Parametric Test Topic Contents

Parametric Test
(The t-test)
Topic Contents
3.1 Introduction
3.2 t-Test
3.3 One Sample Mean Test
3.4. Two Sample Mean Test (Pooled-Variance t Model)
3.5 Correlated t-test
Learning Outcomes
1. Manifest knowledge and skills on the principles and concepts of t-test.
2. Solve problems involving t-test.
Overview
This chapter discusses on the parameter test, I.e., t-test. This chapter also
discussed the determination of one sample mean test, two sample mean test, and
correlated t-test. The difference between one sample mean test and two sample
mean test is also discussed in this chapter. Example problems in this chapter is also
discussed in this chapter. Example problems in this chapter are discussed in step by
step manner. In addition, this chapter provides solution with the aid of computer,
i.e., Microsoft excel.
3. 1 Introduction
When the value of the population variance or population standard deviation
is not known and the sample size is small, assume again that the population being
sampled from has roughly the shape of the normal distribution; the decision shall be
based on the test statistics also called t-distribution or Student-Fisher Distribution.
Recall that t-test is also a parametric test and it is a statistical test for small sample.
Basically, the sample size should be less than 30.
3.2 t-Test
The t-test is commonly used in many educational researches in the

undergraduate and graduate studies. There are two types of t-test formula, i.e., one
sample mean test and two sample mean test. It involves the use of degrees of
freedom for the distribution. The degrees of freedom is important in the t-test and it
varies according to the particular type of t-test to be used. Hence, the formula for
the degree of freedom for one sample mean test is different from the two sample
mean test.
The following conditions should be met before the t-test is used:

1. The population standard deviation is not known and only the sample
variance is available for use as estimate.
2. The sample means are taken from small sample sizes where each n<30.
3.3 One Sample Mean Test
One sample mean test is a test statistics that compares a sample mean with
population mean and the sample standard. This type of test involves the use of
degree of freedom which can be obtained from the sample size minus one. Equation
3.1 and 3.2 are the mathematical formula for the computation of one sample mean
test and degree of freedom respectively.
Equation 3.1 Equation 3.2

t=
(x̄ - µ) √n df= n-1
s Where df= degree of freedom
Where t= t-test n= sample size
x̄ = sample mean
µ = population mean
s = sample standard deviation
n = sample size
Example Problem 3.1
A researcher is conducting a study in which she tests the claim that the
Payatas housing program contains family units of average size of 3.3 people. A
random sample of 24 families from Payatas project showed that there is an average
of 4.3 people per family unit and standard deviation of 1.3 Does this indicate that the
average size of a family unit in Payatas is different from the national average of 3.3?
Use 5% level of significance to test this claim.
Given:
x̄ (sample mean) = 4.3 n (sample size) = 24

µ (population mean) = 3.3 s (population standard deviation) = 1.3
Solution:
Step 1: Write the problem

Problem: Is the average size of a family unit in Payatas within the national
average of 3.3?
Step 2: Determine the null hypothesis.
Null Hypothesis: The average size of a family unit in Payatas within the
national average of 3.3.
Step 3: Determine the alternative hypothesis
Alternative Hypothesis: The average size of a family unit in Payatas is not
within the national average of 3.3.
Step 4: Determine the degrees of freedom using equation 6.2.
df = n – 1 = 24 – 1 = 23
Step 5: Determine the critical value,
Critical Value: The alternative hypothesis is non-directional; hence the two
tailed test shall be used. At 5% level of significance and 23 degrees of freedom, the
critical value is 2.069, see appendix B.
Step 6: Determine the test statistic.
Test Statistic: One sample mean test, population standard deviation is not
known.
Solve the t-test value using 6.1
t=
(x̄ - µ) √n
s
(4.3 – 3.3)√24
=
1.3
(1)(4.90)
=
1.3
= 3.77
Step 8: Write the data in tabular presentation,
The National Size of a family Unit and the Average Size of a family Unit in Payatas
Mean Computed Tabular t Decision Interpretation

Value Value
(0.05), 23
df
Population 3.3 3.77 2.069
Reject Ho Significant
Sample 4.3
Since the computed value of 3.77 is greater than the critical value of 2.069,
reject the null hypothesis. Hence, at 5% level of significance and 23 degrees of
freedom, the average size of a family unit in Payatas is not within the national
average of 3.3.
3.4 Two Sample Means Test (Pooled – Variance t Model)

