Resources

An environmental laboratory evaluates a new municipal water filtration media designed to remove lead. In standard units, the regulatory actionable threshold for lead in drinking water is 15.0 μg/L15.0\,\mu\mathrm{g/L} (15.0 ppb15.0\text{ ppb}), and engineers specify that a filter achieves practical significance only if it lowers lead concentration by at least 8.0 μg/L8.0\,\mu\mathrm{g/L}. In a pilot run of n=36n = 36 treated water samples, the effluent lead concentrations are recorded in parts per billion (ppb, where 1 ppb=1 μg/L=10−6 g/L1\text{ ppb} = 1\,\mu\mathrm{g/L} = 10^{-6}\text{ g/L}), while the raw influent had a constant concentration of 2.20×10−2 mg/L2.20 \times 10^{-2}\text{ mg/L}. The mean effluent concentration is xˉeff=19.50 ppb\bar{x}_{\text{eff}} = 19.50\text{ ppb} with a standard deviation of s=3.00 ppbs = 3.00\text{ ppb}. (a) Convert the influent concentration into μg/L\mu\text{g/L} (ppb), then calculate the sample mean lead reduction xˉreduction\bar{x}_{\text{reduction}} in μg/L\mu\text{g/L}. (b) Test H0:μreduction=0H_0: \mu_{\text{reduction}} = 0 against Ha:μreduction>0H_a: \mu_{\text{reduction}} > 0 at α=0.001\alpha = 0.001 (tcrit=3.344t_{\text{crit}} = 3.344 for df=35df = 35). (c) Determine whether the filter demonstrates practical significance according to the engineering benchmark and municipal safety standards.

Answer and steps

Final answer

(a) 2.20×10−2 mg/L=2.20×10−2×103 μg/L=22.0 μg/L=22.0 ppb2.20 \times 10^{-2}\text{ mg/L} = 2.20 \times 10^{-2} \times 10^3\,\mu\mathrm{g/L} = 22.0\,\mu\mathrm{g/L} = 22.0\text{ ppb}. The mean reduction is xˉreduction=22.0 ppb−19.50 ppb=2.50 μg/L\bar{x}_{\text{reduction}} = 22.0\text{ ppb} - 19.50\text{ ppb} = 2.50\,\mu\mathrm{g/L}. (b) t=2.50−03.0036=2.500.500=5.00t = \frac{2.50 - 0}{\frac{3.00}{\sqrt{36}}} = \frac{2.50}{0.500} = 5.00. Since t=5.00>3.344t = 5.00 > 3.344 (p<0.0001<0.001p < 0.0001 < 0.001), reject H0H_0; the lead reduction is statistically significant. (c) The filter fails the practical significance test on two counts: the mean reduction of 2.50 μg/L2.50\,\mu\mathrm{g/L} is far below the 8.0 μg/L8.0\,\mu\mathrm{g/L} engineering benchmark, and the resulting effluent lead level of 19.50 μg/L19.50\,\mu\mathrm{g/L} remains above the 15.0 μg/L15.0\,\mu\mathrm{g/L} EPA actionable safety threshold.

Steps

Part (a): Unit Conversion and Mean Lead Reduction

  1. Convert the influent concentration to μg/L\mu\text{g/L} (ppb\text{ppb}): Since 1 mg=103 μg1\text{ mg} = 10^3\text{ }\mu\text{g}, we multiply the concentration in mg/L\text{mg/L} by 10310^3: Influent=2.20×10−2 mg/L×103 μgmg=22.0 μg/L=22.0 ppb\text{Influent} = 2.20 \times 10^{-2}\text{ mg/L} \times 10^3\text{ }\frac{\mu\text{g}}{\text{mg}} = 22.0\text{ }\mu\text{g/L} = 22.0\text{ ppb}
  2. Calculate the sample mean reduction xˉreduction\bar{x}_{\text{reduction}}: The influent level is constant at 22.0 ppb22.0\text{ ppb}, and the mean effluent concentration is xˉeff=19.50 ppb\bar{x}_{\text{eff}} = 19.50\text{ ppb}: xˉreduction=Influent−xˉeff=22.0 ppb−19.50 ppb=2.50 μg/L\bar{x}_{\text{reduction}} = \text{Influent} - \bar{x}_{\text{eff}} = 22.0\text{ ppb} - 19.50\text{ ppb} = 2.50\text{ }\mu\text{g/L}

