Scribd
Upload
1K views

Tests of Significance

Tests of significance are used to measure the probability that observed differences between sample means and proportions are due to chance rather than a real biological variation. There are two types of statistical tests: parametric tests which assume a normal distribution, and non-parametric tests which do not. An example is given of using a t-test to analyze blood pressure data from hypertension patients before and after taking a drug, to determine if the observed difference in means is statistically significant and indicates the drug had a hypotensive effect.

Uploaded by

Meer
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
1K views

Tests of Significance

Tests of significance are used to measure the probability that observed differences between sample means and proportions are due to chance rather than a real biological variation. There are two types of statistical tests: parametric tests which assume a normal distribution, and non-parametric tests which do not. An example is given of using a t-test to analyze blood pressure data from hypertension patients before and after taking a drug, to determine if the observed difference in means is statistically significant and indicates the drug had a hypotensive effect.

Uploaded by

Meer
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 111

TESTS OF

SIGNIFICANCE
BY : DR. PRABHJOT KAUR
PG RESIDENT
COMMUNITY MEDICINE DEPT.
TESTS OF
SIGNIFICANCE
• These are the statistical tests or
mathematical methods employed to measure
the probability or chance of occurrence of
biological variation, in samples by standard
error .

• It may be a difference between means and


proportions of sample and universe or
between the estimates of experiment and
control groups .
TESTS OF
SIGNIFICANCE
• Statistical tests can be classified into two :

• 1.) Parametric tests : which follow the


normal distribution.

• 2.) Non parametric tests : which do not


follow the normal distribution .
PROBABILITY
• It means a chance for occurrence of a specific
event .

• P = n/N where P=
probability n= no. of favourable cases
N = total number of both favourable and
unfavourable cases .

• The probability scale varies from 0 to 1 .


STANDARD ERROR OF
DIFFERENCE B/W TWO
MEANS
• In practice it is not possible to find
differences of large number of samples and
then find the SE of these differences . This
test is applied to one pair directly if SD of
two means are known. Assumption of no
difference between the two sample means is
made as per null hypothesis .
• EXAMPLE :

• Systolic blood pressure values of 10


hypertension patients before and after giving
a drug are given. Find out whether the drug
given has hypotensive effect or not ?
• S. N0 SBP Before drug SBP after drug

• 1 160 140

• 2 150 110

• 3 170 165

• 4 130 140

• 5 140 145

• 6 170 120

• 7 160 130

• 8 160 110

• 9 120 120

• 10 140 130
• SOLUTION :

• MEAN SD

• before drug 150 17.00

• after drug 131 17.13

• change 19 22.46

• t = 2.67

• DOF = 10 - 1 = 9
• Checking the value of p = 0.05 with df = 9 ,
value comes out 2.26

• which is less than the calculated value .

• thus , drug have hypotensive action .


NON
-PARAMETRIC
TESTS

You might also like