Resources

A manufacturing firm analyzes the diameter of precision bearings using a sample of size n=16n = 16. The sample mean is xˉ=5.032 mm\bar{x} = 5.032\text{ mm} and the sample standard deviation is s=0.060 mms = 0.060\text{ mm}. The population distribution of diameters is approximately normal. (a) Construct a 95%95\% two-sided confidence interval for the true mean diameter μ\mu using the appropriate critical tt-value (t∗=2.131t^* = 2.131 with df=15df = 15). (b) Compute the sample tt-statistic for the two-sided test H0:μ=5.000 mmH_0: \mu = 5.000\text{ mm} versus Ha:μ≠5.000 mmH_a: \mu \neq 5.000\text{ mm}. (c) Using the duality between two-sided tests and confidence intervals, state the formal hypothesis testing decision at α=0.05\alpha = 0.05, and specify whether the two-tailed PP-value is greater than or less than 0.050.05.

Answer and steps

Final answer

(a) 95%95\% Confidence Interval: [5.000 mm,5.064 mm][5.000\text{ mm}, 5.064\text{ mm}]; (b) Test statistic: t=2.133t = 2.133; (c) Because the hypothesized mean μ0=5.000 mm\mu_0 = 5.000\text{ mm} lies directly on the lower boundary of the 95%95\% confidence interval (and ∣t∣=2.133>2.131|t| = 2.133 > 2.131), reject H0H_0 at α=0.05\alpha = 0.05, which implies that the two-tailed PP-value satisfies P<0.05P < 0.05 (specifically P≈0.0498P \approx 0.0498).

Steps

Part (a): Construct the 95%95\% Confidence Interval

  1. Calculate the Standard Error (SE\text{SE}): The standard error of the sample mean is: SE=sn=0.06016=0.0604=0.015 mm\text{SE} = \frac{s}{\sqrt{n}} = \frac{0.060}{\sqrt{16}} = \frac{0.060}{4} = 0.015\text{ mm}
  2. Compute the Margin of Error (ME\text{ME}): Using the critical value t∗=2.131t^* = 2.131 for degrees of freedom df=n−1=15df = n - 1 = 15: ME=t∗⋅SE=2.131⋅0.015=0.031965 mm≈0.032 mm\text{ME} = t^* \cdot \text{SE} = 2.131 \cdot 0.015 = 0.031965\text{ mm} \approx 0.032\text{ mm}
  3. Determine the Confidence Interval: CI=xˉ±ME=5.032±0.032=[5.000 mm,5.064 mm]\text{CI} = \bar{x} \pm \text{ME} = 5.032 \pm 0.032 = [5.000\text{ mm}, 5.064\text{ mm}]

Part (b): Compute the Sample tt-Statistic

For testing H0:μ=5.000 mmH_0: \mu = 5.000\text{ mm} versus Ha:μ≠5.000 mmH_a: \mu \neq 5.000\text{ mm}, the test statistic measures how many standard errors the sample mean is from the hypothesized mean μ0\mu_0: t=xˉ−μ0SE=5.032−5.0000.015=0.0320.015≈2.133t = \frac{\bar{x} - \mu_0}{\text{SE}} = \frac{5.032 - 5.000}{0.015} = \frac{0.032}{0.015} \approx 2.133


Part (c): Hypothesis Testing Decision and PP-Value via Duality

  1. Duality Between Confidence Intervals and Two-Sided Tests: A two-sided hypothesis test at significance level α\alpha will reject H0:μ=μ0H_0: \mu = \mu_0 if and only if μ0\mu_0 lies outside the (1−α)×100%(1 - \alpha) \times 100\% confidence interval.

    Using exact unrounded figures: Lower Limit=5.032−(2.131)(0.015)=5.000035 mm\text{Lower Limit} = 5.032 - (2.131)(0.015) = 5.000035\text{ mm} Because μ0=5.000 mm<5.000035 mm\mu_0 = 5.000\text{ mm} < 5.000035\text{ mm}, the hypothesized mean falls just outside the interval.

