Chapter 5

z – test and t- test

Assumptions and Appropriate Use of the z
or t test
Parametric tests involve the use of population
parameters such as μ and σ and are known.

Because the z test involves the calculation using the

mean and standard deviation, it is used with interval
or ratio data.

Because we use the area under the normal curve, we

are assuming that the distribution of random samples
is normal.
 Z- Test
Used to measure how many standard deviation
units an individual raw score falls from the
mean of the distribution
Used to compare the means of
two independent samples
two dependent samples
two correlated (matched) samples
Used to determine the difference between two
population proportions.
 Z test
Testing the difference between
two proportions.
Example 1. ( Z-TEST)
Problem: A researcher has found out that 185 of 250
nonsmokers tested were able to succeed at an endurance
test, whereas only 116 of 200 smokers tested were
successful.

Test the hypothesis of equal success proportions against

unequal success proportions at 0.05 level of significance.
or
Test if the two proportions differ significantly from each
other.
Proportion of smokers and non smokers
who succeeded in an endurance test

Succeeded Total Proportion Percent

Smokers 116 200 116/200 58%

Non 185 250 185/250 74%

smokers
 Statistical Objective
To determine if there is a significant difference
between the proportions of smokers and non smokers
who succeeded at an endurance test.

Ho: There is no significant difference between the

proportions of smokers and non smokers who succeeded at
an endurance test.

Ha : There is a significant difference between the

proportions of smokers and non smokers who
succeeded at an endurance test.
 Statistical test: z-test
 = 0.05
 Finding the Critical Value of z
α = 0.05
Critical value of Z = + 1.96
Z - table
α = 0.05 α = 0.01
1- tailed + 1.645 + 2.33

2- tailed + 1.96 + 2.58

 Computing the Z-value
P1 - P2
Z= P1 Q1 P2Q2
n1 + n2

P1 - proportion from group 1 ( nonsmokers) = 185/250 = 0.74

P2 – proportion from group 2 (smokers) = 116/ 200 = 0.58
Q1 – 1 – P1 = 1 – 0.74 = 0.26
Q2 - 1- P2 = 1 – 0.58 = 0.42
n1 – sample size of group 1 = 250
n2 – sample size of group 2 = 200
Computed Z-value = 3.589
P1 - P2
Z= P1 Q1 P2Q2
n1 + n2

0.74 - 0.58
Z= (0.74)(0.26) (0.58)(0.42) = 3.589
250 + 200
 Statistical test: z-test
Critical value = +1.96
Computed value =3.589
 Compare the computed value and the
critical value
 Computed Value is greater than Critical Value, then reject Ho

Reject Ho Do not reject Ho Reject Ho

c.v.= - 1.96 0.0 c.v.=+ 1.96

3.589
 Statistical Conclusion
(Ha) There is a significant difference between the
proportions of smokers and non smokers who succeeded at
an endurance test.
 Research Conclusion
Smoking affects the person’s endurance.
 Z-test
Testing the significance of the
difference between means of two
independent samples.
Example 2. ( Z-TEST)
A researcher wants to find out if diet affects the
diastolic blood pressure of the individual.

Test if there is a significant difference between the mean

diastolic blood pressure of the vegetarians and non
vegetarians at α = 0.05.
 Data showing the mean diastolic blood pressure of
the respondents when they are classified as to diet.

Subgroup Mean – Standard Number of

Diastolic Deviation respondents
blood
pressure
Vegetarian 72.9 11.7 140

Non 71.8 11.4 143

Vegetarian
 Objective
To determine if there is a significant difference
between the mean diastolic blood pressure of the
vegetarians and non vegetarians.

Ho : There is no significant difference between the

mean diastolic blood pressures of the vegetarians and
non vegetarians.