Two sample mean test is a test statistics that uses the sample means X 2 , and
X2 , obtained in small samples of sizes and n, and n, drawn from their respective
populations having and j2, and j2. Equation 3.3 and 3.4 show the mathematical
formula for the computation of two sample mean test and degree of freedom
respectively. The degrees of freedom can be obtained by getting the sum of the two
samples minus two.
Equation 3.3
x 2−x 1
t= 2
√ ( n −1 ) s + ( n2−2 ) s 22 1 1
1 1
n1 +n2−2
.
√ +
n1 n 2
Where: t = t – test
= sample mean of one group
=sample mean of the other group
S2 =sample variance of the other group
n1 = sample size of one group (n1 <30)
n2 = sample size of the other group (n2 <30)
Equation 3.4
df = n₁ + n₂ - 2
Where: df = degrees of freedom
n₁ = sample size of one group (n₁ < 30)
n₂ = sample size of the other group (n₂ < 30)
Example Problem 3.2
A researcher noted that 18 randomly selected private schools had a mean

National achievement test (NAT) score of 86 and a standard deviation of 6.25, while
24 randomly selected public schools got average NAT score of 80 and a standard
deviation of 3.45. Using 0.05 level of significance, can the researcher conclude that
the private school students conformed better than the public school students?
Given:
x1 (sample mean of one group)=80

x2 (sample mean of the other group)=86
s1² (mean variance of one group)=6.25
S2² (mean variance of the other group)=3.45
n1 (sample size of one group)=18
n₂ (sample size of the other group)=24
Solution:
Step 1: Write the problem,

Problem: is there a significant difference between the NAT average performances of
the students of private schools and public schools?
Step 2: Determine the variable,
Variable: The NAT general performances og the students of private schools
and public schools
Step 3: Determine the null hypothesis,
Null Hypothesis: There is no significant difference between the NAT average
performance of private schools and public schools.
Step 4: Determine the alternative hypothesis,
Alternative Hypothesis: The NAT average performance of private schools is high than
the NA, average performance of public schools.
Step 5: Determine the degrees of freedom using equation 3.4,
df = n₁ + n₂ -2 = 18 + 24 -2 = 42 – 2 =40
Step 6: Determine the critical value,
Critical Value: The alternative hypothesis is directional; hence the on-tailed test
shalle be used. At 0.05 significance level and 40 degrees of freedom, the critical
value us 1.684. see Appendix B.
x 2−x 1
t= 2
+ ( n2−2 ) s 22 1 1
√ ( n −1 ) s
1
n1 +n2−2
1
Substitute the data to compute for the amount below.

. +
n1 n 2 √
t=6/ (4.842)(0.312)= 3.97
Comparison of the Average NAT Performances

of Private Schools and Public Schools
School Number Mean Standard Computed Tabular Decision Interpretation

Deviation t Value t Value
(0.05),
40 df
Private 18 86 6.25 3.97 1.684 Reject Significant

Ho
Public 24 80 3.45
Since the computed value of 3.97 is greater than the critical value of 1.684,
reject the null hypothesis. Hence, at 5% level of significance and 40 degrees of
freedom, Average NAT performance of the private schools is higher than the
average NAT performance of the public schools.

Parametric Test Topic Contents

Uploaded by

Copyright:

Available Formats

Parametric Test Topic Contents

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Parametric Test Topic Contents

Uploaded by

Copyright:

Available Formats

Parametric Test

The t-test is commonly used in many educational researches in the

The following conditions should be met before the t-test is used:

3.3 One Sample Mean Test

Equation 3.1 Equation 3.2

Example Problem 3.1

x̄ (sample mean) = 4.3 n (sample size) = 24

Step 1: Write the problem

Step 8: Write the data in tabular presentation,

Mean Computed Tabular t Decision Interpretation

3.4 Two Sample Means Test (Pooled – Variance t Model)

Example Problem 3.2

A researcher noted that 18 randomly selected private schools had a mean

x1 (sample mean of one group)=80

Step 1: Write the problem,

Substitute the data to compute for the amount below.

Comparison of the Average NAT Performances

School Number Mean Standard Computed Tabular Decision Interpretation

Private 18 86 6.25 3.97 1.684 Reject Significant

You might also like