ASRB Agricultural Research Service Agriculture for Statisticians: practice questions

ASRB Agricultural Research Service (ARS) - Agricultural Statistics scientist · India · Agricultural Scientists Recruitment Board (ICAR)

Original practice questions written to the published syllabus · not past papers

10worked here, free

27written to this exam in this subject

10syllabus units

What this subject covers

The agriculture a statistician must know to plan a trial, frame a crop survey or read a breeder's question: how crops, livestock and fisheries are produced, what pests and diseases do to yield, how agricultural research and statistics are organised, and the rural-development programmes surveys evaluate. It is Unit I of AIEEA-PG Group 07.

In the exams: AIEEA-PG Group 07 Unit I (heavy: the agriculture unit); ASRB ARS/NET interviews (present)

Importance of agriculture, forestry, livestock and fisheries in national economies
Basic principles of crop production: seasons, cropping systems and yield components
Soils, nutrients and fertiliser response
Major pests and diseases of crops and how they are measured
Livestock and fisheries production basics
Plant and animal breeding basics: what breeders measure and select
Elementary principles of agricultural economics and extension
Organisation of agricultural research, education and extension (CGIAR and national systems)
Rural development programmes and what they ask statisticians to measure
Agricultural units, conversion factors and crop calendars

Where the 27 questions fall

How our questions for ASRB Agricultural Research Service spread across the units of Agriculture for Statisticians. Difficulty: 0 easy, 18 medium, 9 hard.

Rural development programmes and what they ask statisticians to measure	7
Major pests and diseases of crops and how they are measured	4
Soils, nutrients and fertiliser response	3
Organisation of agricultural research, education and extension (CGIAR and national systems)	3
Importance of agriculture, forestry, livestock and fisheries in national economies	2
Livestock and fisheries production basics	2
Plant and animal breeding basics: what breeders measure and select	2
Elementary principles of agricultural economics and extension	2
Basic principles of crop production: seasons, cropping systems and yield components	1
Agricultural units, conversion factors and crop calendars	1

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.

1 · Importance of agriculture, forestry, livestock and fisheries in national economies · medium

Which feature of livestock makes annual livestock numbers harder to estimate than crop area?

Livestock cannot legally be counted by enumerators on ordinary household visits
Births, deaths, sales and slaughter change herds all year
Livestock numbers are fixed by law in most countries, so surveys add nothing
Animals are counted only at agricultural censuses held once every 25 years

Answer and working

B. Births, deaths, sales and slaughter change herds all year

Herd dynamics mean a count depends on the reference date, and flows (births, offtake) must be recorded separately; mobile pastoral herds add a sampling problem.

Further reading: Martin, Waldren and Stamp, Principles of Field Crop Production; Brady and Weil, The Nature and Properties of Soils; Falconer and Mackay, Introduction to Quantitative Genetics; FAO World Programme for the Census of Agriculture

2 · Soils, nutrients and fertiliser response · medium

Yield response to nitrogen is y = 2,000 + 30x − 0.1x² (kg/ha, x in kg N/ha). What is the maximum yield?

4,500 kg/ha
4,250 kg/ha
6,500 kg/ha
2,150 kg/ha

Answer and working

B. 4,250 kg/ha

Maximum at x = 30/(2 × 0.1) = 150 kg N/ha; y = 2,000 + 4,500 − 2,250 = 4,250 kg/ha. 6,500 forgets the quadratic term; 4,500 is the linear term alone.

3 · Major pests and diseases of crops and how they are measured · hard

Disease severity is 5%, 15%, 40% and 60% at days 0, 7, 14 and 21. What is the AUDPC by the trapezoidal rule?

612 %-days
120 %-days
840 %-days
60 %-days

Answer and working

A. 612 %-days

AUDPC = 7[(5 + 15)/2 + (15 + 40)/2 + (40 + 60)/2] = 7 × (10 + 27.5 + 50) = 612 %-days. 120 is the sum of severities; 840 multiplies that sum by 7 without averaging.

