Stat 130n- Biostatistics

Answers to the Learning Activities in Module 3, Lessons 3.1-3.3

Module 3: Random Variables and Probability Distributions

Lesson 3.1: Random Variables and Related Concepts

• The characteristic of interest in a study can be viewed as a random variable whose

values depend in the outcomes of a random experiment

• Random variable X – a function whose value x is a real number that is determined by

each outcome or sample point in the sample space

• Discrete sample space – contains a finite number of sample points or has as many
sample points as there are counting/natural numbers

• Discrete random variable – a random variable defined over a discrete sample space
• Cumulative distribution function (CDF) – is a function defined for any real number x
as 𝐹(𝑥) = 𝑃 (𝑋 ≤ 𝑥)

• Probability mass function (PMF) or discrete probability distribution – alternative

tool to the CDF that we can use for modeling the behavior of a random variable, a
function defined for any real number x as 𝑓(𝑥) = 𝑃 (𝑋 ≤ 𝑥)

• the sum of 𝑓(𝑥) for all mass points must always be equal to 1 for a derived PMF to be a
valid one

• also used to compute important summary measures such as mean and the standard
deviation of the random variable

• Instead of a PMF, the continuous random variable has Probability density function
(PDF) or continuous probability distribution

• The characteristics of a probability distribution can also be summarized using descriptive

measures through the concept of expectation

• For a random variable X, we used the expected value of X, also referred to as

the mean of X, denoted by E(X) to know where the center of the distribution lies;

• variance of X, denoted by 𝝈𝟐 or Var(X) if we want to know how concentrated or

dispersed the values are of random variable X about its mean 𝜇.
Learning Activity
Perform as indicated.

1. A fair die is tossed twice. Define 𝑿 = number of tosses resulting in a number greater
than 4.

a. Identify all possible values of 𝑿. Present the event associated with each 𝒙.
b. Construct the probability mass function of 𝑿 by completing the table below.
By the multiplication principle (Module 2,
Solution: Lesson 2.3) we have 62 or 6x6=36
equally likely sample points
a) This involves a random experiment with 36 equally likely sample points. In two
tosses, we can have none, one, or both tosses resulting in a number greater than 4.
Thus, the values of 𝒙 with their associated events are as follows:

𝑥 Event associated with 𝑋 = 𝑥

0 {(1,1), (1,2), (1,3), (1,4), (2,1), (2,2), (2,3), (2,4), (3,1), (3,2), (3,3), (3,4),
(4,1), (4,2), (4,3), (4,4)}
1 {(1,5), (1,6), (2,5), (2,6), (3,5), (3,6), (4,5), (4,6), (5,1), (5,2), (5,3), (5,4),
(6,1), (6,2), (6,3), (6,4)}
2 {(5,5), (5,6), (6,5), (6,6)}

b. Using the classical approach of assigning probabilities, the probability mass function
of 𝑿 is as follows:
16/36

𝑥 0 1 2
𝑓(𝑥) = 𝑃(𝑋 = 𝑥) 4/9 4/9 1/9 4/36

16/36

Referring to the event associated with 𝑿 = 𝒙

𝑥 Event associated with 𝑋 = 𝑥

0 {(1,1), (1,2), (1,3), (1,4), (2,1), (2,2), (2,3), (2,4), (3,1), (3,2), (3,3), (3,4), 16
(4,1), (4,2), (4,3), (4,4)}

1 {(1,5), (1,6), (2,5), (2,6), (3,5), (3,6), (4,5), (4,6), (5,1), (5,2), (5,3), (5,4), 16
(6,1), (6,2), (6,3), (6,4)}

2 {(5,5), (5,6), (6,5), (6,6)} 4

Total 36
2. A pen contains 3 black and 4 spotted calves. Suppose a sample of 2 calves are to be
selected at random from the pen. Construct the probability distribution table of the
number of black in the sample.

Let 𝑥 = the number of black calves in the sample

Solution:

This involves a random experiment with C(7, 2)=21 equally likely sample points. In selecting a
sample of 2 calves at random from the pen, we can have no black, one black, or both samples
are black calves.

