Introduction:

3.1 Hypothesis testing

Based on the activity, how do we define hypothesis? According to Oxford
Objective: Languages, hypothesis is defined a proposition made as a basis for
1. Identify null hypothesis reasoning, without any assumption of its truth.
2. Write null There are 2 types of hypothesis: 1. Null hypothesis hypothesis-states that
3. Identify alternate hypothesis there is no difference in the population. It can be denoted as 𝐻𝑜 . 𝜇 = 𝜇𝑜 2.
4. Write alternate hypothesis Alternate hypothesis- states that there is some difference in the population.
Motivation: It can be denoted as 𝐻𝑎 . 𝜇 ≠ 𝜇𝑎 .

Identify the commercials. Examples:

1. Is the average IQ score of all World Campus STAT 200 students

higher than the national average of 100?
a. Null hypothesis: 𝜇 = 100
b. Alternate hypothesis: 𝜇 > 100
2. Is the percent of students enrolled in Penn State's College of
Science who identify as women different from 50%?
3. Is the proportion of men who smoke cigarettes different from the
proportion of women who smoke cigarettes in the United States?
a. Null hypothesis: there is no difference between the
proportion of men who smoke cigarettes from the
proportion of women who smoke cigarettes in the United
States.
b. Alternate Hypothesis: there is a difference between the
What do these 2 commercials have in common? They both claim that
proportion of men who smoke cigarettes from the
their brands are better than brand x. but how do we really identify
proportion of women who smoke cigarettes in the United
whether this is true or not?
State
Activity: 2. Two-tailed test is a non-directional test and a two-sided test.
Meaning, whenever you conduct a two tailed test, you would split
Write a null and alternative hypothesis for the following problems:
the significance level into two.
1. A statistics instructor believes that fewer than 20% of Evergreen
Valley College (EVC) students attended the opening night
midnight showing of the latest Harry Potter movie. She surveys
84 of her students and finds that 11 attended the midnight
showing.
2. Over the past few decades, public health officials have examined
the link between weight concerns and teen girls' smoking.
Researchers surveyed a group of 273 randomly selected teen girls
living in Massachusetts (between 12 and 15 years old). After four
years the girls were surveyed again. Sixty-three said they smoked
to stay thin. Is there good evidence that more than thirty percent Technically speaking, the
of the teen girls smoke to stay thin? • Null hypothesis of a two tailed test is equal to x.
𝜇 = 𝜇𝑜
There are 2 tails in hypothesis test. • Alternate hypothesis of a two tailed test is not equal to x.
𝜇 ≠ 𝜇𝑜
1. One tailed test. It is also known as one-sided and directional test.
Examples: Identify the whether it is a one- tailed or two-tailed test.
This means that the entire significance level goes to the extreme
end of one tail of the distribution. 1. Is the average IQ score of all World Campus STAT 200 students
higher than the national average of 100? Answer: One-tailed
2. Is the percent of students enrolled in Penn State's College of
Science who identify as women different from 50%?
Answer: Two-tailed
3. Is the proportion of men who smoke cigarettes different from the
proportion of women who smoke cigarettes in the United States?

Answer: Two-tailed
Activity: Quiz

Identify the whether it is a one- tailed or two-tailed test. Write the correct null and alternative hypotheses then identify whether
you would use one or two tailed test.
1. A statistics instructor believes that fewer than 20% of Evergreen
Valley College (EVC) students attended the opening night 1. A mayor is concerned about the percentage of city residents who
midnight showing of the latest Harry Potter movie. She surveys express disapproval of his job performance. His political
84 of her students and finds that 11 attended the midnight committee pays for a newspaper ad, hoping to keep his
showing. disapproval rating below 21%. They will use a follow up poll to
2. Over the past few decades, public health officials have examined access effectiveness.
the link between weight concerns and teen girls' smoking. 2. The weight of the average 6th grade student's backpack (with
Researchers surveyed a group of 273 randomly selected teen girls books in it) is 18.4 lbs. The principal of the school thinks that the
living in Massachusetts (between 12 and 15 years old). After four backpack does not weight 18.4 lbs.
years the girls were surveyed again. Sixty-three said they smoked
to stay thin. Is there good evidence that more than thirty percent
of the teen girls smoke to stay thin?
 Ariel is a 16 year old girl who is 66.41 in tall. For girls her age, their
3.2 Z-test average height is 64 in and a standard deviation of 2.66.
Introduction: 1. How many 15 year old girls are shorter than Jessica?
Ms. Gonzales grades his introductory statistics class on a curve. Let’s (𝑥 − 𝜇)
𝑧=
suppose that the average grade in her class is 70 and the standard 𝜎
(66.41−64) 2.41
deviation is 10. Of her many hundreds of students, it turns out that 20 of 𝑧= = = .90601503759
2.66 2.66
them also take psychology classes. Out of curiosity, I find myself When 𝑧 = .90601503759, the % between mean and z would be
wondering: do the psychology students tend to get the same grades as . 8159 − 0.5 = .3159
everyone else (i.e., mean 70) or do they tend to score higher or lower? To identify what percent of the given is shorter than ariel,
o 0.50+0.3159=0.8159 or 81.59% of the class is shorter
What is z-test? than Ariel.
The z test is a statistical test for the mean of a population. 2. What percentage of 15 year old girls are taller than Ariel?
o 0.50-.3159=.1841 or 18.41%
When do we use z-test?
Formula for z-test
when the population is normally distributed and σ(population SD) is 𝑥̅ −𝜇
known. ▪ 𝑧−= 𝜎
√𝑛

