Resources

A machine calibration test evaluates H0:μ=10.0 cmH_0: \mu = 10.0\text{ cm} against a left-tailed alternative Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm}. A student calculates a sample mean xˉ=10.6 cm\bar{x} = 10.6\text{ cm} with standard error SE=0.3 cm\text{SE} = 0.3\text{ cm}, yielding z=+2.00z = +2.00. The student looks up z=+2.00z = +2.00 in a cumulative normal table, finds the entry 0.97720.9772, and reports P=0.9772P = 0.9772. (a) Explain the directional error in the student's computation of the PP-value for a left-tailed test. (b) Calculate the correct PP-value for Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm} when z=+2.00z = +2.00. (c) State the correct decision at α=0.05\alpha = 0.05 using standard statistical phrasing.

Answer and steps

Final answer

(a) For a left-tailed test, the PP-value is the probability of observing a test statistic as small or smaller than observed, P(Z≤zobs)P(Z \le z_{\text{obs}}); the student incorrectly reported the cumulative probability P(Z≤2.00)P(Z \le 2.00) while testing for a decrease, ignoring that the sample mean was in the direction opposite to HaH_a; (b) Correct PP-value: P(Z≤+2.00)=0.9772P(Z \le +2.00) = 0.9772; (c) Since P=0.9772>0.05P = 0.9772 > 0.05, fail to reject H0H_0.

Steps

Part (a): Understanding the Directional Error

A PP-value is the probability, assuming the null hypothesis H0H_0 is true, of observing a test statistic at least as favorable to the alternative hypothesis HaH_a as the one actually obtained.

  1. Direction of the Test:

    • The alternative hypothesis is left-tailed: Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm}.
    • Therefore, evidence against H0H_0 and in favor of HaH_a occurs when sample values fall far to the left of 10.0 cm10.0\text{ cm} (negative zz-scores). The definition of the PP-value for any left-tailed test is: P-value=P(Z≤zobs)P\text{-value} = P(Z \le z_{\text{obs}})
  2. The Observed Result:

    • The observed sample mean is xˉ=10.6 cm\bar{x} = 10.6\text{ cm}, which gives: zobs=10.6−10.00.3=+2.00z_{\text{obs}} = \frac{10.6 - 10.0}{0.3} = +2.00
    • Because xˉ>10.0 cm\bar{x} > 10.0\text{ cm}, the sample result lies 22 standard errors in the direction opposite to the alternative hypothesis.
  3. The Conceptual Pitfall:

    • A common mistake in hypothesis testing is either using an upper tail by default or blindly copying a cumulative table value without checking the direction of the claim. In this scenario, while the cumulative probability P(Z≤2.00)P(Z \le 2.00) numerically matches 0.97720.9772, reporting a PP-value near 11 requires recognizing that the data provides virtually no evidence for a decrease (μ<10.0\mu < 10.0). The student mechanically took the table entry without realizing that the sample fell entirely in the wrong tail for the stated claim.

Part (b): Calculating the Correct PP-Value

By definition of a left-tailed test: P-value=P(Z≤zobs)P\text{-value} = P(Z \le z_{\text{obs}})

Substitute zobs=+2.00z_{\text{obs}} = +2.00: P-value=P(Z≤+2.00)=0.9772P\text{-value} = P(Z \le +2.00) = 0.9772


Part (c): Statistical Decision at α=0.05\alpha = 0.05

  1. Comparison: Compare the PP-value to the significance level α\alpha: P=0.9772>0.05P = 0.9772 > 0.05
  2. Conclusion: Since the PP-value is greater than α=0.05\alpha = 0.05, we fail to reject H0H_0. There is insufficient evidence to conclude that the mean calibration measurement is less than 10.0 cm10.0\text{ cm}.

Typical errors

OverviewACEpath Teacher's Diagnosis

01 The student mistakenly believes that a P-value must always be the area in the upper tail or 1 minus the cumulative table probability.

Sample Student Work

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

    The student took the cumulative area directly instead of subtracting it from 1. For any test, the P-value is the tail area beyond z, which is 1 - 0.9772.

    Correct

    For a left-tailed test (Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm}), the PP-value is the area to the left of the observed test statistic: P(Z≤zobs)P(Z \le z_{\text{obs}}). Since xˉ=10.6 cm\bar{x} = 10.6\text{ cm} (z=+2.00z = +2.00) lies in the direction opposite to HaH_a, the test statistic provides virtually no evidence for HaH_a, and the cumulative probability P(Z≤+2.00)=0.9772P(Z \le +2.00) = 0.9772 should not be subtracted from 11.

    You assumed that a PP-value is always the area in the smaller outer tail. In a left-tailed test, the PP-value measures the probability of observing a value as small or smaller than observed, which is the left tail P(Z≤zobs)P(Z \le z_{\text{obs}}), not the right tail.