Ha : There is a significant difference between the

mean diastolic blood pressures of the vegetarians and
non vegetarians.
 Statistical test: z test
 = 0.05
Critical Value = +1.96
Computed Value = 0.433
 Compare the computed value and the
critical value
 Computed Value is less than Critical Value, then do not reject Ho

Reject Ho Do not reject Ho Reject Ho

c.v.=- 1.96 0.0 c.v.=+ 1.96

0.433
 Statistical Conclusion
(Ho)There is no significant difference between the
mean diastolic blood pressure of the vegetarians and
non vegetarians.
 Research Conclusion
Diastolic blood pressure is not related to person’s diet.

Diet has no effect whatsoever on a person’s diastolic

blood pressure.
 T -test
Test of difference between means of
two independent groups.
Test of difference between means of
two dependent groups.
Test of difference between means of
two matched groups.(identical group)
 (T –test)
Test of difference between means
of two independent groups.
Example 1
A researcher wants to determine if the birth length of
infants is associated to his or her sex.

 Test if the mean birth length of male infants is

significantly different than that of female infants.

 Test at α = 0.05
Measurements of birth lengths of male and female
infants in small clinic

Groups Sample Mean Standard

size Deviation
Males 12 52.2 8.6

Females 9 50.7 9.5

 Objective
To determine if there is a significant difference
between the mean birth lengths of male and female
infants.

Ho : There is no significant difference between the

mean birth lengths of male and female infants.

Ha: There is a significant difference between the

mean birth lengths of male and female infants.
 Test Statistic : t-test
 = 0.05
Critical value = 1.729
Computed value = 0.373
 Compare the computed value and the
critical value
 Computed Value is less than Critical Value, then do not reject Ho

Do not reject Ho Reject Ho

c.v.= 1.729
0.373
 Statistical Conclusion
(Ho)There is no significant difference between the
mean birth lengths of male and female infants.
 Research Conclusion
Sex is not significantly associated to the birth length
of infants.
 Example 2 (t-test)
A researcher wants to find out if sex affects the
smoking practices.

Test if there is a significant difference in the mean number of

cigarettes smoked per day when the respondents are grouped
according to sex.

test at α = 0.05.
Average number of cigarettes smoked
per day by males and females
Groups Mean Standard Sample size
Deviation

Male 7.5 2.59 16

Female 3.5 2.59 14

 Objective
To determine if there is a significant difference in the
average number of cigarettes smoked by males and
females.

 Ho: There is no significant difference in the average

number of cigarettes smoked by males and females.

 Ha : There is a significant difference in the average

number of cigarettes smoked by males and females.
 SPSS output
number of sticks
sex Mean N Std. Deviation
male 7.5000 16 2.58
female 3.5000 14 2.59

T – value = 4.22 df = 28 p-value = 0.007

 Test Statistic – T test
Computed value = 4.22
Significance level:
α-value = 0.05
Significance value:
(p-value)= 0.007
 Compare the p-value and the
α- value
 P-value is less than α- Value, then reject Ho

Reject Ho Do not reject Ho

p=0.007 =0.05
 Statistical Conclusion
(Ha) There is a significant difference in the average
number of cigarettes smoked by males and females.
 Research Conclusion
Males smoke more number of cigarettes per day than
females.

 Smoking practices of respondents is associated to

their sex.

Sex of respondents is related to their smoking

practices.
 T -test
Test of significance between
means of two dependent
groups.
 Example
(t-test, dependent samples )
A researcher wants to find out if the experiment affects
the diastolic blood pressure.

