GATE — Engineering Sciences (XE), Section G Food Technology · India · IITs and IISc on rotation (2027: IIT Madras)
Original practice questions written to the published syllabus · not past papersGATE — Engineering Sciences (XE), Section G Food Technology is a postgraduate admission exam run by IITs and IISc on rotation (2027: IIT Madras), for candidates in India. Computer-based, 3 hours, 100 marks. MCQ, MSQ and NAT. General Aptitude is common to every GATE paper; within XE, Section A Engineering Mathematics is compulsory (stated on the 2026 XE-A syllabus) and the candidate chooses two further sections, of which G Food Technology is one. The split of marks between GA, XE-A and the two chosen sections is from Laisha's knowledge of earlier brochures and was not re-read for 2026.
Source: gate2026.iitg.ac.in.
Checked against the exam body's published information. Formats and fees change; confirm the current edition there before you book.
Probability, inference and the designs used in food research and quality work: factorial and response-surface designs for formulation and process optimisation, mixture designs, the statistics of sensory panels, and multivariate methods (PCA, PLS) used with spectra and chromatograms. Goes further than the general statistics subjects in Agricultural Engineering (probability-statistics-data) and Agribusiness (business-statistics) by focusing on food formulation, sensory and analytical data.
Our 34 GATE XE questions in this subject: 12 easy, 17 medium, 5 hard, across 3 units.
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 GATE XE and are not shown on any other exam's page.
Four replicate determinations of sugar content gave 8, 10, 12 and 10 g per 100 g. What is the sample standard deviation (n − 1 denominator)?
C. 1.63 g per 100 g
Mean = 10; squared deviations 4, 0, 4, 0 sum to 8. Sample variance = 8/3 = 2.67; SD = √2.667 = 1.63. 1.41 uses n = 4 (the population formula), and 2.67 is the variance, not its square root.
Nine replicate fat determinations have a standard deviation of 1.2 g per 100 g. What is the standard error of the mean?
A. 0.40 g per 100 g
SE = s/√n = 1.2/√9 = 1.2/3 = 0.40. Dividing by n instead of √n gives 0.13; multiplying by √n gives 3.6; 1.2 is the standard deviation of single values, not of the mean.
In a triangle test the chance of a correct answer by guessing is 1/3. What is the probability that three independent assessors all pick the odd sample by chance alone?
A. 0.037
Independent events multiply: (1/3)³ = 1/27 = 0.037. 0.333 is one assessor; 0.111 is (1/3)², two assessors; 0.125 is (1/2)³, the chance level of a duo-trio or paired test, not a triangle test.
Five pH readings are 6.4, 6.5, 6.5, 6.6 and 9.8, the last from a faulty electrode. Which measure of location is least affected by the faulty value?
C. The median
The median is the middle value (6.5) and ignores how extreme the outlier is. The mean is pulled up to 7.16; range and standard deviation are measures of spread, not location, and both are inflated by the outlier.
Sixteen cans give a mean drained weight of 20.0 g per unit with s = 0.8 g. Using t(0.05, 15 df) = 2.131, what is the 95% confidence interval for the mean?
C. 19.57 to 20.43 g
SE = 0.8/√16 = 0.2; margin = 2.131 × 0.2 = 0.426; interval 19.57 to 20.43. Using z = 1.96 gives 19.61 to 20.39 (wrong with s estimated from 16 values); using s instead of the SE gives 18.30 to 21.70; dividing by n instead of √n gives an interval far too narrow.
A quality manager rejects a batch that in fact meets specification, because a test was significant by chance. Which error is this?
D. Type I error
Rejecting a true null hypothesis (the batch is fine) is a type I error, with probability α. A type II error is accepting a bad batch. Bias and drift are systematic errors, not errors of a test decision.
A product declares 3.0% fat. Nine units give a mean of 2.9% and s = 0.15%. Using a two-sided one-sample t test at α = 0.05 (critical t, 8 df = 2.306), what is the result?
D. t = -2.0; not significant, since |t| < 2.306
t = (2.9 − 3.0)/(0.15/√9) = −0.1/0.05 = -2.0. With 8 df the critical value is 2.306, so the difference is not significant. Using 1.96 applies the large-sample z value to n = 9; −0.67 divides by s instead of s/√n; −6.0 divides by s/n.
Panel scores and an instrument reading rise together but along a curve, and two samples are extreme. Which correlation measure suits this relationship best?
C. Spearman's rank correlation
Spearman's coefficient works on ranks, measures any monotonic relationship and is robust to outliers. Pearson's r measures only linear association and is sensitive to extreme points; the CV and covariance are not correlation measures on a −1 to +1 scale.
A calibration line gives r = 0.995, but the residuals form a clear U-shaped pattern against concentration. What should the analyst conclude?
A. The response is curved; high r does not prove linearity, so refit or narrow the range
A systematic residual pattern shows lack of fit; r can be very high for curved data. The remedy is a curved (for example quadratic) model or a narrower linear range. Replicates do not remove systematic curvature, and forcing zero intercept usually worsens it.
Five samples have sugar levels x = 1, 2, 3, 4, 5 and sweetness scores y = 3, 5, 4, 7, 6. What is Pearson's r?
B. 0.80
Mean x = 3, mean y = 5. Sxy = 8, Sxx = 10, Syy = 10. r = 8/√(10 × 10) = 0.80. 0.64 is r², the coefficient of determination; 0.89 is the square root of r, a slip; 1.00 would need all points on a straight line.
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.
Read the manual chapters for these first and do every worked question you can find on them.
Aim for one sitting per unit, then mixed questions across them.
Learn the definitions, classifications and standard numbers; these units are more often recalled than worked.
Come back to these after a first pass, untimed, and read every explanation, including for the options you rejected.
Do these early in revision so they are fast on the day.
Last updated · How Laisha writes and checks its pages · Report a mistake
US$10 a month · one price worldwide · cancel any time
| Free | Smarter | |
|---|---|---|
| Grow guides, pest pages, articles | Read all | Read all |
| Ask Laisha | 5 a day | Up to 1,000 a month |
| Exam practice | The sample on each page | Every question, timed papers |
| Crop calendar and pest alerts | — | For your own fields |
| Calculators and tools | — | All |
| Uploads and reports | — | 100 a month, real files |
An AI operating system for agriculture: free guides, manuals and tools written for every country.