Review! Steps:
Let us try to recall how one score (or sample) compares with all other
1. State the hypothesis Compute the mean of each group.
scores (or a population).
2. Compute test value.
3. Solve for critical value
4. Make a decision
Examples: researcher selects a random sample of 35 stroke victims at the
hospital and finds the average cost of their rehab is $25,250. The
1. It has been reported that the average credit card debt for college
standard deviation of the population is $3251. At α = 0.01, can it
seniors is $3262. The student senate at a large university feels
be concluded that the average cost of stroke rehabilitation at a
that their seniors have a debt much less than this, so it conducts a
particular hospital is different from $24,672?
study of 50 randomly selected seniors and finds that the average
debt is $2995, and the population standard deviation is $1100.
Let’s conduct the test based on a Type I error of =0.05
o State the hypothesis
o H0: =$3262 H1: <$3262
o Compute the test value
o 𝑧=
2995−3262
1100 = −1.716341
√50

o Compute Critical value

o Left-tailed test, =0.05  Z will be negative and
have probability 0.05 underneath it.
o -1.65
o Make a decision
o Since our test value (-1.716341) is less than our
critical value (-1.645), we reject the null
hypothesis.

Activity:

2. The medical Rehabilitation Education Foundation reports that the

average cost of rehabilitation for stroke victims is $24,672. To see
if the average cost of rehab is different at a particular hospital, a
 probability (alpha level, level of significance, p) as a criterion for
3.3 T-test acceptance. In most cases, a 5% value can be assumed.
Introduction: 4. The fourth assumption is a reasonably large sample size is used.
A larger sample size means the distribution of results should
Consider a telecom company that has 2 service centers in the city. The approach a normal bell-shaped curve.
company wants to find whether the average time required to service a 5. The final assumption is homogeneity of variance. Homogeneous,
costumer is the same in both stores. or equal, variance exists when the standard deviations of samples
• Suppose that the company measures the average time taken by are approximately equal.
50 random costumers in each store. Store A has an average of 22 TYPES OF T-TEST
mins while store B has an average of 25 mins. Does this mean
that store A is more efficient than store B? 1. One Sample t-test : The One Sample t Test determines whether
the sample mean is statistically different from a known or
WHAT IS T-TEST? hypothesized population mean.
• a type of inferential statistic used to determine if there is a 2. Independent two sample t-test : The two-sample t-test (also
significant difference between the means of two groups, which known as the independent samples t-test) is a method used to
may be related in certain features. test whether the unknown population means of two groups are
equal or not.
T-TEST ASSUMPTIONS 3. Paired sample t-test: used to determine whether the mean
1. The first assumption made regarding t-tests concerns the scale of difference between two sets of observations is zero.
measurement. The assumption for a t-test is that the scale of Steps:
measurement applied to the data collected follows a continuous
or ordinal scale, such as the scores for an IQ test. 3. State the hypothesis Compute the mean of each group.
2. The second assumption made is that of a simple random sample, 4. Compute the variance of each group
that the data is collected from a representative, randomly 5. Compute standard area of difference between means
selected portion of the total population. 6. Compute t- statistic.
3. The third assumption is the data, when plotted, results in a 7. Solve for critical value
normal distribution, bell-shaped distribution curve. When a
normal distribution is assumed, one can specify a level of
Examples: 44.9 − 40
𝑡 − 𝑠𝑡𝑎𝑡 =
8.90
1. Imagine a company wants to test the claim that their √15
batteries last more than 40 hours. Using a simple 𝑡 − 𝑠𝑡𝑎𝑡 ≈ 2.13
random sample of 15 batteries yielded a mean of 44.9 Step 5: Solve for critical value
hours, with a standard deviation of 8.9 hours. Test this Degree of freedom: n-1
claim using a significance level of 0.05. Degree of freedom: 15-1=14
Critical value:1.761
Step 1: write the null and alternate hypothesis Note:

If
𝐻𝑜 : 𝜇 = 40
o Critical value > t-stat, null hypothesis is accepted.
𝐻𝑎 : 𝜇 > 40 o T-stat > critical value, reject null hypothesis.