Part (b): Hypothesis Testing for Statistical Significance

  1. State the hypotheses: H0:μreduction=0H_0: \mu_{\text{reduction}} = 0 Ha:μreduction>0H_a: \mu_{\text{reduction}} > 0
  2. Compute the standard error of the mean: Because the influent level is constant, the standard deviation of the reduction is identical to that of the effluent samples, s=3.00 ppbs = 3.00\text{ ppb}. With sample size n=36n = 36: SE=sn=3.0036=3.006=0.500 μg/LSE = \frac{s}{\sqrt{n}} = \frac{3.00}{\sqrt{36}} = \frac{3.00}{6} = 0.500\text{ }\mu\text{g/L}
  3. Calculate the tt-statistic: t=xˉreduction−μ0SE=2.50−00.500=5.00t = \frac{\bar{x}_{\text{reduction}} - \mu_0}{SE} = \frac{2.50 - 0}{0.500} = 5.00
  4. Conclusion at α=0.001\alpha = 0.001: For degrees of freedom df=n−1=35df = n - 1 = 35, the critical value is tcrit=3.344t_{\text{crit}} = 3.344. Because t=5.00>3.344t = 5.00 > 3.344 (which corresponds to p<0.0001<0.001p < 0.0001 < 0.001), we reject H0H_0. There is statistically significant evidence that the filter produces a non-zero reduction in lead concentration.

Part (c): Practical Significance vs. Statistical Significance

  • Pitfall to avoid: Confusing statistical significance with practical significance. Statistical significance simply confirms that the observed reduction of 2.50 μg/L2.50\text{ }\mu\text{g/L} is unlikely to be due to chance variation alone (p<0.001p < 0.001). It does not automatically mean the effect size is meaningful or useful in the real world.
  • Evaluating practical utility:

    1. Engineering criterion: The engineers require a reduction of at least 8.0 μg/L8.0\text{ }\mu\text{g/L}. The observed average reduction of 2.50 μg/L2.50\text{ }\mu\text{g/L} falls well short of this target.
    2. Regulatory safety threshold: The EPA actionable standard is 15.0 μg/L15.0\text{ }\mu\text{g/L}. Even after filtration, the average effluent concentration is 19.50 μg/L19.50\text{ }\mu\text{g/L}, which remains above the safe drinking limit.

Therefore, despite being statistically significant, the filtration media does not demonstrate practical significance.

Typical errors

OverviewACEpath Teacher's Diagnosis

01 The student confuses the standard error with the standard deviation and omits dividing s by √n when calculating the t-statistic.

Sample Student Work

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

    Influent=2.20×10−2 mg/L×1000=22.0 ppb  ⟹  xˉreduction=22.0−19.50=2.50 ppb\text{Influent} = 2.20 \times 10^{-2}\text{ mg/L} \times 1000 = 22.0\text{ ppb} \implies \bar{x}_{\text{reduction}} = 22.0 - 19.50 = 2.50\text{ ppb}

  2. Step 2

    t=xˉreduction−0s=2.503.00≈0.833t = \frac{\bar{x}_{\text{reduction}} - 0}{s} = \frac{2.50}{3.00} \approx 0.833

    Correct

    t=xˉreduction−μ0s/n=2.50−03.00/36=2.500.500=5.00t = \frac{\bar{x}_{\text{reduction}} - \mu_0}{s / \sqrt{n}} = \frac{2.50 - 0}{3.00 / \sqrt{36}} = \frac{2.50}{0.500} = 5.00

    You divided by the sample standard deviation ss alone rather than the standard error of the sample mean, SE=sn=3.0036=0.500SE = \frac{s}{\sqrt{n}} = \frac{3.00}{\sqrt{36}} = 0.500.

