HYPOTHESIS TESTING

(T-test)
T-Test also called as the “Student’s T Statistics”
Function: The student’s T-test or student’s T statistics may be used as a model for testing a hypothesis
about the mean of a normally distributed population. Typically, the procedure is used to determine whether
the population of interest differs significantly from the norm group.

A). One sample T-test

One Sample T-test is a statistical procedure that is used to know the mean difference the sample and
the known value of the population mean. We draw a random sample from the population and then compare
the sample mean with the population mean and make a statistical decision as to whether or not the sample
mean is different from the population. The sample size should be less than 30.
Where: ̅ − 𝝁)√𝒏
(𝑿
t = one sample t test 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 =
𝒔
̅ = sample mean
𝐗
µ = population mean ̅ )𝟐
∑(𝑿 − 𝑿
s = sample standard deviation 𝒔= √
𝒏−𝟏
n = number of observation in the sample
Degrees of Freedom,
‘df = n-1
Assumptions in One Sample T-test:
1. Samples drawn from the population should be random.
2. Population must be approximately normally distributed.
3. The population mean should be known.
4. Cases of the samples should be independent.
5. Sample size should be less than 30.

PARTIAL TABULAR VALUE FOR T-TEST at 1% and 5% Level of Significance

Level of Level of
Significance 0.05 0.01 Significance 0.05 0.01
(5%) (1%) (5%) (1%)
(a) (a)
Degrees of Degrees of
Freedom, df Freedom, df
1 6.314 31.821 16 1.746 2.583
2 2.920 6.965 17 1.740 2.567
3 2.353 4.541 18 1.734 2.552
4 2.015 3.747 19 1.729 2.539
5 1.943 3.365 20 1.725 2.528
6 1.895 3.143 21 1.721 2.518
7 1.860 2.998 22 1.714 2.508
8 1.833 2.896 23 1.711 2.500
9 1.812 2.821 24 1.708 2.492
10 1.796 2.764 25 1.706 2.485
11 1.782 2.718 26 1.703 2.479
12 1.771 2.681 27 1.699 2.473
13 1.761 2.650 28 1.697 2.467
14 1.753 2.624 29 1.684 2.462
15 1.748 2.602 30 1.671 2.457

Hypothesis Testing Page 1

 Example #1.and #2. (One sample t test).
ABC Company, a manufacturer of automobile tires claims that the average life of its product is
45,600 miles. A random sample of 15 tires was chosen and resulted to a mean life of 43,500 miles with a
standard deviation of 3,000 miles. Based on the above results, test the hypothesis that ABC Company
tires average life is 45,600 miles against the result of less than 45,600 miles. Use level of significance of
1% and 5%.

Given: Sample mean life (𝑿 ̅ ) = 43,500miles Population mean life (µ ) = 45,600miles

Sample size ( n ) = 15 tires Sample standard deviation ( s ) = 3,000miles
Level of Significance = 5% (0.05)

Solution: for #1
1. State the
a). Ho: “ABC Company tires mean life is 45,600miles”…………….. or . .(µ = 45600)
b). Ha: : “ABC Company tires mean life is less than 45,600miles”… or . .(µ < 45600)
2. Determine if the test is One-Tailed or Two-Tailed = One tailed. / (N-A)
3. Specify the Level of Significance (a) = (0.05) or (5%).
4. Determine the degree of freedom, df = for T- test the degree of freedom = n-1 = 15-1 = 14.
5. Determine the TABULAR VALUE or CRITICAL VALUE (from the tabular value of T-test) = 1.753
6. Determine the computed value (using the test statistics or Statistical Tool/Formula).
̅ − 𝝁)√𝒏
(𝑿 (𝟒𝟑𝟓𝟎𝟎 − 𝟒𝟓𝟔𝟎𝟎)√𝟏𝟓
𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = = = −𝟐. 𝟕𝟏
𝒔 𝟑𝟎𝟎𝟎
7. Make a decision. Compare the computed value to the critical/tabular value.
a. If the computed value is LESS than the critical value, ACCEPT the Null Hypothesis.
b. If the computed value is GREATER than or EQUAL to the critical value, REJECT the Null
Hypothesis.
Since the computed value is less than the critical value, therefore ACCEPT Ho!!!

8, Conclusion.

