Chapter 8: Hypothesis Testing for

Population Proportions
Testing a claim
● Earlier we calculated a 95% confidence interval for the
true population proportion of UCLA students who
travelled outside the US.
(0.26, 0.44)
● A 95% confidence interval of 26% to 44% means that
● We are 95% confident that the true population

proportion of UCLA students who travelled outside the

US is between 26% and 44%.
● 95% of random samples of size n = 100 will produce

confidence intervals that contain the true population

proportion.
Testing a claim
● The true population proportion, p, may be outside the interval,
but we would expect it to be somewhat close to .
● In our random sample of 100 students we had found that 35 of
them have at some point in their lives travelled outside the US,
= 0.35.
● Do you think it's possible that:
● p = 0.90?
● p = 0.05?

● p = 0.34 or p = 0.36?

● p = 0.25 or p = 0.45?

● It is difficult to decide how close is close enough, or how far is

too far, and this decision should not be made subjectively.
Hypothesis Testing
● In Statistics when testing claims we use an objective
method called hypothesis testing
● Given a sample proportion, , and sample size, n, we
can test claims about the population proportion, p.
● We call these claims hypotheses
● Our starting point, the status quo, is called the null
hypothesis and the alternative claim is called the
alternative hypothesis
Hypothesis Testing
● If our null hypothesis was that p = 0.35 and our sample
yields = 0.35, then the data are consistent with the null
hypothesis, and we have no reason to not believe this
hypothesis.
● This doesn't prove the hypothesis but we can say that the

data support it.

● If our null hypothesis was different than p = 0.35, let's say
p = 30 and our sample yields = 0.35, then the data
are not consistent with the hypothesis and we need to make
choices as to whether this inconsistency is large enough to
not believe the hypothesis.
● If the inconsistency is significant, we reject the null

hypothesis
Hypothesis Testing for One Proportion
Research conducted a few years ago showed that 35% of UCLA students
had travelled outside the US. UCLA has recently implemented a new
study abroad program and results of a new survey show that out of the
100 randomly sampled students 42 have travelled abroad. Is there
significant evidence to suggest that the proportion of students at UCLA
who have travelled abroad has increased after the implementation of the
study abroad program?
● Population proportion used to be 0.35.
● New sample proportion is 0.42.
● We are testing the claim that the population proportion is now
greater than 0.35.
● The new data indicate that it may be since 0.42 is clearly grater than
0.35.
● But is this difference statistically significant, i.e. are the data
inconsistent enough?
● We do a formal hypothesis test to answer this question.
Setting Up Hypotheses
● Null hypothesis, denoted by Ho, specifies a population model
parameter of interest and proposes a value for that parameter
(p).
Ho: p = 0.35
● Alternative hypothesis, denoted by HA, is the claim we are
testing for.
HA: p > 0.35
● Even though we are testing for the alternative hypothesis, we
check to see whether or not the null hypothesis is plausible.
● If the null hypothesis is not plausible, we reject the null
hypothesis and conclude that there is sufficient evidence to
support the alternative. If the null hypothesis is plausible, we fail
to reject the null hypothesis and conclude that there isn't
sufficient evidence to support the alternative.
Why do we check if Ho is plausible and
not if HA is plausible?
● Think about the logic of jury trials:
● To prove someone is guilty, we start by assuming they are innocent.
● We retain that hypothesis until the facts make it unlikely beyond a
reasonable doubt.
● Then, and only then, we reject the hypothesis of innocence and declare
the person guilty.
● The same logic used in jury trials is used in statistical tests of hypotheses:
● We begin by assuming that the null hypothesis is true.
● Next we consider whether the data are consistent with this hypothesis.
● If they are, all we can do is retain the hypothesis we started with. If
they are not, then like a jury, we ask whether they are unlikely beyond
a reasonable doubt.
A Trial as a Hypothesis Test
● Hypothesis testing is very much like a court trial.
● This doesn't prove the hypothesis but we can say that the
data support it.
● Ho: Defendant is innocent
HA: Defendant is guilty
● We then present the evidence - collect data.
● Then we judge the evidence - “Could these data plausibly have
happened by chance if the null hypothesis were true?”
● If they were very unlikely to have occurred, then the evidence
raises more than a reasonable doubt in our minds about the
null hypothesis.
● Ultimately we must make a decision. How unlikely is unlikely?
A Trial as a Hypothesis Test
● If the evidence is not strong enough to reject the presumption of
innocent, the jury returns with a verdict of “not guilty”
● The jury does not say that the defendant is innocent.
● All it says is that there is not enough evidence to convict, to