  3. Step 3

    t=0.833<3.344  ⟹  Fail to reject H0; there is no significant reduction in lead.t = 0.833 < 3.344 \implies \text{Fail to reject } H_0\text{; there is no significant reduction in lead.}

    Correct

    t=5.00>3.344  ⟹  Reject H0; the lead reduction is statistically significant at α=0.001t = 5.00 > 3.344 \implies \text{Reject } H_0\text{; the lead reduction is statistically significant at } \alpha = 0.001.

    Because your calculated tt-statistic was artificially deflated by omitting n\sqrt{n}, you concluded not to reject H0H_0 instead of rejecting it.

  4. Step 4

    The filter does not achieve practical significance because reduction is not statistically significant, 2.50 ppb<8.0 ppb2.50\text{ ppb} < 8.0\text{ ppb}, and effluent 19.50 ppb>15.0 ppb19.50\text{ ppb} > 15.0\text{ ppb}.

    Correct

    The lead reduction is statistically significant (t=5.00>3.344t = 5.00 > 3.344), but it does not achieve practical significance because the mean reduction of 2.50 μg/L2.50\,\mu\text{g/L} is below the 8.0 μg/L8.0\,\mu\text{g/L} engineering threshold, and the effluent concentration of 19.50 μg/L19.50\,\mu\text{g/L} remains above the 15.0 μg/L15.0\,\mu\text{g/L} regulatory limit.

    You accurately identified the engineering benchmark (2.50<8.02.50 < 8.0) and the actionable safety standard (19.50>15.019.50 > 15.0). However, your explanation listed a lack of statistical significance as a contributing cause, which stems from the calculation error in part (b).

WHAT YOU DID WELL
Compliments from Your Teacher

You performed the unit conversion between mg/L\text{mg/L} and ppb\text{ppb} accurately.

You calculated the mean reduction between influent and effluent correctly.

You correctly cited the practical engineering requirement and municipal safety limit.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Standard error vs standard deviation confusion

You wrote ss in the denominator of the tt-statistic instead of sn\frac{s}{\sqrt{n}}.

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You confused individual sample variability (standard deviation ss) with the variability of the sample mean across repeated samples (standard error SE=snSE = \frac{s}{\sqrt{n}}).

REASONING PATH
Your ACEpath teacher’s suggested approach
One-Sample $t$-Test Statistic Formula
  • For inference about a population mean μ\mu, the test statistic measures distance in standard errors: t=xˉ−μ0s/nt = \frac{\bar{x} - \mu_0}{s / \sqrt{n}}.

  • Always divide the sample standard deviation by n\sqrt{n} to obtain the denominator before computing tt.

Statistical vs. Practical Significance
  • Statistical significance (p<αp < \alpha) confirms an effect is unlikely due to random chance, whereas practical significance evaluates whether the magnitude of the effect meets operational or real-world thresholds.

  • Evaluate statistical significance solely using the test statistic and pp-value, and evaluate practical significance using effect sizes and contextual benchmarks.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Compute the standard error of the mean: SE=sn=3.0036=3.006=0.500 μg/LSE = \frac{s}{\sqrt{n}} = \frac{3.00}{\sqrt{36}} = \frac{3.00}{6} = 0.500\,\mu\text{g/L}.

  2. Calculate the one-sample tt-statistic: t=xˉreduction−0SE=2.50−00.500=5.00t = \frac{\bar{x}_{\text{reduction}} - 0}{SE} = \frac{2.50 - 0}{0.500} = 5.00.

  3. Compare with the critical value: Since t=5.00>3.344t = 5.00 > 3.344, reject H0H_0 at α=0.001\alpha = 0.001. The lead reduction is statistically significant.

  4. Assess practical significance: Despite being statistically significant, the filter does not achieve practical significance because the mean reduction (2.50 μg/L2.50\,\mu\text{g/L}) is less than the required 8.0 μg/L8.0\,\mu\text{g/L}, and the treated water (19.50 μg/L19.50\,\mu\text{g/L}) still exceeds the 15.0 μg/L15.0\,\mu\text{g/L} regulatory limit.

