You are on page 1of 7

COR 006

Statistics and Probability Reviewer


4th Achievement Test, 2nd Semester

MODULE #16

HYPOTHESIS is simply a
statement that something is
true. It is a tentative
explanation, a claim, oran
assertion about people, objects
or events.
STATISTICAL HYPOTHESIS is
any assumption or assertion
made on the distribution of a
population,which is either true
or false.
HYPOTHESIS TESTING is a
statistical procedure in
determining which hypothesis
is more acceptableas true or
which hypothesis is more likely
to be false.
Types of Hypothesis Fact
1.NULL HYPOTHESIS (HO).
Decision Ho is true Ho is false
This expresses the idea of no
existence of relationship or Do not reject Ho Correct decision Type II error (also known as
difference between the variable beta error)
under study. This is usually
designated by a not or no term. Reject Ho Type I error (also known as Correct Decision
alpha error)
TYPE I ERROR. Rejecting the can be expressed as 𝜇 G 𝜇0. A
null hypothesis when in fact two-tailed test uses a
the null hypothesis is true. hypothesis test for which the
rejection region lies on both end
TYPE II ERROR. Failure in
tails of distribution
rejecting the null hypothesis
when in fact the null
hypothesis is false.
MODULE #17
A ONE TAILED TEST is a test
hypothesis test for which the
rejection region lies at only one 1.Formulate the null and
tail of the distribution. One alternative hypothesis.
tailed test is classified as a left-
tailed test or right-tailed test. If
the population mean (𝜇) is less Null Hypothesis: It shows no
than the specified value of 𝜇𝑜, significant difference between
then it is left-tailed test for two parameters. It is denoted
which the alternative by𝐻𝑜.
hypothesis can be expressed as Alternative Hypothesis:
𝜇 < 𝜇0. It is a right-tailed test if contrary to the null hypothesis,
the population mean (𝜇) is which shows that observations
greater than the specified value are the result of a real effect. It
of 𝜇𝑜 for which the alternative is denoted by 𝐻𝑎
hypothesis can be expressed as
𝜇 > 𝜇0.
Two-Tailed Test vs One-Tailed
A TWO-TAILED TEST is used
Test
when the alternative
hypothesis is non-directional When the alternative
which means that the values of hypothesis is two-sided like
who measures of the same kind 𝐻𝑎:𝜇 ≠ 𝜇𝑝, it is called two-tailed
are not equal. A two-tailed tests test.
has a not equal sign (G) in the
When the given statistics
alternative hypothesis. When
hypothesis assumes a less
the population mean (𝜇) is not
than or greater than value, it is
equal to specified value of 𝜇𝑜,
called one-tailed test.
then the alternative hypothesis
standard proportion, or
2.Identify the level of comparing the proportions of
significance (𝛼) and the two populations
appropriate test statistic. Example: Comparing the
average teaching salaries of
men versus women.
Level of Significance
A t-test is used for testing the
The level of significance mean of one population against
denoted by alpha or 𝛼 refers to a standard or comparing the
the degree of significance in means of two populations if you
which we accept or reject the do not know the populations’
null hypothesis. standard deviation and when
100% accuracy is not possible you have a limited sample (𝑛 <
in accepting or rejecting a 30). If you know the
hypothesis. populations’ standard
deviation, you may use a z-test
The significance level α is also
the probability of making the Example: Measuring the
wrong decision when the null average salaries of professors
hypothesis is true.  The most from a certain university when
common levels of significance you have a small sample.
used are 0.01. 0.05 and 0.10. An F-test is used to compare 2
or more populations’ variances.
The samples can be any size. It
3. Types of Test-Statistic
is the basis of ANOVA. It is also
A z-test is used for testing the used in comparing the means of
mean of a population versus a two or more independent
standard, or comparing the groups and in looking at the
means of two populations, with interaction effect between the
large (𝑛 ≥ 30) samples whether variables being analyzed
you know the population
Example: Comparing the
standard deviation or not. It is
variability of score from three
also used for testing the
sections of students
proportion of some
characteristic versus a
4.Determine the critical 6. sate the conclusion
value/tabular value for the test.
Present your
Obtained the value from tables
findings/conclusions and site
using the value of for one-
some sources to support your
tailed test and /2 for two
findings/conclusions.
tailed test.

Illustration of the Rejection MODULE #18


Region TEST STATISTIC is a numerical
 The rejection region (or measure that is used to
critical region) is the set of all determine whether to reject or
values of the test statistic that fail to reject𝐻𝑜.
causes us to reject the null When the population standard
hypothesis. deviation (𝜎) is known and
 The non-rejection region sample size n≥ 30, the test
(or acceptance region) is the set statistics to be used in
of all values of the test statistic performing test of hypothesis
that causes us to fail to reject for the mean is given by
the null hypothesis.
 The critical value is a
point (boundary) on the test
Where X is the sample mean,
distribution that is compared to
µ is the population mean,
the test statistic to determine if
the null hypothesis would be 𝜎 is the population standard
rejected. deviation
5. Decide whether to accept n is the sample size
or reject the null hypothesis.
Example:
The null hypothesis is accepted
A grade 11 researcher reported
when the absolute value of the
that the average allowance of
computed value of z or t is less
Senior High School students
than the absolute value of the
was P100. A sample of 40
critical value, otherwise reject
students has a mean allowance
the null hypothesis.
of P120. At 𝛼 = 0.01 test, it was
the claimed that the students speed different from 550 words
had allowance of more than per minute?
P100. The standard deviation
Here,the sample size (n) is 12
of the population P50.
which is less than 30 and
Here, the sample size (n) is 40 sample standard deviation (5
and the population standard words per minute) was given.
deviation (𝜎 = 50) is known, Therefore, the appropriate test
then the appropriate test is t-test.
statistics to be used is z-test.
Central Limit Theore is to be
When the sample standard used if the population variance
deviation (s) is known, is known or unknown and the
population standard deviation sample size is n≥ 30 or
(𝜎),is unknown and sample size considered sufficiently large.
n < 30, the test statistics to be The test statistic to be used in
used in performing test of performing test of hypothesis
hypothesis for the mean is for the mean is
given by
Where 𝑥 is the sample mean,
Where 𝑥 is the sample mean,
µ is the population mean, µ is the population mean,
𝑠 is the sample standard 𝑠 is the sample standard
deviation deviation
n is the sample size n is the sample size
Example: Example:
An English teacher wanted to A study was conducted to look
test whether the mean reading at the average time student
speed of students is 550 words exercise. A researcher claimed
per minute. A sample of 12 that in average, students
students revealed a sample exercise less than 15 hours per
mean of 540 words per minute month. In a random sample
with a standard deviation of 5 size n=115, it was found that
words per minute. At 0.05 the mean time students
significance level, is the reading exercise is 11.3 hours per
month with s = 6.43 hours per
month.
Since n=115, the sample size
is large and variance is
unknown. Hence, z-test is the
appropriate tool. (Central Limit
Theorem)

MODULE #19

Critical Value of Z

Test Type 0.10

One-tailed 1.28

Two-tailed 1.645
Example:
There is no significant
relationship between the
attitude of students towards
their subjectsand their
performance rating at the end
of the semester.
2.ALTERNATIVE HYPOTHESIS
(HA). This is the opposite of the
null hypothesis. It states
theexistence of relationship or
difference.
Example:
There is a significant
relationship between the
attitude of students towards
their subjects and their
performance rating at the end
of the semester.

You might also like