reject innocence.
● The defendant may, in fact, be innocent, but the jury has no

way to be sure.
● Said statistically, we fail to reject the null hypothesis.
● We never declare the null hypothesis to be true, because we
simply do not know whether it's true or not.
● Therefore we never“accept the null hypothesis”
A Trial as a Hypothesis Test
● In a trial, the burden of proof is on the prosecution.
● In a hypothesis test, the burden of proof is on the
unusual claim.
● The null hypothesis is the ordinary state of affairs (the
status quo), so it's the alternative hypothesis that we
consider unusual (and for which we must gather
evidence).
How do we determine if Ho is plausible?
● In Statistics we can quantify our level of doubt.
● How unlikely is it to get a random sample of 100

students where 42 have travelled abroad if in fact the

true population proportion is 35%?
● To answer this question we use the model proposed

by the null hypothesis as a given and calculate the

probability that the event we have witnessed could
happen.
Prob(observed or more extreme outcome | H0 true)
= Prob( > 0.42 | p = 0.35)
● This probability quantifies exactly how surprised we
are to see our results and is called the p-value
Calculating the p-value.
● First, we calculate the test statistic (z-score)
● The test statistic used for hypothesis testing for
proportions is a z-score.

● Remember the CLT: In calculation of the z-score, we use

the mean and the SD suggested by the CLT, and the
observed value is the

Note: The p comes from the null hypothesis, i.e. it is value proposed for
the parameter in the null hypothesis.
Check the Conditions for the CLT
● Random and Independent: The sample is collected
randomly and the trials are independent of each other.
● Large Sample:
● The sample has at least 10 successes, np ≥ 10, and

● at least 10 failures n (1 – p) ≥ 10.

● Large Population: If the sample is collected without

replacement, then the population size is at least 10 times
the sample size.
Calculating the p-value.
Decision based on the p-value
● When the data are consistent with the model from the null
hypothesis, the p-value is high and we are unable to reject the
null hypothesis.
● In that case, we have to“retain” the null hypothesis.
● We can't claim to have proved it; instead we fail to reject the

null hypothesis and conclude that the difference we observed

between the null hypothesis (p = 0.35) and the observed
outcome ( = 0.42) is due to natural sampling variability
(or chance).
● If the p-value is low enough, we reject the null hypothesis since
what we observed would be very unlikely if in fact the null
model was true.
● We call such results statistically significant
How low a p-value is low enough?
● We compare the p-value to a given α:
If p-value < α → Reject Ho
There is sufficient evidence to suggest that HA is plausible.
If p-value > α → Fail to reject Ho
There is not sufficient evidence to suggest that HA is plausible.
The difference we are seeing between the null model and the
observed outcome is due to natural sampling variability.
● When p-value is low, it indicates that obtaining the observed or
even a more extreme outcome is highly unlikely under the assumption
that Ho is true, therefore we reject that assumption. α level is the
complement of the confidence level.
Remember: When constructing confidence intervals if a confidence
level is not specified use 95% confidence.
● Since 1 − 0.95 = 0.05, if a α level is not specified use α = 0.05.
Decision based on the p-value
With a p-value of 0.0708, do we reject or fail to reject Ho?
a) Reject Ho
b) Fail to reject Ho
Decision based on the p-value
What does the conclusion of the hypothesis mean in context of the
research question?
a) The data provide convincing evidence to suggest that the
true population proportion of UCLA students who have
travelled outside the US is 0.35.
b) The data provide convincing evidence to suggest that the
true population proportion of UCLA students who have
travelled outside the US is 0.42.
c) The data do not provide convincing evidence to suggest
that the true population proportion of UCLA students who
have travelled outside the US has increased.
d) The data do not provide convincing evidence to suggest
that the true population proportion of UCLA students who
have travelled outside the US has not changed.
Interpreting the p-value
● A p-value is a conditional probability - the probability of the
observed or a more extreme outcome given that the null
hypothesis is true.
Prob(observed or more extreme outcome | Ho true)