Final answer

(a) 22.0 μg/L22.0\,\mu\text{g/L} influent; xˉreduction=2.50 μg/L\bar{x}_{\text{reduction}} = 2.50\,\mu\text{g/L}. (b) t=5.00>3.344t = 5.00 > 3.344; reject H0H_0 (statistically significant reduction). (c) Fails practical significance because 2.50 μg/L<8.0 μg/L2.50\,\mu\text{g/L} < 8.0\,\mu\text{g/L} engineering threshold and 19.50 μg/L>15.0 μg/L19.50\,\mu\text{g/L} > 15.0\,\mu\text{g/L} EPA standard.

NEXT STEPS
Final Takeaways from Your Teacher
  • Review the formula card for one-sample tests: verify whether you are testing a single observation (zz-score for an individual) or a sample mean (tt-statistic with SE=s/nSE = s/\sqrt{n}).

  • Practice computing standard error first as an explicit sub-step before writing out the tt-ratio.

  • Remember that a test can be statistically significant (real non-zero effect) without being practically significant (meaningful effect size).

OverviewACEpath Teacher's Diagnosis

02 The student misinterprets statistical significance (rejecting H0) as automatically implying practical significance, ignoring the engineering benchmark of 8.0 µg/L and the regulatory limit.

Sample Student Work

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

    Influent=2.20×10−2×1000=22.0 ppb  ⟹  xˉreduction=22.0−19.50=2.50 ppb\text{Influent} = 2.20 \times 10^{-2} \times 1000 = 22.0\text{ ppb} \implies \bar{x}_{\text{reduction}} = 22.0 - 19.50 = 2.50\text{ ppb}

  2. Step 2

    SE=3.0036=0.50  ⟹  t=2.500.50=5.00  ⟹  Since 5.00>3.344, reject H0; lead reduction is statistically significant.SE = \frac{3.00}{\sqrt{36}} = 0.50 \implies t = \frac{2.50}{0.50} = 5.00 \implies \text{Since } 5.00 > 3.344\text{, reject } H_0\text{; lead reduction is statistically significant.}

  3. Step 3

    Yes, the filter demonstrates practical significance because the null hypothesis was rejected with p<0.001.\text{Yes, the filter demonstrates practical significance because the null hypothesis was rejected with } p < 0.001\text{.}

    Correct

    No, the filter does not demonstrate practical significance because xˉreduction=2.50 \mug/L<8.0 \mug/L and effluent 19.50 \mug/L>15.0 \mug/L.\text{No, the filter does not demonstrate practical significance because } \bar{x}_{\text{reduction}} = 2.50\text{ \mu g/L} < 8.0\text{ \mu g/L}\text{ and effluent } 19.50\text{ \mu g/L} > 15.0\text{ \mu g/L}\text{.}

    Statistical significance only proves that the reduction is distinguishable from zero by chance. Practical significance requires evaluating whether the effect size (2.50 \mug/L2.50\text{ \mu g/L}) meets the required engineering goal (8.0 \mug/L8.0\text{ \mu g/L}) and safety limit (15.0 \mug/L15.0\text{ \mu g/L}).

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly converted the influent concentration to parts per billion.

You accurately calculated the sample mean lead reduction.

You calculated the standard error and test statistic without arithmetic errors.

You properly evaluated the hypothesis test against the critical value and rejected the null hypothesis.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Conflating Statistical and Practical Significance

no

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You assumed that rejecting the null hypothesis with a very small pp-value automatically guarantees real-world or practical utility, rather than comparing the observed effect size against domain-specific benchmarks and thresholds.

REASONING PATH
Your ACEpath teacher’s suggested approach
Statistical vs. Practical Distinction
  • Statistical significance asks 'Is the effect non-zero beyond chance variation?'; practical significance asks 'Is the effect large enough to matter in practice?'