Ho: “ABC Company tires mean life is 45,600miles”…. or . .(µ = 45600)

Given: Sample mean life (𝑿 ̅ ) = 43,500miles Population mean life (µ ) = 45,600miles

Sample size ( n ) = 15 tires Sample standard deviation ( s ) = 3,000miles
Level of Significance = 1% (0.01)

Solution: for #2
1. State the
a). Ho: “ABC Company tires mean life is 45,600miles”…………….. or . .(µ = 45600)
b). Ha: : “ABC Company tires mean life is less than 45,600miles”… or . .(µ < 45600)
2. Determine if the test is One-Tailed or Two-Tailed = One tailed. / (N-A)
3. Specify the Level of Significance (a) = 0.01
4. Determine the degree of freedom, df = for T- test the degree of freedom = n-1 = 15-1 = 14
5. Determine the TABULAR VALUE or CRITICAL VALUE (from the tabular value of T-test) = 2.624
6. Determine the computed value (using the test statistics or Statistical Tool/Formula).
̅ − 𝝁)√𝒏
(𝑿 (𝟒𝟑𝟓𝟎𝟎−𝟒𝟓𝟔𝟎𝟎)√𝟏𝟓
𝒁𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = = = -2.71
𝒔 𝟑𝟎𝟎𝟎
7. Make a decision. Compare the computed value to the critical/tabular value.
a. If the computed value is LESS than the critical value, ACCEPT the Null Hypothesis.
b. If the computed value is GREATER than or EQUAL to the critical value, REJECT the Null
Hypothesis.
Since the computed value is less than the critical value, therefore ACCEPT Ho!!!
8, Conclusion.
Ho: “ABC Company tires mean life is 45,600miles”

Hypothesis Testing Page 2

B). Two-sample T-test
Two Sample T-test is a statistical procedure that is used to know the mean difference/significance of the
two small samples drawn from two different populations. The T-test statistics formula for two-sample means
will be used in computing the value of “tcomputed”. The sample size should be less than 30.
Where:
t = two sample t test The T-test Statistics for Two-Sample Means:
̅̅̅̅
𝐗 𝟏 = sample mean 1 (̅̅̅̅
𝑿𝟏 − ̅̅̅̅
𝑿𝟐 )
̅̅̅̅
𝐗𝟐 = sample mean 2 𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = =
n1 = sample size 1 𝒏𝟏 𝑽𝟏+𝒏𝟐 𝑽𝟐 𝒏𝟏+𝒏𝟐
ට ∗ 𝒏𝟏 𝒏
n2 = sample size 2 𝒏𝟏 +𝒏𝟐−𝟐 𝟐
V1 = sample variance 1
V2 = sample variance 2
Degrees of Freedom,
‘df = n1 + n2 - 2

Example #3. (Two-sample t test).

Two samples are randomly selected from two groups of students who have been taught using
different teaching methods. An examination is given and the results are as shown below:
Group #1: n1 =8, mean=85, variance = 46 Group #2: n2= 10, mean= 87, variance = 36
Using a 5% level of significance, can we conclude that the two different teaching methods are equally
effective?

Given: Sample mean 1 score (𝑿𝟏 ̅̅̅̅ )= 85 Sample variance 1 (V1 ) = 46

̅̅̅̅
Sample mean 2 score (𝑿𝟐 )= 87 Sample variance 2 (V2 ) = 36
Sample size 1 ( n1 ) =8 Sample size 2 ( n2 ) = 10
Level of Significance = 5% (0.05)

Solution:
1. State the
a). Ho: The two different teaching methods are equally effective.
b). Ha: : The two different teaching methods are NOT equally effective
2. Determine if the test is One-Tailed or Two-Tailed = .N-A.
3. Specify the Level of Significance (a) = .0.05
4. Determine the degree of freedom, df = for T- test the degree of freedom =n1+n2 - 2 = 8+10-2=16 .
5. Determine the TABULAR VALUE or CRITICAL VALUE (from the tabular value of T-test) = 1.746
6. Determine the computed value (using the test statistics or Statistical Tool/Formula).

The Test Statistics for Two-Sample Means:

(̅̅̅̅
𝑿𝟏 − ̅̅̅̅
𝑿𝟐 ) (𝟖𝟓 − 𝟖𝟕) −𝟐
𝒕𝑪𝑶𝑴𝑷𝑼𝑻𝑬𝑫 = = = = −𝟎. 𝟔𝟐𝟓
𝒏𝟏𝑽𝟏+𝒏𝟐 𝑽𝟐⬚ 𝒏𝟏+𝒏𝟐 𝟖 (𝟒𝟔)+𝟏𝟎(𝟑𝟔) 𝟖+𝟏𝟎 𝟕𝟐𝟖 𝟏𝟖
ට ⬚
∗ 𝒏𝟏 𝒏 ට 𝟖+𝟏𝟎−𝟐 ∗ 𝟖(𝟏𝟎) ට ∗
𝒏 +𝒏𝟐−𝟐
𝟏 𝟐 𝟏𝟔 𝟖𝟎

7. Make a decision. Compare the computed value to the critical/tabular value.

a. If the computed value is LESS than the critical value, ACCEPT the Null Hypothesis.
b. If the computed value is GREATER than or EQUAL to the critical value, REJECT the Null
Hypothesis.
Since the computed value is less than the critical value, therefore ACCEPT Ho!!!

8, Conclusion.

Ho: The two different teaching methods are equally effective.

Hypothesis Testing Page 3