  2. Decision via Critical Value and PP-Value:

    • Comparing the test statistic to the critical value: ∣t∣=2.133>t∗=2.131|t| = 2.133 > t^* = 2.131 Because the test statistic lies inside the rejection region, we reject H0H_0 at α=0.05\alpha = 0.05.
    • Because ∣t∣>t∗|t| > t^*, the area in the tails beyond ±2.133\pm 2.133 is smaller than the area beyond ±2.131\pm 2.131 (α=0.05\alpha = 0.05). Therefore, the two-tailed PP-value satisfies: P<0.05(P≈0.0498)P < 0.05 \quad (P \approx 0.0498)

Typical errors

OverviewACEpath Teacher's Diagnosis

01 The student forgets to divide the sample standard deviation by the square root of n when computing the margin of error, using s directly instead of the standard error.

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Partially correct
  1. Step 1

    ME=t∗×s=2.131×0.060=0.12786  ⟹  CI=5.032±0.128=(4.904,5.160) mm\text{ME} = t^* \times s = 2.131 \times 0.060 = 0.12786 \implies \text{CI} = 5.032 \pm 0.128 = (4.904, 5.160)\text{ mm}

    Correct

    ME=t∗⋅sn=2.131⋅0.015≈0.032 mm  ⟹  [5.000 mm,5.064 mm]\text{ME} = t^* \cdot \frac{s}{\sqrt{n}} = 2.131 \cdot 0.015 \approx 0.032\text{ mm} \implies [5.000\text{ mm}, 5.064\text{ mm}]

    You calculated the margin of error using the sample standard deviation ss directly (t∗×st^* \times s) rather than multiplying t∗t^* by the standard error of the mean, SE=sn\text{SE} = \frac{s}{\sqrt{n}}.

  2. Step 2

    SE=0.06016=0.015  ⟹  t=5.032−5.0000.015=2.133\text{SE} = \frac{0.060}{\sqrt{16}} = 0.015 \implies t = \frac{5.032 - 5.000}{0.015} = 2.133

  3. Step 3

    5.000 is inside (4.904,5.160), so fail to reject H0 at α=0.05  ⟹  P-value>0.055.000\text{ is inside } (4.904, 5.160)\text{, so fail to reject } H_0 \text{ at } \alpha = 0.05 \implies P\text{-value} > 0.05

    Correct

    μ0=5.000 mm\mu_0 = 5.000\text{ mm} lies outside the unrounded 95%95\% CI [5.000035,5.063965][5.000035, 5.063965], and ∣t∣=2.133>2.131|t| = 2.133 > 2.131, so reject H0H_0 at α=0.05  ⟹  P<0.05\alpha = 0.05 \implies P < 0.05

    Your reasoning logically followed your interval from Part (a), but because that interval was erroneously wide, you reached the opposite conclusion: since 5.000 mm5.000\text{ mm} actually falls outside the true 95%95\% confidence interval, you must reject H0H_0 and conclude that P<0.05P < 0.05.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly identified the formula for standard error when computing the test statistic.

You calculated the sample tt-statistic value accurately as t=2.133t = 2.133.

You understood the theoretical connection between confidence intervals and two-sided hypothesis tests.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Formula Substitution Error

no

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You conflated the sample standard deviation ss, which measures individual variation in the population, with the standard error of the sample mean sn\frac{s}{\sqrt{n}}, which measures the sampling variability of xˉ\bar{x}.

REASONING PATH
Your ACEpath teacher’s suggested approach
Margin of Error vs. Standard Deviation
  • For a sample mean confidence interval, ME=t∗⋅sn\text{ME} = t^* \cdot \frac{s}{\sqrt{n}}, never t∗⋅st^* \cdot s.

  • Always compute SE=sn\text{SE} = \frac{s}{\sqrt{n}} first before computing the margin of error.