  • Always evaluate the magnitude of the point estimate against the practical thresholds specified in the problem statement.

Contextual Benchmark Check
  • Practical utility requires meeting domain criteria: target effect size and safety compliance standards.

  • Compare xˉreduction=2.50 \mug/L\bar{x}_{\text{reduction}} = 2.50\text{ \mu g/L} to the engineering threshold (8.0 \mug/L8.0\text{ \mu g/L}) and the effluent level (19.50 \mug/L19.50\text{ \mu g/L}) to the regulatory limit (15.0 \mug/L15.0\text{ \mu g/L}).

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Compare the observed mean reduction xˉreduction=2.50 \mug/L\bar{x}_{\text{reduction}} = 2.50\text{ \mu g/L} to the engineering threshold of 8.0 \mug/L8.0\text{ \mu g/L}.

  2. Observe that 2.50 \mug/L<8.0 \mug/L2.50\text{ \mu g/L} < 8.0\text{ \mu g/L}, meaning the filter does not meet the specified minimum performance standard.

  3. Compare the remaining effluent concentration of 19.50 \mug/L19.50\text{ \mu g/L} to the municipal safety limit of 15.0 \mug/L15.0\text{ \mu g/L}, noting that water remains unsafe to drink.

  4. Conclude that the filter does not achieve practical significance despite being statistically significant.

Final answer

The filter does not demonstrate practical significance because its mean reduction of 2.50 \mug/L2.50\text{ \mu g/L} fails to reach the required 8.0 \mug/L8.0\text{ \mu g/L} benchmark, and the effluent concentration of 19.50 \mug/L19.50\text{ \mu g/L} still exceeds the regulatory safety threshold of 15.0 \mug/L15.0\text{ \mu g/L}.

NEXT STEPS
Final Takeaways from Your Teacher
  • Review the distinct definitions of statistical significance (p<αp < \alpha) and practical significance (meaningful effect size).

  • When an engineering, clinical, or financial problem gives specific thresholds, set up an explicit inequality to compare your sample estimate to those thresholds before concluding.

OverviewACEpath Teacher's Diagnosis

03 The student uses an incorrect metric conversion factor of 10⁶ instead of 10³ from mg/L to µg/L, leading to an artificially massive reduction calculation.

Sample Student Work

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

    Influent=2.20×10−2×106=22000 ppb  ⟹  xˉreduction=22000−19.50=21980.5 ppb\text{Influent} = 2.20 \times 10^{-2} \times 10^6 = 22000\text{ ppb} \implies \bar{x}_{\text{reduction}} = 22000 - 19.50 = 21980.5\text{ ppb}

    Correct

    Influent=2.20×10−2 mg/L×103 μg/mg=22.0 μg/L=22.0 ppb  ⟹  xˉreduction=22.0−19.50=2.50 μg/L\text{Influent} = 2.20 \times 10^{-2}\text{ mg/L} \times 10^3\,\mu\text{g/mg} = 22.0\,\mu\text{g/L} = 22.0\text{ ppb} \implies \bar{x}_{\text{reduction}} = 22.0 - 19.50 = 2.50\,\mu\text{g/L}

    You multiplied by 10610^6 instead of 10310^3 to convert milligrams to micrograms. Because 1 mg=103 μg1\text{ mg} = 10^3\,\mu\text{g}, the converted influent concentration is 22.0 ppb22.0\text{ ppb}, which makes the mean reduction 2.50 μg/L2.50\,\mu\text{g/L} rather than 21980.5 ppb21980.5\text{ ppb}.

  2. Step 2

    SE=3.0036=0.50  ⟹  t=21980.50.50=43961  ⟹  43961>3.344, reject H0SE = \frac{3.00}{\sqrt{36}} = 0.50 \implies t = \frac{21980.5}{0.50} = 43961 \implies 43961 > 3.344\text{, reject } H_0

    Correct

    SE=3.0036=0.500  ⟹  t=2.50−00.500=5.00  ⟹  5.00>3.344, reject H0SE = \frac{3.00}{\sqrt{36}} = 0.500 \implies t = \frac{2.50 - 0}{0.500} = 5.00 \implies 5.00 > 3.344\text{, reject } H_0

    Your formula for standard error and your rejection rule comparison were mathematically sound, but your tt-statistic was heavily inflated because it inherited the incorrect reduction from part (a).

