10/9/2017

THE NULL AND ALTERNATIVE

TYPES OF STATISTICS TYPES OF STATISTICS
HYPOTHESES
Descriptive Inferential • Descriptive Statistics: Involves the
1. The hypothesis subjected to testing to
Statistics: deals Statistics: involves determine whether its truth can be accepted or
with the methods of organizing, rejected is the null hypothesis denoted by Ho.
collection, the analysis and
summarizing, and presenting data This hypothesis states that there is no significant
organization and interpretation of
presentation of data.
in an informative way. relationship or no significant difference between
data. two or more variables, or that one variable does
• Inferential Statistics: involves the not affect another variable.
analysis and interpretation of data Example:
“There is no significant difference between the
effectiveness of method A and method B.”

THE NULL AND ALTERNATIVE

HYPOTHESIS
HYPOTHESES

• is a conjecture or statement which aims to • Alternative hypothesis denoted by Ha. This is the
explain certain phenomena in the real world. hypothesis that challenges the null hypothesis.
• products of man’s curiosity.
INFERENTIAL STATISTICS Example:
T E STI NG T H E S I GNI FI CANCE OF T H E
“Method A is more effective than method B,”
DI F F ERENCE B E T WEEN T W O M EANS, T W O or
S T A N DA R D D E V I A T IO N S, T W O P R O P O RT I ON S ,
O R T W O P E R C E NT A G ES “Method A is less effective than method B,”

1
 10/9/2017

SIGNIFICANCE LEVEL ONE-TAILED AND TWO-TAILED TESTS TESTING HYPOTHESIS

• To test the null hypothesis of no significance in • Null Hypothesis, Ho: = 6. Find the tabular value using tables of
the difference between two or more variables, • Alternative hypothesis Ha: ≠ . different tests.
one must set the level of significance first. 7. Compare the computed value, CV, to the
• A two-tailed or two-directional test will be used
• Type I error is denoted by the symbol . It is the if there are two possibilities either the left side is tabular value, TV.
probability of accepting the alternative greater than the right, or the left side is less than
hypothesis, Ha, when, in fact, the null the right. • Decision Rule: If CV is less than the TV,
hypothesis, Ho, is true.
• If Ha : < , then a one-tailed or one-directional accept the null hypothesis. If CV is greater
• Type II error is the probability of accepting the test will be used. than the TV, reject the null hypothesis. Make
null hypothesis when, in fact, it is false and is a conclusion using the result of the
denoted by the symbol β. comparison.

ONE-TAILED AND TWO-TAILED TESTS TESTING HYPOTHESIS DEGREE OF FREEDOM, (df)

• A test is called a one-tailed test if the rejection region lies

• The degree of freedom gives the number of
Steps in testing the truth of a hypothesis:
on one extreme side of the distribution pieces of independent information available for
• A two-tailed test - the rejection region is located on
1. Formulate the null hypothesis. Denote it is computing variability.
both ends of the distribution. as Ho and the alternative hypothesis as Ha.
• The number of degrees of freedom required will
2. Set the desired level of significance (). vary depending on the size of the distribution.
3. Determine the state of statistics to be used • For a single group of population, df = N – 1,
in testing the null hypothesis. where N is the population.
4. Compute for the value of the statistics to • For two groups, the formula for df is:
be used.
df = N1 + N2 – 2 for t-test
(a) One-tailed test (b) two-tailed test 5. Compute for the degrees of freedom.
df = N – 2 for Pearson r.

2
 10/9/2017

The tabular values of z can be obtained from

EXAMPLE 1
the following table:

• Table 1. Critical Values of z 1. Ho : The mean lifespan of the battery-operated toy

cars is 5 years.
Level of Significance (µ = 5 years)
Test Type 0.10 0.05 0.025 0.01 Ha : The mean lifespan of the battery-operated toy
TEST CONCERNING MEANS cars is not 5 years.
Z - TEST ON THE COM PARISON BETW EEN One-test type ±1.28 ±1.645 ±1.96 ±2.33 (µ ≠ 5 years)
THE POPULATION M EAN AND THE
SAM PLE M EAN Two-tailed type ±1.645 ±1.96 ±2.33 ±2.58 2.  = 0.05; two-tailed
3. Use z-test as test statistic.
• Decision Rule: Reject Ho if ≥ .

Z-TEST EXAMPLE 1 EXAMPLE 1

• If the population mean (µ) and the population • A company, which makes battery-operated toy 4. Computation:
variance () are known, and µ will be compared cars, claims that its products have a mean life =3 =5
to a sample mean , then, span of 5 years with a standard deviation of 2 = 40 =2
years. Test the null hypothesis that µ = 5 years ( − ) (3 − 5)
against the alternative hypothesis that µ ≠ 5 = = 40 = −6.32
− 2
= years if a random sample of 40 toy cars was 5. Critical regions: z < -1.96 and z > 1.96 (Table 1)
tested and found to have a mean life of only 3 6. Decision: Reject the Ho and accept the proposition
years. Use a 0.05 level of significance. that the mean life span of the toys is not equal to
5 years since , which is 6.32, is greater than
, which is 1.96
7. The difference is significant.