● The p-value is NOT the probability that the null hypothesis is

true.
● It's not even the conditional probability that the null hypothesis is
true given the data.
Interpreting the p-value
● Which of the below is the correct interpretation of the p-value?
a) The probability that 35% of UCLA students have travelled
outside the US is 0.0708.
b) The probability that more than 35% of UCLA students have
travelled outside the US is 0.0708.
c) If in fact the true population proportion of UCLA students
who have travelled outside the US is greater than 0.35, the
probability of getting a random sample where the sample
proportion is 0.42 or higher would be 0.0708.
d) If in fact the true population proportion of UCLA students
who have travelled outside the US is 0.35, the probability of
getting a random sample where the sample proportion is
0.42 or higher would be 0.0708.
Alternative Hypotheses
● When testing for population proportions, there are three possible
alternative hypotheses:
● HA: p < po
● HA: p > po

● HA: p ≠ po

● We decide on which alternative hypothesis to use based on we

hypothesize (or what the wording of the question instructs as to
hypothesize):
● Smaller, less, decreased, fewer
● Larger, greater, more, increased
● Different, not equal to, changed
● We DO NOT decide on which alternative hypothesis to use
based on what the data suggest.
Alternative Hypotheses
● HA: parameter ≠ hypothesized value is known as a two-sided
alternative because we are equally interested in deviations on
either side of the null hypothesis value.
● For two-sided alternatives, the p-value is the probability of
deviating in either direction from the null hypothesis value.
● p-value is the sum of the shaded (red) areas.
Alternative Hypotheses
● The other two alternative hypotheses are called
● A one-sided alternative focuses on deviations from the null
hypothesis value in only one direction.
● Thus, the p-value for one-sided alternatives is the probability of
deviating only in the direction of the alternative away from the
null hypothesis value.
● p-value is the shaded (red) area.
Steps for a Hypothesis Test
Steps for a Hypothesis Test
Write the appropriate hypotheses
A company with a fleet of 150 cars found that the emissions
systems of 7 out of the 22 they randomly chose and tested
failed to meet pollution control guidelines. We would like to
test if this is strong evidence that more than 20% of the fleet
might be out of compliance. What are the appropriate
hypotheses?
Do a Hypothesis Test
A survey conducted five years ago by the health center at a
university showed that 18% of the students smoked at the
time. This year a new survey was conducted on a random
sample of 200 students from this university, and it was found
that 40 of them smoke. Do these data provide convincing
evidence to suggest that the percentage of students who
smoke has changed over the last five years?
Comparing 2 proportions
● Compare 2 sample proportions each from a different
group
● Are the two groups the same? Are they different?

● Same ideas as for 1 proportion, but the formulas change

● Confidence Interval for difference in proportions

● Hypothesis test for difference in proportions

Remember: CI for 1 proportion.
Depending on the confidence level, we choose how many
SEs to add/subtract to/from . So, a more generic
formula for the CI can be written as

where

and z* is a critical z-score that depends on the confidence

level.
Confidence Intervals for the difference in
proportions
When we have two populations to compare, we can also find
a confidence interval for the difference between the two
population proportions:

where z* is still the critical z-score that depends on the

confidence level. The numbers 1 and 2 indicate the two
different groups.
Confidence Intervals for the difference in
proportions
A recent survey asked San Diego residents the following
question:“If you had been aware that before
Schwarzenegger ran for Governor, he had fathered a child
outside of his marriage, would you still have voted or
him?”Out of the 136 males surveyed who had voted for
Schwarzenegger at least once 78 said they would still vote
for him, and out of the 128 females 48 said they would.
Calculate a 95% confidence interval for the difference
between the proportions of male and female SD residents
who have would have voted for Schwarzenegger even if they
knew that he had fathered a child outside of his marriage.
Confidence Intervals for the difference in
proportions
Calculate a 95% confidence interval for the difference
between the proportions of male and female SD residents
who have would have voted for Schwarzenegger even if they
knew that he had fathered a child outside of his marriage.
Confidence Intervals for the difference in
proportions
Randomization condition: We are told that the sample is random.
Large Sample: There are 78 successes and 58 failures in the male
group (both ≥ 10) and 48 successes and 80 failures in the female
group (also both ≥ 10). Since the success/failure condition is met, we
can assume that the distributions of males and females are nearly
normal.
Big Population (10% Condition): 136 < 10% of all male SD residents
who voted for Schwarzenegger before and 128 < 10% of all female
SD residents who voted for Schwarzenegger before.
Independent groups: The two groups should be independent of one
another. Males are independent of females, i.e., you can't be in both
groups.
Confidence Intervals for the difference in
proportions