In this case, since t-stat> critical value, we will reject the null
Step 2: Compute the variance of the group. hypothesis.

𝑠 2 = 8.92 2. Suppose a sample of 16 light trucks is randomly selected off the

assembly line. The trucks are driven 1000 miles and the fuel
𝑠 2 = 79.21
mileage (MPG) of each truck is recorded. It is found that the
Step 3: Compute standard area of difference between means mean MPG is 22 with a SD equal to 3. The previous model of the
light truck got 20 MPG.
𝑥̅ − 𝜇
Questions:
44.9-40 o State the null hypothesis for the problem above
Step 4: Compute t- statistic. ▪ Null hypothesis: Thtere is no significant
𝑥̅ − 𝜇 difference between the average MPG of the 16
𝑡 − 𝑠𝑡𝑎𝑡 = 𝑠 light trucks to the previous model.
√𝑛 ▪ Alternate hypothesis: The average MPG of the 16
light trucks is the same as the previous model.
 o Conduct a test of the null hypothesis at p= .05. BE SURE 2. Twelve subjects with diagnosed hypertension were randomly
TO PROPERLY STATE YOUR STATISTICAL CONCLUSION. selected for this study. The age at which they were diagnosed
𝑥̅ −𝜇 were recorded and listed below. Based on the data, is there
▪ 𝑡 − 𝑠𝑡𝑎𝑡 = 𝑠
√𝑛 any evidence that the age at diagnosis is not equal to 45.0
22−20
▪ 𝑡 − 𝑠𝑡𝑎𝑡 = 3 years? Age at Diagnosis of Hypertension32.8,40.0,41.0,42.0,
√16
45.5, 47.0, 48.5, 50.0, 51.0, 52.0, 54.0, 59.2
▪ 𝑡 − 𝑠𝑡𝑎𝑡 ≈ 2.67
➢ For critical value: Quiz:
▪ Degree of freedom: n-1
1. What is t-test?
▪ Df=16-1=15
2. When do we use t-test?
▪ P=.05
3. What type of statistics uses t-test?
▪ Critical value= 2.131
4. Explain why having a mean grade of 36 does not necessarily
mean that the class performed better than a class that has a
o Provide an interpretation of your statistical conclusion
mean grade of 33.
using the variables from the description given
5. The carbon monoxide (CO) level in a manufacturing plant is
▪ Since t-stat> critical value, therefor, we would
supposed to be about 50 parts per million (ppm). However
reject the null hypothesis.
the actual CO levels are quite variable. Five CO
▪ Interpretation: The average MPG of the 16 light
measurements are taken at various times during the day:58,
trucks is significantly different from the previous
63, 48, 52, 68.
model.
o Test the null hypotheses that μ= 50 ppm for the CO
Activity concentration at the manufacturing plant.
1. A random sample of 22 fifth grade pupils have a grade point
average of 5.0 in maths with a standard deviation of 0.452,
whereas marks range from 1 (worst) to 6 (excellent). The
grade point average (GPA) of all fifth grade pupils of the last
five years is 4.7. Is the GPA of the 22 pupils different from the
populations’ GPA?
 Examples:
3.5 Identifying test analysis
Identify the type of test to be used in the given samples:
What are the two types of test used in
inferential statistics? How are they • Humerous bones from the same species ofanimal have approximately
different from each other? the same length-to-width ratios. It is
Z-test T-test
It is a test used to compare T-test is a test used to test known thatSpecies A has a mean ratio of 8.5. Suppose that 41 fossil
population mean to sample hypothesis. they are most useful humerous boneswere unearthed at a site
when where Species A is known to have flourished. (Weassume that all bones
we need to determine if there is a
are from the same species.) The
statistically significant difference
between two independent sample length-to-width ratiosof these bones has sample mean 9.26 and sample
groups. standard deviation 1.20.Can we
require data with a normal distribution, which
means that the sample (or population) data is distributed evenly conclude that these bones belong to Species A?
around
• Average heart rate for Americans is 72 beats/minute. A group of 25
the mean.
Population Standard deviation is Population Standard deviation is individuals participated in an aerobics
known unknown fitness program to lower their heart rate. After six months the group was
evaluated to identify is the program
N>30 N<30
had significantly slowed their heart. The mean heart rate for the group
was 69 beats/minute with a standard

deviation of 6.5. Was the aerobics program effective in lowering heart

rate?

• The mean Verbal SAT score for the population of first students at
Radford is 520. The standard deviation of
scores in this population is 95. An investigator believes that the mean
Verbal SAT of first year psychology

majors is significantly different from the mean score of the population.

The mean of a sample of 36 first year

psychology majors is 548. Please test the investigator's prediction using

an alpha level of .05.

• A colleague of the investigators is problem 3 repeats the experiment but

matches the samples on the

dimensions of sex and job type. the raw data appear below. Evaluate her
experiment using the criteria of p <

.05. Assume it is a two tailed test.