Duality Consistency Check
  • A test statistic ∣t∣>t∗|t| > t^* must lead to the same rejection decision as checking if μ0\mu_0 lies outside the (1−α)(1 - \alpha) confidence interval.

  • If your test statistic ∣t∣=2.133>2.131|t| = 2.133 > 2.131 suggests rejecting H0H_0, but your confidence interval suggests failing to reject, recheck your confidence interval width immediately.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Compute the standard error of the mean: SE=sn=0.06016=0.015 mm\text{SE} = \frac{s}{\sqrt{n}} = \frac{0.060}{\sqrt{16}} = 0.015\text{ mm}.

  2. Compute the margin of error: ME=t∗⋅SE=2.131×0.015=0.031965≈0.032 mm\text{ME} = t^* \cdot \text{SE} = 2.131 \times 0.015 = 0.031965 \approx 0.032\text{ mm}.

  3. Construct the 95%95\% confidence interval: xˉ±ME=5.032±0.032=[5.000 mm,5.064 mm]\bar{x} \pm \text{ME} = 5.032 \pm 0.032 = [5.000\text{ mm}, 5.064\text{ mm}] (exact lower bound is 5.000035 mm5.000035\text{ mm}).

  4. Apply duality: Since μ0=5.000 mm<5.000035 mm\mu_0 = 5.000\text{ mm} < 5.000035\text{ mm}, it lies outside the 95%95\% confidence interval (consistent with ∣t∣=2.133>2.131|t| = 2.133 > 2.131).

  5. State conclusion: Reject H0H_0 at α=0.05\alpha = 0.05, which implies P<0.05P < 0.05.

Final answer

(a) [5.000 mm,5.064 mm][5.000\text{ mm}, 5.064\text{ mm}]; (b) t=2.133t = 2.133; (c) Reject H0H_0 at α=0.05\alpha = 0.05; P<0.05P < 0.05.

NEXT STEPS
Final Takeaways from Your Teacher
  • Write down SE=sn\text{SE} = \frac{s}{\sqrt{n}} as the very first step in any problem involving sample means.

  • Compare the conclusion from the test statistic to the conclusion from the confidence interval whenever both are requested to ensure they agree.

OverviewACEpath Teacher's Diagnosis

02 The student misinterprets the duality rule, believing that a test statistic lying beyond critical limits or a null value outside the interval means failing to reject the null hypothesis.

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Partially correct
  1. Step 1

    SE=0.06016=0.015  ⟹  ME=2.131×0.015=0.031965  ⟹  CI=(5.000035,5.063965) mm\text{SE} = \frac{0.060}{\sqrt{16}} = 0.015 \implies \text{ME} = 2.131 \times 0.015 = 0.031965 \implies \text{CI} = (5.000035, 5.063965)\text{ mm}

  2. Step 2

    t=5.032−5.0000.015=2.133t = \frac{5.032 - 5.000}{0.015} = 2.133

  3. Step 3

    5.000 is outside the 95% confidence interval.  ⟹  Fail to reject H0 at α=0.05.5.000\text{ is outside the } 95\% \text{ confidence interval.} \implies \text{Fail to reject } H_0 \text{ at } \alpha = 0.05.

    Correct

    μ0=5.000 lies outside the 95% CI [5.000035,5.063965]  ⟹  Reject H0 at α=0.05\mu_0 = 5.000\text{ lies outside the } 95\%\text{ CI } [5.000035, 5.063965] \implies \text{Reject } H_0 \text{ at } \alpha = 0.05

    A (1−α)×100%(1 - \alpha) \times 100\% confidence interval contains all plausible values of μ\mu that would not be rejected. Because the hypothesized value μ0=5.000\mu_0 = 5.000 lies outside the 95%95\% confidence interval, it is statistically implausible at the α=0.05\alpha = 0.05 level, so you must reject H0H_0.

  4. Step 4

    P-value>0.05P\text{-value} > 0.05

    Correct

    P-value<0.05P\text{-value} < 0.05

    Because H0H_0 is rejected at α=0.05\alpha = 0.05 (or equivalently, because ∣t∣=2.133>t∗=2.131|t| = 2.133 > t^* = 2.131), the tail area beyond the observed statistic is smaller than the critical area, meaning P<0.05P < 0.05.