We are 95% confident that the proportion of male SD residents

who would have voted for Schwarzenegger even if they knew
that he had fathered a child outside of his marriage is 13% to
25% higher than the proportion of females.
Determining if there is a difference in
proportions between groups
If the confidence interval for the difference between two
population proportions includes 0, then it is possible that the
difference between the two population proportions equals 0, i.e.
that there is no significant difference between the two
population proportions.
95% CI : (13%, 25%)
The confidence interval does not include 0, which means that we
are 95% confident that the two population proportions are not
equal; that proportion of males that would have voted for
Schwarzenegger is higher than the proportion of females.
CI for the difference in proportions
We buy a bag of Skittles and M&M's. The bag of Skittles has
24% red out of a total 65 Skittles. The bag of M&M's has 19%
red out of a total of 55 M&M's. Construct a 90% confidence
interval for the difference in percent red for Skittles versus
M&M's. Interpret the interval. Is there evidence for a difference
in proportions between the two candies?
CI for the difference in proportions
The number of Prius owners in a random sample of 500
Californians is 50. The number of Prius owners in random
sample of 1000 Iowans is 25. Create a 95% CI for the
difference in proportions of Californians and Iowans who own
Priuses. Is this evidence of a difference that proportion of Prius
owners is different in CA than in IA?
Hypothesis Testing for Two Proportions
Hypotheses: state the null and alternative hypotheses
Ho: p1 = p2
HA: p1 < p2 OR HA: p1 > p2 OR HA: p1 ≠ p2
Note:
If p1 = p2, then p1 − p2 = 0. This is always what goes in the null
hypothesis since we start by assuming that the two population means
are equal, i.e. the two samples come from the same population.
If p1 > p2, then p1 − p2 > 0.
If p1 < p2, then p1 − p2 < 0.
These are one-sided alternative hypotheses.

If p1 ≠ p2, then p1 − p2 < 0. This is a two-sided alternative

hypothesis.
Hypothesis Test for the difference in
proportions
A recent survey asked San Diego residents the following question:“If
you had been aware that before Schwarzenegger ran for Governor,
he had fathered a child outside of his marriage, would you still have
voted or him?”Out of the 136 males surveyed who had voted for
Schwarzenegger at least once 78 said they would still vote for him,
and out of the 128 females 48 said they would. Do these data
provide convincing evidence to suggest that there is a difference
between the proportion of male and female SD residents who would
have voted for Schwarzenegger had they known he had fathered a
child outside of his marriage?
Hypothesis Test for the difference in
proportions
Out of the 136 males surveyed who had voted for Schwarzenegger at
least once 78 said they would still vote for him, and out of the 128
females 48 said they would. Do these data provide convincing
evidence to suggest that there is a difference between the proportion
of male and female SD residents who would have voted for
Schwarzenegger had they known he had fathered a child outside of
his marriage?
Hypotheses: Ho: pM = pF
HA: pM ≠ pF

alternatively... Ho: pM − pF = 0
HA: pM − pF ≠ 0
Check the Conditions
The conditions are the same as in the CI for two proportions. We have
already checked these and found that the sampling distributions for
male and female voters follow a normal distribution.

We can proceed with the hypothesis test.

Combine the sample proportions
Combine the sample proportions
Calculate the test statistic (z-score)
Calculate the test statistic (z-score)
Find the p-value
Make a conclusion
Based on the p-value of 0.0020, which of the below is the most
appropriate conclusion for this hypothesis test?
(a) We can Reject Ho. There is evidence to suggest that males are
more likely ...
(b) Reject Ho. There is evidence to suggest that males and females are
not equally likely ...
(c) Fail to reject Ho. There males and females are equally likely ...
(d) Fail to reject Ho. There is evidence to suggest that males and
females are not equally likely ...
to have voted for Schwarzenegger even if they knew that he had
fathered a child outside of his marriage.
2 proportion hypothesis test example
A USA Today/Gallup poll surveyed 1,145 unemployed and 675
underemployed randomly sampled people. 27% of the unemployed
and 25% of the underemployed people said they had major problems
in relationships as a result of their employment status. Do the data
provide convincing evidence to suggest a difference between the true
population proportions of unemployed and underemployed people who
had such problems. Use a 5% significance level.
2 proportion hypothesis test example
Do the data provide convincing evidence to suggest a difference
between the true population proportions of unemployed and
underemployed people who had such problems. Use a 5% significance
level.