3
 10/9/2017

EXAMPLE 2 EXAMPLE

• A manufacturer of bicycle tires has developed a • The average length of time for people to vote
new design which he claims has an average life using the old procedure during a presidential
span of 5 years with standard deviation of 1.2 election period in precinct A is 55 minutes.
years. A dealer of the product claims that the Using a computerization as a new election
average life span of 150 samples of the tire only method, a random sample of 20 registrants was
3.5 years. Test the difference of the population T-TEST used and found to have a mean length of voting
and sample means at 5% level of significance. time of 30 minutes with a standard deviation of
1.5 minutes. Test the significance of the
difference between the population mean and the
sample mean.

t-test on the comparison between the

population mean and the sample mean
1. Ho: = Ha: <
• The t-test can be used to compare the means when the 1. Ho: ̅ = μ Ha: ̅ < μ
2.  = 5%; one-tailed population mean (µ) is known but population variance 2. α =5%; one-tailed
3. Test statistics: z-test () is unknown.
• When the population standard deviation is unknown
3. t−test is the appropriate test statistic.
( ) .
4. Computation: = = 12.25 = −15.31 but the sample standard deviation can be computed, the 4. Computation:
.
5. Critical areas: 5% level of significance, z < -1.645 and z > t-test can also be used instead of z-test. The formula is
( )
1.645 (Table 1) given below: = = 20 = −74.54
.
( − )
6. Since > , reject the null hypothesis and = 5. df = n -1 = 20 – 1 = 19
accept the alternative hypothesis. The mean life span of the 6. Tabular value of t = 1.729 (one-tailed).
samples is less than the mean life span of the population. • Where
7. It implies that the difference between two means is • Decision Rule: Reject Ho if ≥ . 7. Decision: Reject the Ho since , which is 74.54,
significant at 5%. greater than , which is 1.729.
8. There is significant difference between the means.

4
 10/9/2017

EXAMPLE

• A course in physics was taught to 10 students using 4. Computation: = 82 = 78 = 10

traditional method. Another group of 11 students = 11 =5 =6
went through the same course using another
method. At the end of the semester, the same test 82 − 78
=
was administered to each group. The 10 students
10 − 1 5 + 11 − 1 6 10 + 11
under method A got an average of 82 with a 10 + 11 − 2 (10)(11)
standard deviation of 5, while the 11 students under
t = 1.65
method B got an average of 78 with a standard
deviation of 6. Test the null hypothesis of no
significant difference in the performance of the two
groups of students at 5% level of significance.

t-test Concerning Means of Independent

Samples
• When two samples are drown from normally distributed 1. Ho: there is no significant difference between the 5. df = + − 2 = 10 + 11 – 2 = 19
populations with the assumption that their variances average scores of the two group students. = 6. Tabular value = 1.729
are equal, the t-test with the given formula should be Ha: the mean score of first group is higher than the 7. The null hypothesis will be accepted because
used. mean score of the second group. > < .
−
= 2. α = 5%; one-tailed 8. This means that the difference between the
−1 + −1 + 3. Use the t-test as test statistic. means is not significant at 5% level of significance.
+ −2 − This also implies that method A is as effective as
= method B.
where , =
−1 + −1 +
, = + −2
, = variances
• Decision Rule: Reject Ho if ≥ .

5
 10/9/2017

t-test on the significance of the difference

between two correlated means Pre-test Post-test Difference
Student d2
Score Score d
• When comparing two correlated means, the t-test is A 34 45 -11 121 −77
the appropriate test statistic. A typical example is B 23 32 -9 81 = = 10 = −3
when comparing the results of the pre-test and the C 40 46 -6 36 8.11
post-test administered to a group of individuals. The 10
two tests must be the same. D 31 57 -26 676
5. df = n – 1 = 10 – 1 = 9
E 24 39 -15 225
6. Tabular value = 2.821 (one-tailed)
= F 45 48 -3 9
G 27 27 0 0 7. Reject Ho since > .
H 32 33 -1 1 8. This means that the performance of the students in
where: d = difference between the pre-test and post- College Algebra significantly improved.
test scores I 12 18 -6 36
n = number of observations J 45 45 0 0
sd = standard deviation of the differences Σ d = -77 Σ d2 = 1,185

z- test on the significance of the difference

EXAMPLE
between two independent proportions
• To determine whether the students’ performance in 1. Ho: The students’ performance in Algebra did not improve. • To determine if there is a significant difference between
College Algebra improved after enrolling in the (µ1 = µ2) proportions two variables, the z- test will be used.
−
subject for one term, a 60-item pre-test and post- Ha: The students’ performance in Algebra improved. (µ1 < µ2) =
test were administered to them on the first and the +
2. α = 1%; one-tailed test.
last days classes, respectively. The same test was
given as pre-test and post-test. 3. t - test will be used. where =
4. Computations =
• The results are as follows:
∑ ∑ ) , )
=1−
Sample variance: = = = 65.79 = 1−
= 8.11 = ℎ
= ℎ

