Resources
A sociologist tests versus for the proportion of urban commuters using public transit. The study fixes the significance level at . (a) Determine the exact critical value that marks the boundary between rejecting and failing to reject . (b) If a sample of size produces an observed test statistic of exactly , find the -value to four decimal places and state whether the researcher rejects or fails to reject at .
Answer and steps
Final answer
(a) Critical value: ; (b) -value: . Since , fail to reject .
Steps
Part (a): Finding the Critical Value
- Identify the direction of the test:
The alternative hypothesis is , which indicates a one-tailed (right-tailed) test. - Locate the critical boundary:
The significance level is . For a right-tailed test, the critical region lies in the upper tail of the standard normal distribution , such that: Equivalently, the cumulative probability to the left of is: -
Determine from the standard normal table:
Looking up an area of in a standard normal table:- For , the cumulative area is .
- For , the cumulative area is .
Since lies exactly halfway between these values, the conventional critical value is:
Part (b): Finding the -Value and Making the Decision
- Calculate the -value:
The observed test statistic is . Because this is a right-tailed test, the -value is the probability of observing a -score at least as extreme as under the null hypothesis: From the standard normal cumulative distribution table, . Therefore: -
Compare with the significance level :
Since , the -value is greater than (and the observed test statistic does not exceed the critical threshold ).
- Conclusion:
We fail to reject at the significance level. There is insufficient evidence to conclude that the proportion of urban commuters using public transit exceeds .
Common Pitfall: A test statistic of is very close to , but in hypothesis testing, failing to meet the strict threshold means we do not reject . Always compare to directly: rejection requires .
Typical errors
01 The student treats the hypothesis test as two-tailed instead of right-tailed, splitting alpha to find the critical value and doubling the tail area for the P-value.
Sample Student Work
Reviewed by ACEpath TeacherStep-by-Step Review
Walk Through It with Your Teacher- Step 1Correct
The alternative hypothesis specifies a right-tailed test, so the entire rejection area belongs in the upper tail rather than being divided into two tails.
- Step 2Correct
You correctly found the tail probability , but you doubled it as if the test were two-tailed. For a right-tailed test (), the -value is simply the single tail area.
- Step 3Correct
You correctly applied the rule that leads to failing to reject , but your comparison relied on the doubled -value from the previous step.
WHAT YOU DID WELL
Compliments from Your TeacherYou correctly calculated the upper-tail probability .
You correctly applied the decision rule to fail to reject when the -value exceeds .
THE ERROR
Where Your ACEpath Teacher Spotted the MistakeNo, this was not a simple arithmetic slip; you treated a directional hypothesis test as two-tailed throughout both parts of the problem.
WHY IT HAPPENED
ACEpath Teacher's DiagnosisYou divided by and doubled the tail area because you defaulted to the procedure for a two-tailed test () instead of checking the inequality in the alternative hypothesis ().
REASONING PATH
Your ACEpath teacher’s suggested approach- Direction of the Test
is right-tailed (use all of in the upper tail); is left-tailed; is two-tailed (split and double the tail area).
Look at the inequality sign in before splitting or multiplying tail probabilities.
CORRECTED PATH
ACEpath Teacher's DemonstrationIdentify that indicates a one-tailed (right-tailed) test with significance level .
Set , which gives cumulative area and critical value .
Calculate the -value as the single right-tail probability: .
Compare the -value directly to : since , fail to reject .
(a) Critical value: ; (b) -value: ; fail to reject .
02 The student calculates the cumulative left-tail probability Φ(z) instead of the upper-tail probability for a right-tailed test, leading to an incorrect decision to reject H₀.
Sample Student Work
Reviewed by ACEpath TeacherStep-by-Step Review
Walk Through It with Your Teacher- Step 1
- Step 2Correct
Because the alternative hypothesis is right-tailed (), the -value is the area in the upper tail, , not the cumulative left-tail probability .
- Step 3Correct
The decision rule states that we reject only when the -value is less than or equal to (). Since the correct -value is , we fail to reject .
WHAT YOU DID WELL
Compliments from Your TeacherYou correctly set up the upper-tail condition to find the critical value.
You looked up the correct standard normal table probability corresponding to .
THE ERROR
Where Your ACEpath Teacher Spotted the MistakeYou read directly from the standard normal table without subtracting it from for an upper-tail test.
WHY IT HAPPENED
ACEpath Teacher's DiagnosisYou treated the cumulative left-tail probability as the -value in a right-tailed test and reversed the rejection criteria ( rejects ).
REASONING PATH
Your ACEpath teacher’s suggested approach- Tail Direction for P-Values
For a right-tailed test (), .
Always align the inequality of the -value calculation with the inequality in .
- P-Value Decision Rule
Reject if and only if .
Remember the rhyme 'If is low, must go.' When , you must fail to reject .
CORRECTED PATH
ACEpath Teacher's DemonstrationCalculate the upper-tail probability: .
Compare the -value to : .
Apply the decision rule: because , do not reject the null hypothesis.
Fail to reject (with ).
03 The student reverses the hypothesis testing decision rule, incorrectly deciding to reject the null hypothesis whenever the P-value is greater than alpha.
Sample Student Work
Reviewed by ACEpath TeacherStep-by-Step Review
Walk Through It with Your Teacher- Step 1
- Step 2
- Step 3
, so reject
CorrectSince , fail to reject .
You reversed the statistical decision rule. We reject the null hypothesis only when the -value is less than or equal to the significance level (). Because , the evidence is not strong enough to reject , so you must fail to reject .
WHAT YOU DID WELL
Compliments from Your TeacherYou correctly identified that the test is right-tailed and found the critical value .
You accurately found the cumulative normal probability and computed the -value as .
THE ERROR
Where Your ACEpath Teacher Spotted the Mistakeno
WHY IT HAPPENED
ACEpath Teacher's DiagnosisYou treated a -value greater than as evidence against the null hypothesis, confusing the criterion for rejecting () with failing to reject ().
REASONING PATH
Your ACEpath teacher’s suggested approach- The P-Value Decision Rule
If , reject . If , fail to reject .
Remember the rhyme: 'If the is low, must go; if the is high, will fly (keep ).'
- Comparing Test Statistic to Critical Value
For a right-tailed test, reject only if .
Cross-check with the critical value: here , confirming that the test statistic does not fall into the rejection region.
CORRECTED PATH
ACEpath Teacher's DemonstrationCompare the calculated -value to the significance level: and .
Apply the decision rule: because , the sample result is not unusual enough under to justify rejection.
State the conclusion: fail to reject at the significance level.
(a) ; (b) . Since , fail to reject .
Lock it in
How does the alternative hypothesis determine whether a -test is one-tailed or two-tailed?
is right-tailed and is left-tailed (one-tailed). Only is two-tailed.
Similar question
A researcher tests versus at . Find the critical value and determine the -value if the observed test statistic is .
Answer
Critical value: . Since this is a left-tailed test, . Reject because .
A quality control manager tests versus at . An observed test statistic is . Find the critical values and the -value.
Answer
Critical values: (since ). Because this is a two-tailed test, . Fail to reject because .
In a one-tailed test of versus , a student mistakenly computed a two-tailed -value of and concluded to fail to reject at . What was the correct one-tailed -value, and what decision should have been made?
Answer
The one-tailed -value is half of the two-tailed value: . Since , the researcher should reject at .