Test if the pre experiment systolic blood pressure is different

from the post experiment systolic blood pressure.

test at α = 0.05.
Table showing the pre experiment and post
experiment systolic blood pressure
Res. # Pre Post Res. # Pre Post Res. # Pre Post
1 134 134 6 120 126 11 134 138
2 103 106 7 128 130 12 108 111
3 116 110 8 122 118 13 111 115
4 113 115 9 123 125 14 125 125
5 124 122 10 108 110 15 134 130
Respondent Number Pre Experiment SBP Post Experiment SBP
1 134 134
2 103 106
3 116 110
4 113 115
5 124 122
6 120 126
7 128 130
8 122 118
9 123 125
10 108 110
11 134 138
12 108 111
13 111 115
14 125 125
15 134 130
Table showing the mean of pre experimental and
post experimental systolic blood pressure
Mean Standard Sample size
deviation

Pre experimental
systolic blood 120.2 10.1 15
pressure

Post experimental
systolic blood 121.0 9.7 15
pressure
 Objective
To determine if there is a significant difference in the
average systolic blood pressure before and after the
experiment.

Ho: There is no significant difference in the average

systolic blood pressure before and after the
experiment.

Ha : There is a significant difference in the average

systolic blood pressure before and after the
experiment.
 SPSS output
T Test for dependent samples
(Pre) (Post) Mean Standard df t Sig.
Mean/ sd Mean/ sd difference Deviation

120.2 / 10.1 121.0 / 9.7 -0.800 3.4268 14 -.904 0.381

T – value = -0.904 critical value =2.145

 Test Statistic – T test
α = 0.05 ,
Critical Value = +2.145
Test statistic value = -0.904
 Compare the computed value and the
critical value
 Computed Value is less than Critical Value, then do not reject Ho.

Reject Ho Do not reject Ho Reject Ho

c.v.= -2.145 c.v. = + 2.145

-0.904
 Statistical Conclusion
 (Ho) There is no significant difference in the average
systolic blood pressure before and after the
experiment.
 Research Conclusion
The experiment does not have an effect on the systolic
blood pressure of respondents.
 T -test

Test of difference between means of

two matched groups.(identical group)
 Example
(t-test, two matched groups )
A researcher wants to find out if the teaching method
affects the reading performance of students.

Test if the new teaching method is better than the standard

teaching method.

test at α = 0.01
 Reading test scores for eight pairs of slow
learners
Pair New Method Standard Method
1 77 72
2 74 68
3 82 76
4 73 68
5 87 84
6 69 68
7 66 61
8 80 76
 Table showing the mean performance score of
students under new and standard method
Mean Standard Sample size
deviation

New method
76.00 6.93 8
Standard Method
71.63 7.0 8
 Objective
To determine if there is a significant difference in the
mean performance score of students when they are
grouped according to teaching methods.

Ho: There is no significant difference in the mean

performance score of students when they are grouped
according to teaching methods.

Ha : There is a significant difference in the mean

performance score of students when they are grouped
according to teaching methods.
 SPSS output
T Test for paired samples
(Pre) (Post) Mean Standard df t Sig.
Mean/ sd Mean/ sd difference Deviation

76 / 6.93 71.63 / 7.0 4.38 1.68 7 7.344 0.002

T – value = 7.344 df = 7 p-value = 0.002

 Test Statistic – T test
Test statistic value= 7.344
α -value = 0.01
p-value = 0.002
 Compare the p-value and the
α- value
 P-value is less than α- Value, therefore Reject Ho.

Reject Ho Do not reject Ho

α= 0.01
p=0.002
 Statistical Conclusion
( Ha) There is a significant difference in the mean
performance score of students when they are grouped
according to teaching methods.
 Research Conclusion
The new teaching method is significantly better than
the standard method in improving the “reading IQs”
of slow learners.
The new teaching method has significantly improved
the reading IQ of the students.
 Check your progress
 Determine if the mean level of surgical skills of nurses differ significantly when
they are grouped according to sex. Test at 5% significance level.

Group n Mean SD t p
Male 16 8.9 1.2
-0.123 0.183
Female 18 8.5 1.15

 Is this a test between means of independent groups or test of means between

dependent groups?
 Formulate the null hypothesis.
 What is the value of test statistic? What is the significance value?
 Is the result significant?
 Write your statistical conclusion.
END of Chapter 5