WHAT YOU DID WELL
Compliments from Your Teacher

You calculated the standard error and margin of error with full precision.

You set up and computed the correct sample tt-statistic.

You correctly observed that the hypothesized null value 5.0005.000 falls outside the constructed interval.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Confidence interval and hypothesis test duality

no

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You reversed the logical connection between confidence intervals and two-sided hypothesis tests, treating values outside the interval as acceptable rather than rejectable.

REASONING PATH
Your ACEpath teacher’s suggested approach
Duality Rule
  • For a two-sided test at level α\alpha, reject H0:μ=μ0H_0: \mu = \mu_0 if and only if μ0\mu_0 falls outside the (1−α)(1 - \alpha) confidence interval.

  • Check whether μ0∈CI\mu_0 \in \text{CI}. If μ0∉CI\mu_0 \notin \text{CI}, reject H0H_0 and conclude P<αP < \alpha.

Test Statistic vs Critical Value
  • If ∣t∣>t∗|t| > t^*, the sample statistic lies deeper in the rejection region than the critical cutoff.

  • Verify your duality conclusion by comparing ∣t∣|t| to t∗t^*: since 2.133>2.1312.133 > 2.131, the tail probability is strictly less than 0.050.05.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Evaluate the position of μ0=5.000\mu_0 = 5.000 relative to the interval: μ0<5.000035\mu_0 < 5.000035, so it lies strictly outside the 95%95\% confidence interval.

  2. Apply the duality principle: an hypothesized value outside the 95%95\% confidence interval is not plausible at α=0.05\alpha = 0.05, so reject H0H_0.

  3. Determine the PP-value inequality: rejecting H0H_0 at α=0.05\alpha = 0.05 implies P<0.05P < 0.05.

Final answer

(a) [5.000 mm,5.064 mm][5.000\text{ mm}, 5.064\text{ mm}] (exact unrounded limits: [5.000035,5.063965][5.000035, 5.063965]); (b) t=2.133t = 2.133; (c) Reject H0H_0 at α=0.05\alpha = 0.05; P-value<0.05P\text{-value} < 0.05.

NEXT STEPS
Final Takeaways from Your Teacher
  • Memorize the core equivalence: 'Inside CI   ⟺  \iff Plausible   ⟺  \iff Fail to reject H0H_0 (P≥αP \ge \alpha)' and 'Outside CI   ⟺  \iff Implausible   ⟺  \iff Reject H0H_0 (P<αP < \alpha)'.

  • Cross-check your decision by comparing the absolute test statistic ∣t∣|t| to the critical value t∗t^*: when ∣t∣>t∗|t| > t^*, H0H_0 is always rejected.

  • Practice matching confidence intervals directly to two-sided test decisions across varied significance levels.

OverviewACEpath Teacher's Diagnosis

03 The student divides s by n rather than √n when calculating the standard error of the mean.

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Incorrect
  1. Step 1

    SE=sn=0.06016=0.00375\text{SE} = \frac{s}{n} = \frac{0.060}{16} = 0.00375

    Correct

    SE=sn=0.06016=0.0604=0.015 mm\text{SE} = \frac{s}{\sqrt{n}} = \frac{0.060}{\sqrt{16}} = \frac{0.060}{4} = 0.015\text{ mm}

    You divided the sample standard deviation by the sample size nn rather than its square root n\sqrt{n}.

  2. Step 2

    ME=2.131×0.00375=0.00799  ⟹  CI=5.032±0.008=(5.024,5.040) mm\text{ME} = 2.131 \times 0.00375 = 0.00799 \implies \text{CI} = 5.032 \pm 0.008 = (5.024, 5.040)\text{ mm}

    Correct

    ME=2.131×0.015≈0.032 mm  ⟹  CI=[5.000 mm,5.064 mm]\text{ME} = 2.131 \times 0.015 \approx 0.032\text{ mm} \implies \text{CI} = [5.000\text{ mm}, 5.064\text{ mm}]

    Your margin of error and confidence interval calculation followed the correct formula xˉ±t∗⋅SE\bar{x} \pm t^* \cdot \text{SE}, but inherited the incorrect standard error value from Step 1.