  3. Step 3

    The filter achieves practical significance because \text{The filter achieves practical significance because } 21980.5 ppb is much greater than \text{ ppb is much greater than } 8.0 ppb and effluent \text{ ppb and effluent } 19.50 ppb is close to standard.\text{ ppb is close to standard.}

    Correct

    The filter fails practical significance: the true reduction of \text{The filter fails practical significance: the true reduction of } 2.50,\mu$$\text{g/L}  is below the \text{ is below the } 8.0,\mu$$\text{g/L}  engineering benchmark, and effluent \text{ engineering benchmark, and effluent } 19.50,\mu$$\text{g/L}  exceeds the \text{ exceeds the } 15.0,\mu$$\text{g/L}  regulatory threshold.\text{ regulatory threshold.}

    In addition to propagating the conversion mistake, treating an effluent concentration of 19.50 ppb19.50\text{ ppb} as acceptable overlooks that regulatory safety standards are strict non-exceedance limits; 19.50 ppb19.50\text{ ppb} is above the 15.0 ppb15.0\text{ ppb} maximum allowed threshold.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly calculated the standard error of the mean using SE=sn=0.50SE = \frac{s}{\sqrt{n}} = 0.50.

You applied the correct critical value decision rule to reject H0H_0 when t>tcritt > t_{\text{crit}}.

You recognized that evaluating practical significance requires checking both the engineering reduction target and the effluent safety standard.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Metric prefix conversion error

You multiplied by 10610^6 instead of 10310^3 when converting from milligrams (mg\text{mg}) to micrograms (μg\mu\text{g}).

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

Conflating the conversion between grams and micrograms (1 g=106 μg1\text{ g} = 10^6\,\mu\text{g}) with the conversion between milligrams and micrograms (1 mg=103 μg1\text{ mg} = 10^3\,\mu\text{g}).

REASONING PATH
Your ACEpath teacher’s suggested approach
Metric prefix bridge
  • 1 g=103 mg=106 μg1\text{ g} = 10^3\text{ mg} = 10^6\,\mu\text{g}, so 1 mg=103 μg1\text{ mg} = 10^3\,\mu\text{g}.

  • Always convert through the base unit (grams) or directly apply the three-order-of-magnitude difference between milli- (10−310^{-3}) and micro- (10−610^{-6}).

Statistical vs. Practical Significance
  • A statistically significant test (p<αp < \alpha) confirms only that an effect is non-zero, not that the effect size meets real-world standards.

  • Check whether the observed effect size satisfies operational benchmarks and regulatory limits independently of the test statistic.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Convert influent to μg/L\mu\text{g/L}: 2.20×10−2 mg/L×103 μg/mg=22.0 μg/L=22.0 ppb2.20 \times 10^{-2}\text{ mg/L} \times 10^3\,\mu\text{g/mg} = 22.0\,\mu\text{g/L} = 22.0\text{ ppb}.

  2. Compute the true mean reduction: xˉreduction=22.0−19.50=2.50 μg/L\bar{x}_{\text{reduction}} = 22.0 - 19.50 = 2.50\,\mu\text{g/L}.

  3. Compute the tt-statistic: t=2.50−00.500=5.00t = \frac{2.50 - 0}{0.500} = 5.00. Since 5.00>3.3445.00 > 3.344, reject H0H_0.

  4. Assess practical significance: 2.50 μg/L<8.0 μg/L2.50\,\mu\text{g/L} < 8.0\,\mu\text{g/L} (fails engineering target) and 19.50 μg/L>15.0 μg/L19.50\,\mu\text{g/L} > 15.0\,\mu\text{g/L} (fails EPA threshold).

