Large-Sample Statistical Test for  Example 1. Example 1.

Extension Tasks

a binomial population, the sample proportion 𝒑 has an

• When a random sample of n identical trials is drawn from Problem: Regardless of age, about 20% of Filipino adults 1. Compute the test statistic using the alternative formula.

Then, q0 = 1 – .20 = .80 𝒛 = 𝟏𝟓 −𝟏𝟎𝟎(.𝟐𝟎)

participate in fitness activities at least twice a week. Given: p0 = .20, n = 100, x = 15

mean 𝒑 and standard error 𝟏𝟎𝟎 .𝟐𝟎 (.𝟖𝟎) = −𝟏. 𝟐𝟓

approximately normal distribution when n is large, with However, these fitness activities change as the people get
older, and occasionally participants become

𝑝 = 15
nonparticipants as they age. In a local survey of n = 100 Ƹ

• Null hypothesis: H0: 𝒑 = 𝒑𝟎

adults over 40 years old, a total of 15 people indicated that
they participated in a fitness activity at least twice a week. 100 = .15

Two-tailed test: H1: 𝒑 ≠ 𝒑𝟎

• Alternative hypothesis: Do these data indicate that the participation rate for adults 2. Calculate the p-value and use it to make the decision.

Left-tailed test: H1: 𝒑 < 𝒑𝟎

over 40 years old is significantly less than the 20% figure? The p-value associated to zc = -1.25 is .1056.

Right-tailed test: H1: 𝒑 > 𝒑𝟎

( = .10) Since p = .1056 >  = .10, then there is no
sufficient evidence to reject the null hypothesis.
• Test statistic: Hence, there is insufficient evidence to conclude
Step 1. State the null and alternative hypotheses.
that the percentage of Filipino adults over age 40
and 𝒒 = 𝟏 − 𝒑
with Ho: p = .20 H1: p < .20
who participate in fitness activities twice a week is
Step 2. Select the level of significance
less than 20%.
• Decision Rule: Reject H0: 𝒑 = 𝒑𝟎 when
where x is the number of successes in n binomial trials.  = .10
Step 3. Determine the critical value and rejection region 3. Compute the test statistic when n = 400 and the same

Two-tailed test: 𝒛 > 𝒛𝒂/𝟐 or 𝒛 < −𝒛𝒂/𝟐

• Alternative hypothesis:
Given: p0 = .20, n = 400, Ƹ 𝑝 = .15
sample proportion.

Left-tailed test: 𝒛 < 𝒛𝒂 Step 4. State the decision rule

Then, q0 = 1 – .20 = .80 𝒛 = .𝟏𝟓 −.𝟐𝟎
Right-tailed test: 𝒛 > 𝒛𝒂 Reject the H0 if zc < zt = -1.28.

𝟒𝟎𝟎
.𝟐𝟎 (.𝟖𝟎)

= −𝟐. 𝟓𝟎
Assumptions Step 5. Compute the test statistic .

Then, q0 = 1 – .20 = .80 𝒛 = 𝟏𝟓 − .𝟐𝟎

• The experiment consists of n identical trials. Given: p0 = .20, n = 100, x = 15
• Each trial has one of two outcomes. One outcome is The p-value associated to z = -2.50 is .0062.

𝟏𝟎𝟎
called a success while the other is a failure. .𝟐𝟎 (.𝟖𝟎) Since p = .0062 <  = .10, then there is sufficient

= −𝟏. 𝟐𝟓
• The probability of success in a single trial is equal to p evidence to reject the null hypothesis. Hence,
and remains the same from trial to trial. The probability of there is sufficient evidence to conclude that the

𝑝 = 15
failure is q = 1 – p. Ƹ percentage of Filipino adults over age 40 who
• The trials are independent from each other. participate in fitness activities twice a week is
• The number of successes observed during n trials is x = 0, 100 = .15 less than 20%.
Step 6. Make a decision.
• ��ෞ𝑝0 > 5 and ��ෞ𝑞0 > 5
1, 2,…, n.
Since zc = -1.25 > zt = -1.28 , then there is no sufficient
evidence to reject the null hypothesis. Hence, there
𝒛 = 𝒙 − 𝒏𝒑𝟎
Alternative Formula for the Test statistic:
is insufficient evidence to conclude that the percentage
𝒏𝒑𝟎𝒒𝟎 of Filipino adults over age 40 who participate in fitness
activities twice a week is less than 20%.
Checkpoint 1 two binomial
populations, the focus of the experiment may be the
Problem: A random sample of n = 1000 observations from
difference (p1 – p2) in
a binomial population produced x = 279.
the proportions of individuals or items possessing a
1. If your research hypothesis is that p is less than .30,
specified characteristic
what should you choose as your alternative hypothesis?
in the two populations. In this situation, you can use the
Null hypothesis?
difference in the
2. What is the critical value that determines the rejection
sample proportions ෝ𝒑𝟏 − ෝ𝒑𝟐 along with its standard
region for your test with  = .05?

𝑺𝑬 = 𝒑𝟏𝒒𝟏
error,
3. Do the data provide sufficient evidence to Indicate that

𝒏𝟏
p is less than .30? ( = .05)

+ 𝒑𝟐𝒒𝟐
4. Calculate the corresponding p-value.

Checkpoint 2 𝒏𝟐

• Null hypothesis: H0: 𝒑𝟏 = 𝒑𝟐 or (𝒑𝟏 −𝒑𝟐) = 𝟎

Problem: A random sample of n = 1400 observations from .
a binomial population produced x = 529.