  3. Step 3

    t=5.032−5.0000.00375=0.0320.00375=8.533t = \frac{5.032 - 5.000}{0.00375} = \frac{0.032}{0.00375} = 8.533

    Correct

    t=xˉ−μ0SE=5.032−5.0000.015≈2.133t = \frac{\bar{x} - \mu_0}{\text{SE}} = \frac{5.032 - 5.000}{0.015} \approx 2.133

    You set up the ratio for the sample tt-statistic correctly as xˉ−μ0SE\frac{\bar{x} - \mu_0}{\text{SE}}, but substituted the incorrect standard error.

  4. Step 4

    5.000 is outside (5.024,5.040), so reject H0 at α=0.05. P-value<0.055.000\text{ is outside }(5.024, 5.040)\text{, so reject }H_0\text{ at }\alpha = 0.05\text{. } P\text{-value} < 0.05

    Correct

    Because μ0=5.000 mm<5.000035 mm\mu_0 = 5.000\text{ mm} < 5.000035\text{ mm} lies just outside the interval (or ∣t∣=2.133>2.131|t| = 2.133 > 2.131), reject H0H_0 at α=0.05\alpha = 0.05, which implies P<0.05P < 0.05.

    Your reasoning about interval duality and the PP-value relationship was conceptually sound, though based on the shifted interval boundaries.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly recalled the formula relating the margin of error to the critical value and standard error.

You correctly set up the tt-statistic structure as the difference between the sample mean and hypothesized mean divided by the standard error.

You accurately connected confidence interval inclusion to the two-sided hypothesis test decision and the size of the PP-value.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Formula Misrecollection

You forgot to take the square root of n=16n = 16 in the denominator when calculating the standard error.

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

Confusing the standard error of the mean, σxˉ=sn\sigma_{\bar{x}} = \frac{s}{\sqrt{n}}, with a simple per-unit variance ratio or omitting the square root operation on sample size.

REASONING PATH
Your ACEpath teacher’s suggested approach
Standard Error Formula
  • The standard error of the sample mean is SE=sn\text{SE} = \frac{s}{\sqrt{n}}, never sn\frac{s}{n}.

  • Always write down the formula SE=sn\text{SE} = \frac{s}{\sqrt{n}} before substituting numerical values.

Square Root Check
  • When nn is a perfect square like 1616, n\sqrt{n} simplifies cleanly (here, 16=4\sqrt{16} = 4).

  • Check whether the denominator represents the sample size nn or the square root n\sqrt{n} before computing.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Recompute the standard error: SE=0.06016=0.0604=0.015 mm\text{SE} = \frac{0.060}{\sqrt{16}} = \frac{0.060}{4} = 0.015\text{ mm}.

  2. Calculate the margin of error: ME=2.131×0.015=0.031965 mm≈0.032 mm\text{ME} = 2.131 \times 0.015 = 0.031965\text{ mm} \approx 0.032\text{ mm}.

  3. Construct the 95%95\% confidence interval: [5.032−0.032,5.032+0.032]=[5.000 mm,5.064 mm][5.032 - 0.032, 5.032 + 0.032] = [5.000\text{ mm}, 5.064\text{ mm}].

  4. Compute the tt-statistic: t=5.032−5.0000.015=0.0320.015≈2.133t = \frac{5.032 - 5.000}{0.015} = \frac{0.032}{0.015} \approx 2.133.

  5. Compare with critical value t∗=2.131t^* = 2.131: since ∣2.133∣>2.131|2.133| > 2.131, reject H0H_0 at α=0.05\alpha = 0.05, confirming P<0.05P < 0.05.

