LAISHA
Open Laisha

ASRB NET Principles of Experimental Design & Basic Designs: 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
65written 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).

Official site: asrb.org.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.

Principles of Experimental Design & Basic Designs in ASRB NET

The same Unit 4 content with applied judgment: which design for which field, which comparison procedure for which question, how to read an ANOVA.

Fisher's three principles and how they shape a field or animal trial: uniformity trials, plot size and shape, the CRD, RCBD and Latin square with their analysis and efficiency, missing plots, transformations, multiple comparisons and analysis of covariance.

Our 65 ASRB NET questions in this subject: 2 easy, 39 medium, 24 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 a feeding trial each ration is given to a pen of 10 birds and weight gain is recorded per bird. Which is the experimental unit?

  1. The individual bird, because each bird is weighed separately
  2. The shed, because all pens share the same housing conditions
  3. The pen, because the ration is allocated to whole pens
  4. The ration, because it is the factor under study in the trial
Answer and working

C. The pen, because the ration is allocated to whole pens

The experimental unit is the smallest unit to which a treatment is independently applied; rations go to pens, so pens are the replicates. Birds within a pen are sampling units. The ration is the treatment, not the unit, and the shed is a possible block.

Question 2 · hard

Which statement about randomisation is correct?

  1. It guarantees that every layout drawn will be free of any fertility trend
  2. It is needed only in a CRD, because blocks already remove soil bias
  3. It can be replaced by a systematic order once the field is well levelled
  4. It justifies the tests over all layouts, even if one layout is poor
Answer and working

D. It justifies the tests over all layouts, even if one layout is poor

Randomisation makes the procedure unbiased on average and gives the error its validity; a particular random layout can still be unlucky. It is needed within blocks of an RCBD and in rows and columns of a Latin square too. A systematic order can coincide with hidden gradients, which is exactly what randomisation guards against.

Question 3 · medium

On the same 20 plots, a CRD with 5 treatments has 15 error degrees of freedom, while an RCBD with 4 blocks has 12. When does the RCBD still give the more precise comparison?

  1. When blocks differ enough to cut the error mean square despite lost df
  2. Only when the number of blocks is larger than the number of treatments
  3. Only when the block mean square is smaller than the error mean square
  4. Never, because fewer error degrees of freedom always mean less precision
Answer and working

A. When blocks differ enough to cut the error mean square despite lost df

Blocking spends r - 1 = 3 df but removes block variation from the error; if blocks really differ, the reduction in MSE outweighs the slightly larger t value. If the block MS is no larger than the error MS, blocking gained nothing.

Question 4 · medium

How is a Latin square layout randomised?

  1. Randomise treatments within each row only, as in an RCBD with rows as blocks
  2. Pick a standard square at random, then randomise rows, columns and letters
  3. Randomise treatments over all t^2 plots, then move repeats to empty rows
  4. Use a cyclic square in its standard order without any further shuffling
Answer and working

B. Pick a standard square at random, then randomise rows, columns and letters

Randomising a standard square by permuting rows, columns and the allocation of letters keeps the Latin property and gives every layout a chance. Free randomisation over all plots breaks it; the unshuffled cyclic square is systematic; within-row randomisation alone does not ensure each treatment once per column.

Question 5 · hard

An RCBD with 4 blocks and 5 treatments has block mean square 8.000 and error mean square 3.1667. Ignoring the degrees-of-freedom adjustment, what is its efficiency relative to a CRD?

  1. 393.0%
  2. 252.6%
  3. 124.1%
  4. 132.1%
Answer and working

C. 124.1%

RE = [(r - 1)Eb + r(t - 1)Ee] / [(rt - 1)Ee] = [3 x 8.000 + 16 x 3.1667] / [19 x 3.1667] = 74.667 / 60.167 = 1.241, i.e. 124.1%. The simple ratio Eb/Ee overstates it (252.6%); 132.1% swaps r and t.

Question 6 · hard

In an RCBD with 5 treatments and 4 blocks one plot was missing and estimated; the error mean square is 3.0. What is the standard error of the difference between the mean of the treatment with the missing plot and the mean of another treatment?

