theorem (3 easy of 3); Multiple testing: family-wise error rate and false discovery rate (3 easy of 3).

    Do these early in revision so they are fast on the day.

  6. Finish timed. Sit the subject as a timed set without notes, mark it, and send each wrong answer back to its unit above.

Last updated · How Laisha writes and checks its pages · Report a mistake

Ask Laisha about this

Opens Laisha's chat with your question ready to send. Sign in or start free.

Unlock every question — SSC Combined Graduate Level Statistical Inference II: Testing of Hypotheses & Sequential A

US$10 a month · one price worldwide · cancel any time

  • Every practice question in this subject, with full working
  • Timed exam papers and progress by unit
  • Ask Laisha to explain any answer — up to 1,000 questions a month
Free or Smarter
FreeSmarter
Grow guides, pest pages, articlesRead allRead all
Ask Laisha5 a dayUp to 1,000 a month
Exam practiceThe sample on each pageEvery question, timed papers
Crop calendar and pest alerts—For your own fields
Calculators and tools—All
Uploads and reports—100 a month, real files
10 days free for every new account, no card.