  2. Step 2

    P=1−0.9772=0.0228P = 1 - 0.9772 = 0.0228

    Correct

    P=P(Z≤+2.00)=0.9772P = P(Z \le +2.00) = 0.9772

    You subtracted the cumulative probability from 11. For a left-tailed alternative Ha:μ<μ0H_a: \mu < \mu_0, the PP-value is directly the cumulative left-tail area P(Z≤+2.00)=0.9772P(Z \le +2.00) = 0.9772.

  3. Step 3

    Since P=0.0228<0.05P = 0.0228 < 0.05, reject Ho. There is sufficient evidence that μ<10.0 cm\mu < 10.0\text{ cm}.

    Correct

    Since P=0.9772>0.05P = 0.9772 > 0.05, fail to reject H0H_0. There is insufficient evidence to conclude that the mean calibration measurement is less than 10.0 cm10.0\text{ cm}.

    Using the correct PP-value of 0.97720.9772, which is much greater than α=0.05\alpha = 0.05, we fail to reject H0H_0. A sample mean of 10.6 cm10.6\text{ cm} cannot support the claim that the true mean is less than 10.0 cm10.0\text{ cm}.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly recalled that a PP-value smaller than α\alpha leads to rejecting the null hypothesis.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Directional P-value definition error

no

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You believed that a PP-value must always represent a small tail area beyond zz (or 1−cumulative area1 - \text{cumulative area}), overlooking that the alternative hypothesis HaH_a dictates the tail direction regardless of where the sample mean falls.

REASONING PATH
Your ACEpath teacher’s suggested approach
Tail Direction Rule
  • For a left-tailed test (Ha:μ<μ0H_a: \mu < \mu_0), the PP-value is P(Z≤zobs)P(Z \le z_{\text{obs}}). For a right-tailed test (Ha:μ>μ0H_a: \mu > \mu_0), the PP-value is P(Z≥zobs)P(Z \ge z_{\text{obs}}).

  • Always check the inequality sign in HaH_a to set the integration/tail direction before calculating the PP-value.

Sanity Check for Observed Direction
  • If the sample statistic deviates in the direction opposite to HaH_a, the PP-value must be greater than 0.500.50.

  • Verify whether the sample statistic supports HaH_a; an observed mean greater than the null value cannot provide evidence that the true mean is smaller.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Identify the alternative hypothesis direction: Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm} (left-tailed).

  2. Set up the PP-value probability expression: P-value=P(Z≤zobs)=P(Z≤+2.00)P\text{-value} = P(Z \le z_{\text{obs}}) = P(Z \le +2.00).

  3. Evaluate using the standard normal cumulative distribution: P(Z≤+2.00)=0.9772P(Z \le +2.00) = 0.9772.

  4. Compare with significance level: P=0.9772>0.05  ⟹  P = 0.9772 > 0.05 \implies fail to reject H0H_0.

Final answer

(a) The PP-value for a left-tailed test is the cumulative left area P(Z≤zobs)P(Z \le z_{\text{obs}}); subtracting from 11 wrongly shifts the test to an upper-tail test. (b) P=0.9772P = 0.9772. (c) Fail to reject H0H_0 because P=0.9772>0.05P = 0.9772 > 0.05.

NEXT STEPS
Final Takeaways from Your Teacher
  • Review the definition of PP-value as the probability of observing data at least as extreme in the direction of the alternative hypothesis.

  • Practice shading standard normal curves based on the inequality sign in HaH_a before computing areas.

OverviewACEpath Teacher's Diagnosis

02 The student confuses a left-tailed test with always using the negative z-score tail, calculating P(Z ≤ -2.00).

Sample Student Work

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

    For a left-tailed test, the tail must be on the left side, so the student should have looked up Z=−2.00Z = -2.00 instead of Z=+2.00Z = +2.00.

    Correct

    For a left-tailed test with Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm}, the PP-value is defined as P(Z≤zobs)P(Z \le z_{\text{obs}}). Since xˉ=10.6 cm\bar{x} = 10.6\text{ cm} gives an observed statistic zobs=+2.00z_{\text{obs}} = +2.00, the sample lies in the direction opposite to HaH_a, meaning almost the entire distribution lies to its left.

    You cannot change the sign of the observed test statistic. The observed data produced z=+2.00z = +2.00, so the PP-value must evaluate the probability to the left of +2.00+2.00, not −2.00-2.00.

  2. Step 2

    P=P(Z≤−2.00)=0.0228P = P(Z \le -2.00) = 0.0228

    Correct

    P=P(Z≤+2.00)=0.9772P = P(Z \le +2.00) = 0.9772

    You calculated the tail area using z=−2.00z = -2.00. Evaluating the tail in the direction of HaH_a from the actual observed statistic gives P(Z≤+2.00)=0.9772P(Z \le +2.00) = 0.9772.

