Z - test
Standard Error of mean: Standard error is a measure or
chance variation and is not an error or mistake. While
selecting the sample, the sample estimates of statistics will
differ from population parameters because of chance or
biological variability. Such a difference between sample and
population values is measured by statistic known as sampling
error or SE) or SE().
1. will cover 68% of sample means.
2. will cover 95% of sample means.
3. will cover 99% of sample means.
If calculate value of test statistic is outside this range then it is
considered that abnormal and statistically significant
difference. Here the level of significance is 5%.
EX: Mean haemoglobin is 12.0 gm%. SD is 1.5 gm% in a
sample size of 100. calculate 95% confidence interval for
the population mean.
Soln. Here,
Now 95% confidence interval will be
So, Hb from 11.7 gm% to 12.3 gm% is 95% confidence
interval that population or universe mean will lie within
EX: Systolic blood pressure of 566 males was taken.
Mean BP was found to be 128 mm and SD 13.05 mm.
Find 95% confidence limits of BP within which the
population mean would lie.
Soln. Here,
Now 95% confidence interval will be
- If the samples are large in size ( ≥30 ) the means of samples
in a population of differences between pairs of samples in a
population also follow normal distribution.
- When ‘Z’ test is applied to the sampling variability the
difference observed between a sample estimate and that of
population is expressed in terms of ‘Standard Error”.
The four criteria to apply Z – test for mean are:
1. The sample or samples must be randomly selected.
2. The data must be quantitative.
3. The variable is assumed to follow normal distribution in
the population.
4. The sample size must be larger than 30.
The ‘Z’ test for means has two applications
1. To test the significance of difference between a sample
mean and a known value of population mean .
Where, (when and are known)
(when is unknown and is known)
Conclusion:
If at level of significance.
Null hypothesis is rejected at level of significance.
Test is significant.
Otherwise
If at level of significance.
Null hypothesis is accepted at level of significance.
Test is insignificant.
EX:1 In a health survey of school children the mean
haemoglobin level of 55 boys was found to be 10.2 gm
per 100 ml with a standard deviation of 2.1 gm. Can it
be considered that this group of boys is identified from
a population with a mean Hb. Of 11.0 gm/100 ml?
Soln. Data given
The sample is taken from the given population with
mean Hb. Level of .
at 5% level of significance.
at 5% level of significance.
We reject our at 5% level of significance.
Test is significant.
The sample is not taken from the given population with mean
Hb. Level of .
EX:2 A random sample of 100 students have a mean weight
of 58 kg. with SD of 4 kg. Can it be concluded that this
group of students is identified from a population with a
mean weight of 60 kg.
Soln. Data given
There is no difference between sample mean weight of
58 kg. and population mean weight 60 kg.
at 1% level of significance.
at 1% level of significance.
We reject our at 1% level of significance.
Test is significant.
There is significant difference between sample mean weight of
58 kg. and population mean weight 60 kg.
Standard Error of difference between two means : The
test is required to be applied very often in Medical practice.
Means of a normally distributed variables in the two like or
unlike group are compared such as height, weight, blood
pressure, pulse rate etc. The action of a drug on a variable such
as BP or pulse rate is compared in two different groups when
placebo is given to the control group. Same way, the action of
two different drugs or of two different doses of the same drug
can be compared. As this test is based on normal distribution,
the sample should be large and random.
According to the null hypothesis, if the samples drawn are
random and of sufficiently large size, their means should not
differ from population mean or from means of each other. The
difference should be nil or zero.
If calculate value of test statistic is outside this range then it is
considered that abnormal and statistically significant
difference. Here the level of significance is 5%.
2. To test the significance of difference between two
sample means or between experiment sample mean and
control sample mean.
Where (when and are unknown)
(when is known and is unknown)
(when and are unknown)
EX: 1 In a study on growth of children, one group of 100
children had a mean height of 60 cm and standard
deviation of 2.5 cm while another group of 150
children had a mean height of 62 cm and standard
deviation of 3 cm. Is the difference between the two
groups statistically significant?
Soln. Data given
There is no significant difference between the means
of height in two groups or the two samples have come
from the same population.
at 1% level of significance.
at 1% level of significance.
We reject our at 1% level of significance.
Test is significant.
There is statistically significant difference between the means
of height in two groups.
EX:2 Hearing levels in two groups of school children with
normal hearing abnormality in the frequency of 500
cycles per second was found to be as follows.
Children Mean hearing Standard
threshold Deviation
Group I 62 15.5 decibels 6.5 decibels
Group II 76 20 decibels 7.1 decibels
Test if there is any difference between hearing levels
recorded in two groups?
Soln. Data given
There is no significant difference between the means
of hearing thresholds in two groups or the two samples
have come from the same population.
at 5% level of significance.
at 5% level of significance.
We reject our at 5% level of significance.
Test is significant.
There is statistically significant difference between the means
of hearing thresholds in two groups of children.