  1. 1.031
  2. 1.225
  3. 1.346
  4. 1.323
Answer and working

C. 1.346

SEd = sqrt(MSE [2/r + t / (r(r - 1)(t - 1))]) = sqrt(3.0 x [0.5 + 5/48]) = sqrt(1.8125) = 1.346. The usual sqrt(2 MSE/r) = 1.225 ignores the estimation; treating the design as a CRD with 3 and 4 replicates gives 1.323; using 1/r instead of 2/r gives 1.031.

Question 7 · medium

Two plots are missing in different blocks and treatments of an RCBD. How are their values usually found by hand?

  1. By iteration or simultaneous equations, losing 2 error df
  2. Each by the single-value formula once, ignoring the other gap
  3. By inserting treatment means and keeping the full error df
  4. By the Latin-square formula, since two gaps act like two directions
Answer and working

A. By iteration or simultaneous equations, losing 2 error df

Each estimate depends on the other, so guess one, estimate the other with the single-value formula, re-estimate the first and repeat until stable (or solve the two equations together). One error df is lost per missing value.

Question 8 · medium

Insect counts range from 5 to 2000 across treatments and the standard deviations are roughly proportional to the means. Which transformation is indicated?

  1. Logarithm, as log(x + 1) if zeros occur
  2. Square root, as the square root of (x + 0.5)
  3. Arcsine of the square root of x
  4. No transformation, as the range is so wide
Answer and working

A. Logarithm, as log(x + 1) if zeros occur

When SD is proportional to the mean (constant CV), the log transformation equalises variances and turns multiplicative effects into additive ones. Square root suits Poisson-like data where variance, not SD, is proportional to the mean; arcsine is for proportions.

Question 9 · hard

Why is Bartlett's test not always preferred for checking equal variances in agricultural data?

  1. It is sensitive to non-normality, so Levene's test is often used instead
  2. It requires the data to be transformed to logarithms before it is run
  3. It tests the additivity of effects rather than the equality of variances
  4. It can only be applied when there are exactly two treatments to compare
Answer and working

A. It is sensitive to non-normality, so Levene's test is often used instead

Bartlett's test can signal unequal variances when the real problem is non-normal (e.g. heavy-tailed) data; Levene's test, based on absolute deviations, is more robust. Bartlett's test handles any number of groups and tests variances, not additivity.

Question 10 · medium

In a covariance analysis the error sums of squares and products are Eyy = 100, Exy = 30 and Exx = 60. What is the adjusted error sum of squares of Y?

  1. 99.5
  2. 91.0
  3. 85.0
  4. -20.0
Answer and working

C. 85.0

Adjusted Eyy = Eyy - Exy^2 / Exx = 100 - 900/60 = 85.0. Forgetting to square Exy gives 99.5; dividing by Eyy gives 91.0.

A study plan for Principles of Experimental Design & Basic Designs 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.

  1. First, the heavy units (46% of the 65 questions): Assumptions of ANOVA and variance-stabilising transformations (8, 12%); Multiple comparison procedures: LSD, Tukey, Duncan, Dunnett and Scheffe (8, 12%); Analysis of covariance (7, 11%); Latin square design and Graeco-Latin squares (7, 11%).

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

  2. Then the middle band: Missing plot techniques (7, 11%); Size and shape of plots and blocks; Smith's variance law (7, 11%); Relative efficiency of designs (6, 9%); Uniformity trials, fertility contour maps and soil heterogeneity (4, 6%).

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

  3. Then the lighter units: Principles: replication, randomisation and local control (3, 5%); Randomised complete block design (3, 5%); Sample size and number of replications (3, 5%); Completely randomised design (2, 3%).

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

  4. Leave time for the hard ones: Relative efficiency of designs (4 hard of 6), Sample size and number of replications (2 hard of 3), Completely randomised design (1 hard of 2).

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

  5. Bank the easy marks: Analysis of covariance (1 easy of 7); Multiple comparison procedures: LSD, Tukey, Duncan, Dunnett and Scheffe (1 easy of 8).

    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 — ASRB NET Principles of Experimental Design & Basic Designs

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.