Final answer

(a) [5.000 mm,5.064 mm][5.000\text{ mm}, 5.064\text{ mm}]; (b) t≈2.133t \approx 2.133; (c) Reject H0H_0 at α=0.05\alpha = 0.05; P<0.05P < 0.05.

NEXT STEPS
Final Takeaways from Your Teacher
  • Review the sampling distribution of the mean to reinforce why sample variation decreases as n\sqrt{n} rather than nn.

  • Practice calculating standard error across different sample sizes before plugging values into confidence intervals and hypothesis tests.

Lock it in

Similar question

01

A quality control engineer takes a random sample of n=25n = 25 bolts and finds a sample mean diameter of xˉ=10.05 mm\bar{x} = 10.05\text{ mm} with standard deviation s=0.10 mms = 0.10\text{ mm}. Using t∗=2.064t^* = 2.064 for a 95%95\% confidence interval, calculate the margin of error and determine the confidence interval for the true mean diameter μ\mu.

Answer

SE=0.1025=0.02 mm\text{SE} = \frac{0.10}{\sqrt{25}} = 0.02\text{ mm}. ME=2.064×0.02=0.04128 mm≈0.041 mm\text{ME} = 2.064 \times 0.02 = 0.04128\text{ mm} \approx 0.041\text{ mm}. The 95%95\% confidence interval is 10.05±0.041=[10.009 mm,10.091 mm]10.05 \pm 0.041 = [10.009\text{ mm}, 10.091\text{ mm}].

02

A sample of n=9n = 9 batteries yields a mean lifespan of xˉ=48.6 hours\bar{x} = 48.6\text{ hours} and s=1.8 hourss = 1.8\text{ hours}. Test H0:μ=50.0 hoursH_0: \mu = 50.0\text{ hours} versus Ha:μ≠50.0 hoursH_a: \mu \neq 50.0\text{ hours} using a 95%95\% confidence interval. Given t∗=2.306t^* = 2.306 for df=8df = 8, construct the interval, state your testing decision at α=0.05\alpha = 0.05, and indicate whether P>0.05P > 0.05 or P<0.05P < 0.05.

Answer

SE=1.89=0.6 hours\text{SE} = \frac{1.8}{\sqrt{9}} = 0.6\text{ hours}. ME=2.306×0.6=1.3836 hours\text{ME} = 2.306 \times 0.6 = 1.3836\text{ hours}. The 95%95\% CI is 48.6±1.384=[47.216,49.984] hours48.6 \pm 1.384 = [47.216, 49.984]\text{ hours}. Because the hypothesized value μ0=50.0\mu_0 = 50.0 lies outside the interval (and ∣t∣=∣48.6−50.00.6∣=2.333>2.306|t| = |\frac{48.6 - 50.0}{0.6}| = 2.333 > 2.306), reject H0H_0 at α=0.05\alpha = 0.05, which implies P<0.05P < 0.05.

03

A researcher constructs a 90%90\% confidence interval for μ\mu as [19.8,24.2][19.8, 24.2] from a sample of size n=16n = 16. However, they realize they mistakenly calculated ME=t∗⋅s\text{ME} = t^* \cdot s instead of t∗⋅snt^* \cdot \frac{s}{\sqrt{n}}. Find the corrected 90%90\% confidence interval for μ\mu.

Answer

The center is xˉ=19.8+24.22=22.0\bar{x} = \frac{19.8 + 24.2}{2} = 22.0. The incorrect margin of error was 24.2−22.0=2.224.2 - 22.0 = 2.2. Because the student used t∗st^* s instead of t∗sn\frac{t^* s}{\sqrt{n}}, the actual margin of error should be divided by n=16=4\sqrt{n} = \sqrt{16} = 4. Correct ME=2.24=0.55\text{ME} = \frac{2.2}{4} = 0.55. The corrected 90%90\% CI is 22.0±0.55=[21.45,22.55]22.0 \pm 0.55 = [21.45, 22.55].

Loading workspace…