A student wants to estimate the mean of population for some sample. The probability is 0.95 that sample mean will not differ by true mean by more than 50% of standard deviation. \((Z_{0.025}=1.96)\) The size of sample taken by student is :
Answer & explanation
Correct answer: option 2
The correct answer is Option (2) → 15
Note: The given answer is as per the official NTA Answer Key. However, the answer given by NTA appears to be incorrect, and the correct answer should be 16, which is not among the options, as explained below.
The sample mean \(\overline{x}\) of a sample of size n has standard error \(\frac{\sigma}{\sqrt{n}}\). With probability 0.95 it lies within \(1.96\) standard errors of the true mean \(\mu\):
\( \displaystyle P\left(|\overline{x}-\mu| \le 1.96\,\frac{\sigma}{\sqrt{n}}\right) = 0.95 \)
The difference must not exceed 50% of the standard deviation, that is \(0.5\sigma\). So the 95% error bound must be at most \(0.5\sigma\):
\( \displaystyle 1.96\,\frac{\sigma}{\sqrt{n}} \le 0.5\,\sigma \)
\( \displaystyle \sqrt{n} \ge \frac{1.96}{0.5} = 3.92 \)
\(n \ge (3.92)^2 = 15.3664\)
A sample size is a whole number, so it must be rounded up: the smallest sample that meets the condition is n = 16. Check:
\( \displaystyle n = 16:\ \frac{1.96}{\sqrt{16}} = 0.49 \le 0.5 \)
\( \displaystyle n = 15:\ \frac{1.96}{\sqrt{15}} = 0.506 \gt 0.5 \)
Option (1) 17 — Incorrect. With 17 the error bound is \(\frac{1.96}{\sqrt{17}}\sigma = 0.475\sigma\), within \(0.5\sigma\), but 17 is not the size needed: 16 already meets the condition.
Option (2) 15 — The answer given by NTA, but it seems to be incorrect. With 15 the error bound is \(\frac{1.96}{\sqrt{15}}\sigma = 0.506\sigma\), which is more than \(0.5\sigma\), so the condition is not met. The value 15.37 must be rounded up to 16, not down to 15.
Option (3) 14 — Incorrect. With 14 the error bound is \(\frac{1.96}{\sqrt{14}}\sigma = 0.524\sigma\), more than \(0.5\sigma\).
Option (4) 13 — Incorrect. With 13 the error bound is \(\frac{1.96}{\sqrt{13}}\sigma = 0.544\sigma\), more than \(0.5\sigma\).
Advice for the exam: In the real exam, mark the answer that is mathematically correct. If NTA’s provisional answer key marks a different option, challenge it during the answer-key challenge window, giving proper reasoning. NTA does revise its key when a challenge is well justified, so a sound challenge can raise your marks.