6
 10/9/2017

EXAMPLE

• A sample survey of a presidential candidate 4. Computation:

120 175
in the Philippines shows that 120 of 200 male − −
200 250
voters dislike candidate X and 175 of 250 = =
female voters dislike the same candidate. + 120 80 175 75 SIGNIFICANCE OF THE
200 200 250 250
Determine whether the difference between 200
+
250 DIFFERENCE BETWEEN
the two sample portions, and , is =
−0.1
=
−0.1
= −2.22 VARIANCES
0.24 0.21 0.045
significant or not at 10 % level of +
200 250 AN AL Y S I S O F V AR I AN C E S
significance.

ANOVA

1. Ho: There is no significant difference between the 5. The tabular value of z is 2.58 (two-tailed) (Use • When variances of two or more independent
proportion of the male votes and the proportion Table 1). samples differ, the appropriate test statistic to
of female votes. ( = ) 6. Since = 2.22, is less than = 2.58, determine the significance of such difference
Ha: There is no significant difference between the accept Ho. is the analysis of variance (ANOVA), which
proportion of the male votes and the proportion of 7. Therefore, there is no significant difference
female votes. ( ≠ ) makes use of the F ratio or variance ratio. The
between the proportions of male and female various groups being compared are assumed
2. Set α = 0.01; two-tailed voters in their dislikes for candidate X.
3. Use z-test as test statistic.
to belong to a population with a normal
distribution, each group randomly selected
and independent from the other groups. The
variables from each group also have standard
deviations that are approximately equal.

7
 10/9/2017

Steps in solving analysis of variance

1. State the null hypothesis. 4. Find the tabular value of F at the given level of 1. Ho: There is no significant difference in the mean sales of the three
2. Set the level of significance. significance (from Appendix E) candidates for promotion.
3. Accomplish the ANOVA Table. 5. Accept the null hypothesis if the computed value 2. Set α = 5%.
F is less than the tabular value and reject if it is 3. Sum of Squares
Source greater than the tabular value.
Sum of = (∑ +∑ +∑ )−
(∑ ∑ ∑ )
of df Mean Square F 6. Interpret the result.
Square
Variation ( , , , )
= 1,951,220,000 + 1,895,218,544 + 1,858,674,201 −
Between SSB = −1 = ( , )
TSS = 5,705,112,745 − = 5,705,112,745 − 5,569,583,606
=
Within SSW = − = TSS = 135, 529, 139
Total TSS = −1

EXAMPLE: Determine who among the three salesmen

will most likely be promoted based on their monthly sales in
pesos? Use 5% level of significance.

• Where: Sales of Three Candidate for Promotion (A, B, C) (∑ ) + (∑ ) + (∑ ) (∑ +∑ +∑ )

(∑ ) (∑ ) A B C A2 B2 C2
= −
= −
12,000 15,500 12,899 144,000,000 240,250,000 166,384,201 ( , , , ) ( , , , )
10,000 12,500 16,000 100,000,000 156,250,000 256,000,000 = −
(∑ )
= − 10,900 12,000 15,000 118,810,000 144,000,000 225,000,000 ( , )
18,000 13,000 12,700 324,000,000 169,000,000 161,290,000
= 5,569,693,572 -
SSW = TSS – SSB 16,000 14,000 15,000 256,000,000 196,000,000 225,000,000 = 5,569,693,572 – 5,569,583,606
N = sample size k = number of columns 14,400 15,888 13,000 207,360,000 252,428,544 169,000,000 = 109,966
X = observed value n = number of rows 14,400 12,300 12,000 207,360,000 151,290,000 144,000,000
= − = 135,529,139 − 109,966 = 135,419,173
A = given factor or category 15,500 15,000 16,000 240,250,000 225,000,000 256,000,000
18,800 19,000 16,000 353,440,000 361,000,000 256,000,000
i = individual observation cell
∑ = 130,000 ∑ = 129,188 ∑ = 128,599
∑ = 1,951,220,000 ∑ = 1,895,218,544 ∑ = 1,858,674,201

8
 10/9/2017

degrees of freedom
= − 1 = 27 − 1 = 26
= −1= 3−1=2
= − = 27 − 3 = 24
Mean squares
,
= = = 54,983
, ,
= = = 5,642,465.542
,
= = = 0.00974
, , .

The ANOVA Table

Source of
Sum of Square df Mean Square F
Variation
Between 109,966 2 54,983
0.00974
Within 135,419,173 24 5,642,465.542
Total 135,529,139 26
4. Tabular value of F = 3.40
5. The computed value, 0.00974, is less than the tabular
value, 3.40, so the null hypothesis is accepted.
6. Therefore, there is no significant difference between
the sales of Salesmen A, B, and C. Hence, the three
salesmen have almost equal chances promotion.