Sei sulla pagina 1di 32

TESTING

Definition 1 (Statistical Hypothesis)


A statistical hypothesis is an
assumption or statement concerning a
characteristic of a population.
Remarks 2
If θ represents a characteristic or
parameter, the statement “θ has some
specified value θ0” is a statistical
hypothesis.
Examples 3:
•There is no significant relationship
between the mathematics attitude and
competency levels of grade 11 ABM
students of Iligan City National High
School.
•The percentage of shoppers who buy
their favorite products regardless of
price is not 25%.
•The average monthly allowance of all
students of Iligan City National High
School is at least P1, 000.
Definition 4 (Hypothesis Testing)
Hypothesis testing is a procedure in
making decisions based on a sample
evidence or probability theory used to
determine whether the hypothesis is acc
epted or rejected.
There are Two Types of Statistical Hypotheses

NULL HYPOTHESIS.
The null hypothesis, denoted by H0, specifies
the statement to be tested. It is assumed to
be true throughout the statistical analysis.

Example 5:If one hypothesizes that the


average Filipino weight is 100 lbs., or
symbolically μ = 100 lbs. Then H0 :µ = 100.
ALTERNATIVE HYPOTHESIS.
The alternative hypothesis, denoted by HA or
H1 specifies the parameter value or values
which are to be accepted if H0 is rejected.

Example 6: A possible alternative hypothesis


of example 6.5 is:
 the average weight is not equal to 100 lbs
, i.eHA :μ ≠ 100 lbs
 the average weight is greater than 100 lbs,
i.e., HA :μ > 100 or
 the average weight is less than 100 lbs, i.e.
, HA :μ < 100.
Types of Errors
Using sample information, two different
decisions can be reached; namely
Reject H0 – if there is sufficient evidence in
the sample that indicates H0 is more likely to
be false
Accept H0 – if there is insufficient evidence in
the sample to indicate that H0 is false, thus,
H0 is accepted as true.
Type I and Type II Errors

Type I Error - Rejecting the null


hypothesis when in fact the null
hypothesis is true.

Type II Error - Not rejecting the null


hypothesis when in fact the null
hypothesis is false.
This procedure, which helps us to decide whe
ther to reject or not the null hypothesis, is ca
lled test of hypothesis. In doing so, we may
commit either of two types of errors which ar
e shown in the table below:
Remark 7
These errors are measured in terms of
probability and the Greek letters α and β
are used to denote them, i.e.
•The level of significance
α α of the test is
defined to be the probability of
•committing a type I error.
•The probability of committing a Type II
error is β, called the power of the test.
CRITICAL REGION

Definition 8 (Critical Region)


A critical region or rejection region
is a part of the set of all possible values
of a test statistics for which Ho is rejected
.
Remarks 9.1:
The test statistic is used as a basis for
deciding whether the null hypothesis
should be rejected, e.g. mean.
The set of all possible values of the test
statistic, which could sometimes range
from negative infinity to positive
infinity, is divided into two regions:
critical region and acceptance region.
The values of the test statistic that separate
the rejection and acceptance regions are
called critical values.
Illustration:
Two Types of Statistical Test

1. One – Tailed Test


One-tailed test or one-sided test is a test
where critical region is in one direction
only. It is used when the alternative
hypothesis is directional, e.g. HA :μ> μ0 or
HA :μ < μ0.
Example 10: Suppose you want to answer any
of the following questions:
• Is the average Filipino weight greater than
100 lbs.? In this case the null and alternative
hypotheses are H0: μ = 100 lbs and
HA: μ >100 lbs.
• Is the average Filipino weight less than 100
lbs.? In this case the null and alternative
hypotheses are H0: μ = 100 lbs and
HA: μ <100 lbs.
2. Two – Tailed Test
A two-tailed test is used when the
alternative hypothesis is non-directional
which means that the values of two measures
of the same kind are not equal, i.e,
HA :μ ≠ μ0.
Example 11: Suppose you want to answer
the question: Do the sample data
provide sufficient evidence to indicate
that the average Filipino weight is
100 lbs.?. A two-sided test is employed
where H0: μ = 100 lbs and
HA: μ ≠ 100 lbs.
Example 10: Suppose you want to answer any
of the following questions:
• Is the average Filipino weight greater than
100 lbs.? In this case the null and alternative
hypotheses are H0: μ = 100 lbs and
HA: μ >100 lbs.
• Is the average Filipino weight less than 100
lbs.? In this case the null and alternative
hypotheses are H0: μ = 100 lbs and
HA: μ <100 lbs.
Steps in a Stati
stical Test of H
ypothesis:
Formulate the null a
he
appropriate alterna
Specify the leve
significance and
test to be used.
Select the appr
opriate test sta
Formulate the
decision rule (i.
e., indicating o
r establishing t
he
Compute the
value of the
statistical t
est.
Decide whether to ac
cept or reject the null
hypothesis that is by ma
king a statistical decision
, which is to reject H0 wh
en the computed value
of the test statistic is wit
Draw the
appropriate
conclusion.
TEST FOR POPULATION MEAN

The Z – Test
The z-test is a parametric test (a test
that requires of normal distribution). It
utilizes the two population parameters µ
and σ.
Remarks 9.3:
 The z-test is used to compare two
means, the sample mean, and the
perceived population mean.
 It is also used to compare the two sam
ple means taken from the same
population and
 When samples are equal or greater
than 30.
 The z-test can be applied in two ways:
The One-Sample Mean Test and Two-Sam
ple Mean Test.
T – Test
The t-test is also a parametric test.
The t-distribution was developed by an
employee of Irish Brewery in the person
of William S. Gosset (1876 – 1936). He
chose to publish his findings using pen
name “Student”.
Remarks 9.4:
o The T-test is used when the sample
size involves small cases (n<30) and the
population standard deviation is
unknown.
o The sample standard deviation is used
as an estimator of the population
standard deviation.
o The t-test is used to compare two mea
ns, the means of independent samples
and the means of correlated samples
before and after the treatment.

Potrebbero piacerti anche