Black= 3 calves Spotted = 4 calves

• 0 black and 2 spotted = C(3, 0) x C(4, 2) = (1) (6) = 6 sample points/ outcomes
• 1 black and 1 spotted = C(3, 1) x C(4, 1) = (3) (4) =12 sample points/ outcomes
• 2 black and 0 spotted = C(3, 2) x C(4, 0) = (3) (1) = 3 sample points/ outcomes

Black calves = {B1, B2, B3} Spotted calves = {S1, S2, S3, S4}

The values of 𝑥 with their associated events are as follows:

𝑥 Event associated with 𝑋 = 𝑥

0 {(S1, S2), (S1, S3), (S1, S4), (S2, S3), (S2, S4), (S3, S4)} 6
1 {(B1, S1), (B1, S2), (B1, S3), (B1, S4), (B2, S1), (B2, S2), (B2, S3), (B2, S4), (B3, S1), 12
(B3, S2, (B3, S3), (B3, S4)}

2 {(B1, B2), (B1, B3), (B2, B3)} 3

Total 21

Thus, using the classical approach of assigning probabilities the probability distribution table of
the number of black (x) in the sample is as follows:

𝑥 0 1 2
𝑓(𝑥) = 𝑃(𝑋 = 𝑥) 6/21 or 2/7 12/21 or 4/7 3/21 or 1/7
3. It is useful for Fire Departments to have a model for 𝑿 = number of direct injuries
occurring in a fire incident. They can use this to determine the level of medical services
to make available each time. Suppose the PMF of 𝑿 is as follows:

𝑥 0 1 2 3 4 5 6 7 8 9
𝑓(𝑥) .08 .09 .09 .14 .18 .16 .09 .06 .06 .05

Use this PMF to determine the probabilities of the following events:

a. Event that there will be at least 1 direct injury

𝑃(𝑋 ≥ 1) = 1 − 𝑃(𝑋 < 1) = 1 − 𝑃(𝑋 = 0) = 1 − .08 = 𝟎. 𝟗𝟐

Note: You can also find the 𝑃(𝑋 ≥ 1) by adding all the probabilities 𝑃(𝑋 = 1) +𝑃(𝑋 = 2) +𝑃(𝑋 = 3)
+𝑃(𝑋 = 4) +𝑃(𝑋 = 5) +𝑃(𝑋 = 6) +𝑃(𝑋 = 7)+𝑃(𝑋 = 8) +𝑃(𝑋 = 9) =0.92

b. Event that there will be more than 2 but less than 8 direct injuries

𝑃(2 < 𝑋 < 8) = 𝑃(𝑋 = 3, 4, 5, 6, or 7)

= 𝑃(𝑋 = 3) + 𝑃(𝑋 = 4) + 𝑃(𝑋 = 5) + 𝑃(𝑋 = 6) + 𝑃(𝑋 = 7)
= 0.14 + 0.18 + 0.16 + 0.09 + 0.06
= 𝟎. 𝟔𝟑

4. Consider the following probability distribution:

x 0 1 2 3 4
f(x) 0.1 0.4 0.3 0.1 0.1

Find the standard deviation of 𝑋.

Solution:

Let’s find the 𝐸(𝑋) = 𝜇 first,

𝐸(𝑋) = 𝜇 = 𝑥1 𝑓(𝑥1 ) + 𝑥2 𝑓(𝑥2 ) + ⋯ + 𝑥𝑛 𝑓(𝑥𝑛 ) = ∑ 𝑥𝑖 𝑓(𝑥𝑖 )

𝑖=1

𝐸(𝑋) = 𝜇 = 0(0.1) + 1(0.4) + 2(0.3) + 3(0.1) + 4(0.1)

𝐸(𝑋) = 𝜇 = 1.7 ≅ 2
Now, let’s find the standard deviation of X, recall that the standard deviation is just the square
root of the variance. We know that 𝜎 2 = Var(𝑋) = 𝐸(𝑋 − 𝜇)2, thus, standard deviation of X is
𝜎 = √𝐸(𝑋 − 𝜇)2 .

𝜎 = √(0 − 1.7)2 (0.1) + (1 − 1.7)2 (0.4) + (2 − 1.7)2 (0.3) + (3 − 1.7)2 (0.1) + (4 − 1.7)2 (0.1)

𝜎 = √(−1.7)2 (0.1) + (−0.7)2 (0.4) + (0.3)2 (0.3) + (1.3)2 (0.1) + (2.3)2 (0.1)

𝜎 = √0.289 + 0.196 + 0.027 + 0.169 + 0.529

𝜎 = √1.21
𝝈 = 𝟏. 𝟏

x 0 1 2 3 4
f(x) 0.1 0.4 0.3 0.1 0.1

Lesson 3.2: The Normal Distribution

• The Normal Distribution – is a continuous probability distribution that is very important

in the theory and application of classical statistical inference
- Use to model the behavior of quantitative characteristics of interest whose
relative frequency or density histograms can be approximated by the normal
curve

• The Standard normal random variable Z – can be used to determine the probabilities
of events associated with any normal random variable X if its mean and variance
are known

• Mesokurtic, leptokurtic, platykurtic

• Using Table B.1 (Z- table)

Learning Activity
Perform as indicated.