4 · Major pests and diseases of crops and how they are measured · hard

Control costs 1,500 currency units/ha; produce is worth 30 units/kg; each insect causes 0.5 units of injury; each unit of injury loses 0.4 kg/ha; control reduces injury by 80%. What is the EIL (insects per unit sample) by EIL = C/(VIDK)?

250.0
50.0
312.5
4.8

Answer and working

C. 312.5

EIL = 1,500/(30 × 0.5 × 0.4 × 0.8) = 1,500/4.8 = 312.5. Omitting K gives 1,500/6 = 250.0; 50.0 divides only by V; 4.8 is the denominator VIDK, not the EIL.

5 · Major pests and diseases of crops and how they are measured · medium

Which statistical feature of insect counts per plant usually requires a transformation or a count model before ANOVA?

Counts are always normally distributed with a constant variance across plots
Counts can never be zero, so their logarithm is always defined for analysis
Counts are ordinal scores whose gaps between categories are unknown
Clumping: variance exceeds the mean

Answer and working

D. Clumping: variance exceeds the mean

Aggregated pests give over-dispersed counts (variance > mean). Square-root or log(x + 1) transformations, or Poisson and negative binomial models, deal with this.

6 · Plant and animal breeding basics: what breeders measure and select · medium

Two varieties rank differently at two locations: A is best at the first, B at the second. What does this show?

High heritability, so testing at one location is enough for selection
Genotype-by-environment interaction, so testing at many locations is needed
Additive gene action, so the two varieties should now be crossed
Measurement error at one site, so both results should be discarded

Answer and working

B. Genotype-by-environment interaction, so testing at many locations is needed

Rank changes across environments are crossover G × E. Multi-location, multi-year trials and stability analysis address it.

7 · Rural development programmes and what they ask statisticians to measure · hard

A subsidy is given only to farms below 2 ha. Which design compares farms just below and just above that limit?

Propensity score matching
Randomised controlled trial
Instrumental variables
Regression discontinuity

Answer and working

D. Regression discontinuity

Farms just either side of the cut-off are similar except for eligibility, so the jump in outcomes at the threshold estimates the effect there.

8 · Rural development programmes and what they ask statisticians to measure · medium

Mexico's conditional cash transfer programme of the late 1990s is often cited in evaluation teaching. Why?

Its phased roll-out was randomised across villages, giving a credible control group
It was the first programme ever to be evaluated using only administrative records
It transferred agricultural land rather than cash to the poorest rural households
It was evaluated only by asking the participants their opinion of its benefits

Answer and working

A. Its phased roll-out was randomised across villages, giving a credible control group

Randomly choosing which eligible villages entered first allowed early and late villages to be compared, a model later used widely.

9 · Rural development programmes and what they ask statisticians to measure · hard

When planning the sample for a village-level programme evaluation, why must clustering be accounted for?

Clustering matters only for crop-cutting experiments
Clustering matters only for crop-cutting experiments, not household surveys
Households in a village are alike, so each adds less information
Clustering is removed entirely by surveying every household in each village

Answer and working

C. Households in a village are alike, so each adds less information

With intra-cluster correlation ρ and m households per village, the design effect 1 + (m − 1)ρ inflates variance, so power calculations must use it.

10 · Agricultural units, conversion factors and crop calendars · medium

Why should a crop-cutting yield survey be scheduled from the crop calendar of each region?

The crop calendar sets the price at which the harvested crop is valued
Sampling weights depend on the calendar month in which the survey is run
Maturity dates differ by region
The calendar shows in advance which farmers will refuse to cooperate

Answer and working

C. Maturity dates differ by region

Crop cutting needs mature crops; sowing and harvest dates differ with climate, altitude and season, so field timing follows regional calendars.

Ten here, 27 in the app. The Laisha app has every practice question we have written to ASRB Agricultural Research Service in this subject, marked as you go, with timed papers, a jump-to-unanswered list and a certificate when you pass at 70 per cent.

Free for ten days, then one subscription of US$10 a month. We would rather say so here than have you find it later.

Practise the full set in Laisha →