Two-tailed test: H1:(𝒑𝟏 −𝒑𝟐) ≠ 𝟎

1. If your research hypothesis is that p differs from .40, • Alternative hypothesis:

Left-tailed test: H1: (𝒑𝟏 −𝒑𝟐) < 𝟎

what should you choose as your alternative hypothesis?

Right-tailed test: H1: (𝒑𝟏 −𝒑𝟐) > 𝟎

Null hypothesis?
2. What is the critical value that determines the rejection

Test statistic: 𝒛 = ෝ𝒑 𝟏−ෝ𝒑𝟐

region for your test with  = .01?
3. Do the data provide sufficient evidence to

𝒏𝟏𝒏𝟐 )
Indicate that p is different from .40? ( = .01) ෝ𝒑ෝ𝒒(𝒏𝟏+𝒏𝟐

with ෝ𝒑𝟏 = 𝒙𝟏
4. Calculate the corresponding p-value.

Checkpoint 3 𝒏𝟏
Problem: A random sample of n = 120 observations from a and ෝ𝒑𝟐 = 𝒙𝟐
binomial population produced x = 72. 𝒏𝟐
1. If your research hypothesis is that p greater than .50,
• Decision Rule: Reject H0: 𝒑𝟏= 𝒑𝟐 when
where x is the number of successes in n binomial trials.
what should you choose as your alternative hypothesis?
Null hypothesis?
• Two-tailed test: 𝒛 > 𝒛𝒂/𝟐 or 𝒛 < −𝒛𝒂/𝟐
• Alternative hypothesis:
2. What is the critical value that determines the rejection
region for your test with  = .10? • Left-tailed test: 𝒛 < 𝒛𝒂
3. Do the data provide sufficient evidence to • Right-tailed test: 𝒛 > 𝒛𝒂
Indicate that p is greater than .50? ( = .10)
• Assumption: 𝑛1ෞ𝑝1, 𝑛1ෞ𝑞1, 𝑛2ෞ𝑝2, and 𝑛2ෞ𝑞2 must
• p-value is < 
4. Calculate the corresponding p-value.
Example 2.
all
Comparing 2 binomial proportions Problem: The records of a hospital shows that 52 men in a
be greater than 5.
• When random and independent samples are drawn from sample of 1000 men versus 23 women in a sample of 1000
women were admitted because of heart disease. Do these Hence, there is a higher rate of heart disease
data present sufficient evidence to indicate that a higher among men admitted to the hospital.
rate of heart disease among men admitted to the hospital?
Checkpoint 4
Use  = .05.
Problem: Independent random samples of n1 = 140 and n2
Ho: 𝒑𝟏 = 𝒑𝟐 H1: 𝒑𝟏 > 𝒑𝟐
Step 1. State the null and alternative hypotheses.
= 140 observations were randomly selected from binomial
populations 1 and 2, respectively. Sample 1 had 74
Step 2. State level of significance and critical value.
successes, and sample 2 had 81 successes.
 = .05
1. Suppose you simply want to detect only whether a
1.645
difference exist between the two parameters, p1 and p2,
Step 3. State the decision rule.
what must be your alternative hypothesis? Null
Reject Ho if z > 1.645.
hypothesis?
Step 4. Calculate the test statistic.
2. Find the standard error of the difference
Given: x1 = 52, n1 = 1000, x2 = 23 n2 = 1000
in the sample proportions, ෝ𝒑𝟏 − ෝ𝒑𝟐 . Use
The pooled estimate of p required for the standard error is
the pooled estimate.
𝑝 = 52+23
Ƹ
3. Calculate the test statistic and make a decision. ( = .01)
4. Test the significant difference using the
1000+1000 = .0375 so ො𝑞 = .9625
p-value approach. ( = .01)
𝑝1 = 52
Ƹ
Checkpoint 5
1000 = .052 Problem: Independent random samples of n1 = 140 and n2

𝑝2 = 23
Ƹ = 140 observations were randomly selected from binomial
populations 1 and 2, respectively. Sample 1 had 74

𝒛 = . 𝟎𝟓𝟐−. 𝟎𝟐𝟑
1000 = .023 successes, and sample 2 had 81 successes.

. 𝟎𝟑𝟕𝟓 . 𝟗𝟔𝟐𝟐 [ 𝟏𝟎𝟎𝟎 + 𝟏𝟎𝟎𝟎

1. Suppose p1 cannot be larger than p2, what must be

𝟏𝟎𝟎𝟎 𝟏𝟎𝟎𝟎 ]
your alternative hypothesis? Null hypothesis?

= 𝟑. 𝟒𝟏
2. Find the standard error of the difference
in the sample proportions, ෝ𝒑𝟏 − ෝ𝒑𝟐 . Use
Step 5. Make a decision. the pooled estimate.
Since zc = 3.41 > zt = 1.645, then there is sufficient 3. Calculate the test statistic and make a decision. ( = .05)
evidence to reject the null hypothesis that p1 = p2. Hence, 4. Test the significant difference using the p-value
the percentage of men entering the hospital because of approach. ( = .05)
heart disease is higher than that of women.
Step 6. Compute the p-value and make a decision.
The p-value associated with zc = 3.41 is .0003
Since p = .0003 <  = .005, then there no
sufficient evidence to reject the null hypothesis.