Probability

Rules of Probability
Course: Research Methodology and Biostatistics
Course code: MSDURMB-521
Teacher Name: MEHREEN FATIMA
SAP CODE: 834
Department: University Institute of Radiological Sciences and Technology
The University of Lahore
Content
• Additive rule of probability for mutually exclusive events
• Independent events and Multiplicative Rule of probability
• Additive rule for non mutually exclusive events
• Conditional probability
• Screening Test
Example 1: Relation between blood type and
disease
• Statement : In
a study of the relation between
blood type and disease, a sample of patients
ulcers, patients with gastric cancer and
control persons classified into the blood
group
(O,A, B).
• Table 1: Distribution of table by disease and blood group

Blood Peptic Ulcer Gastric Controls Total Probability

Group Cancer
O 983 383 2892 4285 4258/8766
=0.486
A 679 416 2625 3720 ?

B 134 84 570 788 ?

Total 1796 883 6087 8766

Probability 1796/8766
=0.205

(snedecor and Cochran 1980)

In this problem one can easily determine the
Probability that a patient selected at random will fall in blood group O
or A or B or he/ she is suffering from peptic ulcer or gastric cancer.

a) The probability that a person selected at random from 8766 cases

falls in blood type O group will be

P(blood type O) = 4285/8766 = 0.486

b) The probability that a person selected at random from 8766 cases
belongs to peptic ulcer group will be
P(peptic ulcer) = 1796/8766 = 0.205
Additive Rule of probability for Mutually
Exclusive Events
• What is the probability that person selected at random having a blood
group A or O .
• P(O) = 4258/8766 = 0.486
• P(A) = 3720/8766 = 0.424
• Rule : P(O or A) = P(O) + P(A)
=0.486 + 0.424
= 0.91
Independent Events and Multiplicative Rule of
Probability
• Independent Event
If the outcome of one event does not affect the
outcome of another event then these events are called
Independent event.
Independent Events and Multiplicative Rule of
Probability
• Rule
If two events A and B are independent then the probability that both A
and B occur is equal to the product of their respective probabilities i.e.

P(A and B) = P(A) P(B)

Example 2: Two coin tossed
• Suppose two coins are tossed. The probability that heads occur on
both coins.
P(two heads) = P(H1 and H2) =P( H1) P(H2)

P( H1) = 1/2
P( H2) = 1/2

P(H1 and H2) = 1/4

Additive Rule for non mutually Exclusive
Events
• Concept Joint Probability
• Non mutually exclusive events can jointly occur. What is the
probability that a person suffering from peptic ulcer and has a blood
type O.
P( peptic ulcer and blood type O) = P(A and B)
=983/8766 = 0.112

What is the Additive Rule?

Additive Rule for non mutually Exclusive
Events
• This Rule used to find out the probability that either of the two
events occur , When they are not mutually exclusive
• Rule P(A or B) = P(A) + P(B) – P(A and B)
• P(peptic ulcer or blood group O) = P(peptic ulcer) + P(blood group O)
-P(peptic ulcer and blood group O)
= 0.205 + 0.486 – 0.112 = 0.579
Conditional Probability
• The probability of an event A, given that an event B
i.e. P(A|B) has occurred , is called the conditional probability of A
given B, is defined as:
P(A|B) = P(A and B)/P(B)
P(B|A) = P(A and B)/P(A)
a) P( peptic ulcer and blood type O) = P(A and B)
=983/8766 = 0.112
b) P ( blood type O) = P(A) = 4258/ 8766 = 0.486
P(A|B) = P(A and B)/P(B)= 0.112/0.486 = 0.231
Exercise 1

• The following data relates to Chinese smoking and lung cancer study
in Beijing during 1990. Various types of probabilities can be calculated
based on data.
• Table 2: status of Lung Cancer by Smoking

Smoking Lung Cancer Total

Present Absent
Yes 126 100 226
No 35 61 96
Total 161 161 322
Questions

• 1) the probability that a selected person has lung cancer ?

• 2) the probability that a selected person is smoker?
• 3) the probability that a man has a lung cancer given that he is
smoker?
• 4) the probability that a man is smoker given that he has lung cancer?
• 5) the probability that a man is not smoker given that he has lung
cancer?
• 6) the probability that a man is smoker and does not have the lung
cancer?
• 1) the probability that a selected person has lung cancer
= 161/322 = 0.50
• 2) the probability that a selected person is smoker
= 226/322 = 0.702
• 3) the probability that a man has a lung cancer given that he is
smoker
= (126/322)/(226/322)
126 322
= ×
322 226
126
=
226
= 0.56
• ) the probability that a man is smoker given that he has lung cancer
=126/322)/(161/322)
126 322
= ×
322 161
126
=
161
=0.78
• 5) the probability that a man is not smoker given that he has lung
cancer?
• 6) the probability that a man is smoker and does not have the lung
cancer?
Screening Test
• Reliable:
A test is reliable if it provides consistent results when performed more than
once.
• Valid
The test is valid if it correctly identifies those who probably have the disease
And those who probably free from disease
Validity is measured by both sensitivity and specificity
Example 2
• In a BCP screening test of 1600 cancer for patients, the result are
given below:
• Table 3
Test Disease Total
Present Absent
Positive 570 TP a 150 FP b 720
Negative 30 FN c 850 TN d 880
Total 600 1000 1600
Specificity

• Specificity is the proportion of truly healthy people who are

identified by the screening test.
Specificity = d/b+d
= 850/1000
= 0.85
So 85% persons identified as normal or free from disease.
Sensitivity
• Sensitivity is the proportion of truly ill people in the screened
population who are identified as ill by the screening test
• Sensitivity = a/a+b
= 570/600
= 0.95
Report in %
It shows 95% pf patients are correctly identified.