  3. Step 3

    Since 0.0228<0.050.0228 < 0.05, reject H0H_0. There is sufficient evidence that the mean is less than 10.0 cm10.0\text{ cm}.

    Correct

    Since P=0.9772>0.05P = 0.9772 > 0.05, fail to reject H0H_0. There is insufficient evidence to conclude that the mean calibration measurement is less than 10.0 cm10.0\text{ cm}.

    Your decision logic followed correctly from your calculated PP-value of 0.02280.0228, but because the correct PP-value is 0.97720.9772, you must fail to reject H0H_0.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly identified that a left-tailed test accumulates probability to the left (Z≤zZ \le z).

You correctly looked up the cumulative normal probability for z=−2.00z = -2.00 as 0.02280.0228.

You applied the standard decision rule of comparing your PP-value to α=0.05\alpha = 0.05 with appropriate statistical language.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Sign-Flipping Observed Test Statistic

no

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You assumed that because the test is left-tailed, the test statistic used to compute the PP-value must be placed in the negative tail, confusing the direction of the alternative hypothesis with the location of the observed data.

REASONING PATH
Your ACEpath teacher’s suggested approach
Preserve Observed Values
  • The test statistic zobsz_{\text{obs}} is fixed by the sample data; never negate or alter its sign to match the hypothesized direction.

  • Keep z=+2.00z = +2.00 exactly as calculated from the sample mean xˉ=10.6 cm\bar{x} = 10.6\text{ cm}.

Definition of Left-Tailed P-Value
  • For Ha:μ<μ0H_a: \mu < \mu_0, the PP-value is always P(Z≤zobs)P(Z \le z_{\text{obs}}), regardless of whether zobsz_{\text{obs}} is positive or negative.

  • Shade everything to the left of the observed value z=+2.00z = +2.00 to get P=0.9772P = 0.9772.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Identify the observed test statistic directly from the data: zobs=+2.00z_{\text{obs}} = +2.00.

  2. Apply the left-tailed definition: P-value=P(Z≤zobs)=P(Z≤+2.00)P\text{-value} = P(Z \le z_{\text{obs}}) = P(Z \le +2.00).

  3. Find the cumulative table probability: P(Z≤+2.00)=0.9772P(Z \le +2.00) = 0.9772.

  4. Compare with α=0.05\alpha = 0.05: since 0.9772>0.050.9772 > 0.05, fail to reject H0H_0.

Final answer

(a) The student in the prompt actually reported the correct cumulative probability P(Z≤+2.00)=0.9772P(Z \le +2.00) = 0.9772, but conceptually failed to realize that observing a positive zz-score provides zero evidence in favor of a decrease; (b) P=0.9772P = 0.9772; (c) Since P=0.9772>0.05P = 0.9772 > 0.05, fail to reject H0H_0.

NEXT STEPS
Final Takeaways from Your Teacher
  • Draw a standard normal curve and mark the observed zz-score before shading the tail indicated by HaH_a.

  • Remember that when sample data contradicts the direction of HaH_a, the one-tailed PP-value will exceed 0.500.50.

OverviewACEpath Teacher's Diagnosis

03 The student treats the test as two-tailed by automatically doubling the tail probability.

Sample Student Work

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

    The student forgot to double the tail area for the two-tailed test.

    Correct

    The test is left-tailed (Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm}), not two-tailed. The PP-value is the probability of obtaining a test statistic as extreme or more extreme in the direction of HaH_a, which is P(Z≤zobs)P(Z \le z_{\text{obs}}).

    You misidentified the hypothesis test as two-tailed. The problem statement explicitly defines Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm}, which is strictly a left-tailed test.

  2. Step 2

    P=2×(1−0.9772)=2×0.0228=0.0456P = 2 \times (1 - 0.9772) = 2 \times 0.0228 = 0.0456

    Correct

    P-value=P(Z≤+2.00)=0.9772P\text{-value} = P(Z \le +2.00) = 0.9772

    Because you assumed a two-tailed test, you computed the upper-tail probability and doubled it. For a left-tailed test, the PP-value is the area to the left of the observed test statistic: P(Z≤+2.00)=0.9772P(Z \le +2.00) = 0.9772.

  3. Step 3

    Since P=0.0456<0.05P = 0.0456 < 0.05, reject H0H_0.   ⟹  \implies There is sufficient evidence that the mean is less than 10.0 cm10.0\text{ cm}.

    Correct

    Since P=0.9772>0.05P = 0.9772 > 0.05, fail to reject H0H_0. There is insufficient evidence to conclude that the mean calibration measurement is less than 10.0 cm10.0\text{ cm}.

