A random sample of $n = 16$ production batches from an active pharmaceutical ingredient synthesis yields a sample mean reaction yield of $\bar{x} = 88.50\%$ and a sample standard deviation of $s = 2.40\%$. Assuming the batch yields are normally distributed, the lower limit of the $95\%$ two-sided confidence interval for the true population mean yield $\mu$ is most nearly:
- A86.85%
- B87.22%correct
- C87.52%
- D88.10%
Show the solution
Step 1: Identify degrees of freedom for the sample: $$\nu = n - 1 = 16 - 1 = 15$$ Step 2: Find the critical Student's t-distribution value from the FE Handbook table for $95\%$ confidence ($1 - \alpha = 0.95 \implies \alpha/2 = 0.025$): $$t_{0.025, 15} = 2.131$$ Step 3: Calculate the standard error of the mean: $$SE = \dfrac{s}{\sqrt{n}} = \dfrac{2.40\%}{\sqrt{16}} = \dfrac{2.40\%}{4.0} = 0.600\%$$ Step 4: Compute the margin of error and the lower confidence limit: $$E = t_{\alpha/2, \nu} \times SE = 2.131 \times 0.600\% = 1.2786\% \approx 1.28\%$$ $$\mu_{lower} = \bar{x} - E = 88.50\% - 1.28\% = 87.22\%$$ Distractor Analysis: - Option A (86.85%) uses the standard normal critical value $z = 1.96$ but multiplies by $s$ without dividing by $\sqrt{n}$. - Option C (87.52%) uses the standard normal distribution $z_{0.025} = 1.960$ instead of Student's $t$ ($88.50 - 1.96 \times 0.60 = 87.32$). - Option D (88.10%) uses a one-sided confidence interval with $t_{0.05, 15} = 1.753$.