HYPOTHESIS TESTING

USING P-VALUE
APPROACH
What is p-value?
• The p-value (probability value) of test
statistic.
• It is the smallest probability value at which
the null hypothesis would be rejected. 
• A smaller p-value indicates that there is
stronger evidence against the null
hypothesis.
 Comparing the P-Value Method to the
Traditional Method
Both approaches will ensure the same conclusion and
either one will work. However, using the p-value approach
has the following advantages:

1. In p-value method, there is no need to compare the test

p-value = P(ztest < zcomp) p-value = P(ztest > zcomp)

left right

p-value = 2 x P(ztest < - zcomp) or 2 x P(ztest > zcomp)

Reject H0 if p ≤ α. Otherwise,
do not reject H0 .
Steps using the P-value Approach
1. Null and Alternative Hypotheses
2. Type of Test, α level
3. Test Statistic (z or t), p-value
4. Decision Rule: Reject H0 if p ≤ α. Otherwise, do not
reject H0 .
5. Compare p-value with α and decide whether to reject
H0 or not.
6. Conclusion
Note: Only z-test will be used
for all examples in this module
since a graphing calculator is
needed to get the p-value for t-
test.
Example 1:
A dietitian claims that the average weight loss
of women using a certain drug is 4.55 kg per
month. Ten women chosen at random have
reported losing an average of 4.05 kg within a
month. If the population standard deviation is
0.91 kg, does this data support the claim of the
dietitian? Use 0.05 level of significance.
 Step p-value = 2 x P(z < -1.74)
1: Hypotheses
= 2(0.5 - 0.4591)
H0 : The average weight loss per
= 2(0.0409)
month of women using the drug is
4.55 kg/mo. (µ = 4.55) Claim p-value = 0.0818
Step 4: Decision rule
H1 : The average weight loss per
month of women using the drug is Reject H0 if p ≤ 0.05. Otherwise, do not
not equal to 4.55 kg/mo. (µ ≠ 4.55) reject H0.
Step 5: Decision
Since 0.0818 > 0.05, do not reject H0.
Step 2: Type of test
(Weak evidence against H0.)
two-tailed, α = 0.05
Step 6: Conclusion
There is no significant difference between
Step 3: Test Statistic and p-value the sample mean and the population mean.
Thus, the average weight loss per month of
women using the drug is 4.55kg.
Example 2:
The dean claims that the average height
of HRM students is more than 155 cm. To
verify this, a random sample of 100 HRM
students of a certain college was selected
and their average height is 157 cm.
Previous study shows that σ = 5 cm. Use α
= 0.01 to test the claim.
 Step 1: Hypotheses p-value = P(z > 4) = (0.5 – 0.4999)
p-value = 0.0001
H0 : The average height of HRM
students is less than or equal to 155 Step 4: Decision rule
cm. (µ ≤ 155) Reject H0 if p ≤ 0.01. Otherwise, do not
H1 : The average height of HRM reject H0.
students is more than 155 cm. Step 5: Decision
(µ > 155) Claim Since 0.0001 < 0.01, reject H0.
(very strong evidence against H0.)
Step 2: Type of test Step 6: Conclusion
right-tailed, α = 0.01 There is a significant difference
between the sample mean and the
Step 3: Test Statistic and p-value population mean.
Thus, the average height of HRM
students is more than 155 cm.
 Activity 7: P-value Method
1. In a plant nursery, a botanist thinks that the seedlings sprayed with a
new kind of fertilizer has an average height of 26 cm after three days.
The botanist randomly selected 80 such seedlings and obtained a
mean height of 25 cm. Assume that the population standard
deviation is 10 cm. Test the claim using α = 0.01 and p-value method.
2. A manufacturing firm has been averaging 18 orders per week with a
population standard deviation of 2.3 orders for several years.
However, during a recession, orders appear to have slowed down.
Suppose the firm’s production manager randomly selects 32 weeks
and finds a sample mean of 15 orders per week. Use α = 0.05 and
the p-value approach to determine whether the average number of
orders has decreased during recession.
 Step 1: Hypotheses Step 4: Decision rule
H0 : µ = 26 Claim Reject H0 if p ≤ 0.05. Otherwise, do not
H1 : µ ≠ 26 reject H0.
Step 2: Type of test Step 5: Decision
two-tailed, α = 0.01 Since 0.3734 > 0.05, do not reject H0.
Step 3: Test Statistic and p-value (weak evidence against H0)
Step 6: Conclusion
p-value = 2 x P(z < -0.89) There is no significant difference
= 2(0.5 - 0.3133) between the sample mean and the
population mean.
p-value = 0.3734
Therfore, the average height of the
seedlings sprayed with new fertilizer is
equal to 26 cm after three days.
 Step 1: Hypotheses
P(z < -7.37) = 0.5 – 0.4999
H0 : The average number of orders
P-value = 0.0001
during recession greater than or
equal to 18 orders/wk Step 4: Decision rule
(µ ≥ 18) Reject H0 if p ≤ 0.05. Otherwise, do not
reject H0.
H1 : The average number of orders
during recession is lower than 18 Step 5: Decision
orders/wk. (µ < 18) Claim Since 0.0001 < 0.05, reject H0.
Step2: Type of test (Very strong evidence against H0.)
left tailed, α = 0.05 Step 6: Conclusion
Step 3: Test Statistic There is a significant difference between
the sample mean and the population mean.
Thus, the average number of orders per
week during recession is lower than 18.