    Using your calculated PP-value led to an incorrect decision. Comparing the true PP-value 0.97720.9772 to α=0.05\alpha = 0.05 shows P>αP > \alpha, so we fail to reject H0H_0. Notice that an observed sample mean of 10.6 cm10.6\text{ cm} (z=+2.00z = +2.00) cannot provide evidence that the true mean is less than 10.0 cm10.0\text{ cm}.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly calculated the upper-tail probability 1−0.9772=0.02281 - 0.9772 = 0.0228.

You followed the correct statistical decision logic of rejecting the null hypothesis whenever P<αP < \alpha.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Treated one-tailed test as two-tailed

You overlooked the inequality sign in Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm} and defaulted to a two-tailed calculation.

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You assumed that tail areas must always be doubled, conflating two-tailed test procedures with directional one-tailed tests. In addition, you did not verify whether the sample evidence (barx=10.6>10.0\\bar{x} = 10.6 > 10.0) physically supported the claimed decrease before deciding to reject H0H_0.

REASONING PATH
Your ACEpath teacher’s suggested approach
Check the Alternative Hypothesis Direction
  • If HaH_a contains '<<', it is left-tailed (P=P(Z≤z)P = P(Z \le z)). If HaH_a contains '>>', it is right-tailed (P=P(Z≥z)P = P(Z \ge z)). Only double the tail when HaH_a contains '≠\ne'.

  • Always circle the inequality symbol in HaH_a before looking up or calculating tail areas.

Reality-Check the Sample Direction
  • A sample mean that is greater than μ0\mu_0 (z>0z > 0) provides no evidence in favor of Ha:μ<μ0H_a: \mu < \mu_0.

  • Verify that your observed statistic lies in the direction claimed by HaH_a. If it lies in the opposite direction, the one-tailed PP-value will exceed 0.500.50.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Identify the direction of the test from Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm}: the test is left-tailed, so the PP-value is the area to the left of the observed test statistic.

  2. Calculate the PP-value using the standard normal distribution: P-value=P(Z≤+2.00)=0.9772P\text{-value} = P(Z \le +2.00) = 0.9772.

  3. Compare the PP-value to α=0.05\alpha = 0.05: since 0.9772>0.050.9772 > 0.05, do not reject H0H_0.

Final answer

(a) The test is left-tailed (Ha:μ<10.0 cmH_a: \mu < 10.0\text{ cm}), so the PP-value is P(Z≤zobs)P(Z \le z_{\text{obs}}); the observed sample mean was higher than 10.0 cm10.0\text{ cm}, lying in the direction opposite to HaH_a. (b) P-value=P(Z≤+2.00)=0.9772P\text{-value} = P(Z \le +2.00) = 0.9772. (c) Since P=0.9772>0.05P = 0.9772 > 0.05, fail to reject H0H_0.

NEXT STEPS
Final Takeaways from Your Teacher
  • Review the distinction between one-tailed (<<, >>) and two-tailed (≠\ne) hypotheses.

  • Sketch the standard normal curve and shade the tail indicated by HaH_a before writing down PP-value formulas.

  • Practice checking whether xˉ\bar{x} falls above or below μ0\mu_0 as a quick sanity check on your conclusion.

Lock it in

Similar question

01

A test of H0:μ=25H_0: \mu = 25 versus Ha:μ<25H_a: \mu < 25 yields a test statistic z=+1.50z = +1.50. Find the PP-value and state the decision at α=0.05\alpha = 0.05.

Answer

P=P(Z≤+1.50)=0.9332P = P(Z \le +1.50) = 0.9332. Since P=0.9332>0.05P = 0.9332 > 0.05, fail to reject H0H_0.

02

A quality inspector tests H0:μ=100H_0: \mu = 100 against Ha:μ>100H_a: \mu > 100. A sample produces xˉ=98\bar{x} = 98 with standard error SE=1.0\text{SE} = 1.0, giving z=−2.00z = -2.00. Compute the PP-value and state whether this sample provides evidence for HaH_a.

Answer

P=P(Z≥−2.00)=1−P(Z≤−2.00)=1−0.0228=0.9772P = P(Z \ge -2.00) = 1 - P(Z \le -2.00) = 1 - 0.0228 = 0.9772. The sample provides no evidence for HaH_a since xˉ\bar{x} is below 100100 and P>0.05P > 0.05.

03

An experiment tests H0:μ=0H_0: \mu = 0 against Ha:μ<0H_a: \mu < 0. An analyst mistakenly calculates P=0.01P = 0.01 by finding the upper tail of z=+2.33z = +2.33. Explain why this calculation is contradictory and provide the correct PP-value.

Answer

The test is left-tailed, so P-value=P(Z≤zobs)P\text{-value} = P(Z \le z_{\text{obs}}). Observing z=+2.33z = +2.33 means the sample mean is positive, which directly contradicts a claim that μ<0\mu < 0. The correct PP-value is P(Z≤+2.33)=0.9901P(Z \le +2.33) = 0.9901.

Loading workspace…