Final answer

(a) Influent is 22.0 ppb22.0\text{ ppb}; xˉreduction=2.50 μg/L\bar{x}_{\text{reduction}} = 2.50\,\mu\text{g/L}. (b) Reject H0H_0 (t=5.00>3.344t = 5.00 > 3.344). (c) The filter does not demonstrate practical significance.

NEXT STEPS
Final Takeaways from Your Teacher
  • Review unit conversions involving adjacent metric prefixes (such as milli- to micro-).

  • Practice checking effect sizes against external domain thresholds after running hypothesis tests.

Lock it in

Similar question

01

A new air scrubber is tested in n=25n = 25 industrial chambers. The mean reduction in particulate matter is xˉ=4.0 mg/m3\bar{x} = 4.0\text{ mg/m}^3 with sample standard deviation s=5.0 mg/m3s = 5.0\text{ mg/m}^3. Test H0:μ=0H_0: \mu = 0 versus Ha:μ>0H_a: \mu > 0 at α=0.05\alpha = 0.05 (critical value t24=1.711t_{24} = 1.711).

Answer

t=4.00t = 4.00; reject H0H_0. Steps: SE=sn=5.025=1.0 mg/m3SE = \frac{s}{\sqrt{n}} = \frac{5.0}{\sqrt{25}} = 1.0\text{ mg/m}^3; t=4.0−01.0=4.00t = \frac{4.0 - 0}{1.0} = 4.00. Since 4.00>1.7114.00 > 1.711, the reduction is statistically significant.

02

A medical device company develops an automated blood pressure cuff. In a trial of n=100n = 100 patients, the device reduces systolic blood pressure by an average of xˉ=1.2 mmHg\bar{x} = 1.2\text{ mmHg} with s=4.0 mmHgs = 4.0\text{ mmHg}. Physicians state that an intervention requires a reduction of at least 5.0 mmHg5.0\text{ mmHg} to be clinically meaningful. (a) Is the reduction statistically significant at α=0.01\alpha = 0.01 (tcrit=2.364t_{\text{crit}} = 2.364)? (b) Is it practically significant?

Answer

(a) Yes, statistically significant: SE=4.0100=0.40 mmHgSE = \frac{4.0}{\sqrt{100}} = 0.40\text{ mmHg}, t=1.2−00.40=3.00>2.364t = \frac{1.2 - 0}{0.40} = 3.00 > 2.364. (b) No, it lacks practical significance because the observed reduction of 1.2 mmHg1.2\text{ mmHg} is well below the 5.0 mmHg5.0\text{ mmHg} clinical threshold.

03

An algorithm optimization team tests a new caching routine on n=400n = 400 server batches. The latency reduction has sample mean xˉ=0.60 ms\bar{x} = 0.60\text{ ms} and standard deviation s=4.00 mss = 4.00\text{ ms}. Management requires a latency drop of at least 3.00 ms3.00\text{ ms} to justify deployment costs. Evaluate both statistical significance at α=0.001\alpha = 0.001 (zcrit≈3.090z_{\text{crit}} \approx 3.090) and practical significance.

Answer

Statistically significant, but practically insignificant. Steps: SE=4.00400=4.0020=0.20 msSE = \frac{4.00}{\sqrt{400}} = \frac{4.00}{20} = 0.20\text{ ms}; test statistic z=0.60−00.20=3.00z = \frac{0.60 - 0}{0.20} = 3.00 (or with t399t_{399}, p≈0.0014p \approx 0.0014; if using two-tailed/one-tailed benchmark z=3.00z = 3.00, at one-tailed α=0.001\alpha=0.001, zcrit=3.090z_{\text{crit}}=3.090, so it narrowly misses or borders on statistical significance depending on cutoff, but with a massive sample size even a tiny effect of 0.60 ms0.60\text{ ms} yields a large statistic). Regardless, the practical threshold of 3.00 ms3.00\text{ ms} is five times larger than the observed 0.60 ms0.60\text{ ms} effect, so management should not deploy the routine.

Loading workspace…