1. Suppose a random variable Y is normally distributed with mean 200 and standard
deviation of 20. Calculate the following probabilities:

Given: 𝜇 = 200 𝜎 = 20
a) P(Y<170)
𝑌−𝜇 170−200
For 𝑌 = 170, the corresponding value of 𝑍 is 𝑧 = = = −1.5. Therefore,
𝜎 20

𝑃(𝑌 < 170) = 𝑃(𝑍 < −1.5) = 𝐹(−1.5) = 0.0668

From Table B.1-Z-table

Note: The Z-table is biased to the area below the value of Z under the standard normal curve.
 b) P(150<Y<250)

Given: 𝜇 = 200 𝜎 = 20
The 𝑧 values corresponding to 𝑌 = 150 and 𝑌 = 250 are

150−200 250−200
𝑧= ≅ −2.5 and 𝑧 = ≅ 2.5, respectively. We calculate
20 20

𝑃(150 < 𝑌 < 250) = 𝑃(−2.5 ≤ 𝑍 ≤ 2.5)

= 𝐹(2.5) − 𝐹(−2.5)
= 𝑃(𝑍 ≤ 2.5) − 𝑃(𝑍 ≤ −2.5)
= 0.9938 − 0.0062
= 0.9876

From Table B.1: Z-table

𝑃(𝑍 ≤ 2.5)

𝑃(𝑍 ≤ −2.5)
 c) P(Y>225)

Given: 𝜇 = 200 𝜎 = 20
225−200
For 𝑌 = 225, the corresponding value of 𝑍 is 𝑧 = = 1.25. We find
20

𝑃(𝑌 > 225) = 𝑃(𝑍 > 1.25) = 1 − 𝑃(𝑍 < 1.25) = 1 − 0.8944 = 0.1056

From Table B. 1

1.2+0.05=1.25

𝑃(𝑍 < 1.25)

2. A large group of students took a test in Biostatistics and their scores have a mean
of 70 and a standard deviation of 10. If we can approximate the distribution of these
scores by a normal distribution.

a) What percent of the students scored higher than 80?

b) What percent of the students have a score between 50 and 90?
c) Suppose the passing score for the test is 60 points. What percent of the students should
pass the test?
d) What should be the score of a student to be in the top 10% of all takers?

Solution:

Let X denote the score of a student who took a test in Biostatistics. With a mean score of the
students who took the test, 𝜇 = 70 and standard deviation, 𝜎 = 10, we can transform X into the
standard normal random variable Z by

𝑋 − 70
𝑍=
10
 a) What percent of the students scored higher than 80?

The percent of the students scored higher than 80 also denotes the probability that a
randomly chosen student scored higher than 80 points.

80−70
For 𝑥 = 80, the corresponding value of 𝑍 is 𝑧 = 10
=1

Therefore,

𝑃(𝑋 > 80) = 𝑃(𝑍 > 1) = 𝐹(1)

𝑃(𝑋 > 80) = 𝑃(𝑍 > 1)
= 1 − 𝑃(𝑍 < 1)
= 1 − 0.8413 (value taken from Table B.1)
𝑷(𝑿 > 𝟖𝟎) = 𝟎. 𝟏𝟓𝟖𝟕

The probability value 0.1587 corresponds to the area above 𝑧 = 1 under the standard normal
curve. We can say that the probability of this event happening that is a student scored higher
than 80 in the test is 0.1587. We can conclude that 15.87% or about 16% of the students
scored higher than 80 points under this model.

Using Table B.1 to find P(Z<1)

b) What percent of the students have a score between 50 and 90?

The percent of the students who scored between 50 and 90 points also denotes the
probability that a randomly chosen student scored between 50 and 90 points.

Given: 𝑥1 = 50 kg, 𝑥2 = 90 kg

The corresponding standardized value, 𝑧1 and 𝑧2 are:

 50 − 70
𝑧1 = = −2
10
90 − 70
𝑧2 = =2
10
Thus, the probability that the value of 𝑥 are between 50 and 90 is:

𝑃(𝑥1 ≤ 𝑥 ≤ 𝑥2 ) = 𝑃(50 ≤ 𝑥 ≤ 90) = 𝑃(𝑧1 ≤ 𝑧 ≤ 𝑧2 ) = 𝑃(−2 ≤ 𝑧 ≤ 2)

𝑃(50 ≤ 𝑋 ≤ 90) = 𝑃(−2 ≤ 𝑍 ≤ 2)
= 𝐹(2) − 𝐹(−2)
= 𝑃(𝑍 ≤ 2) − 𝑃(𝑍 ≤ −2)
= 0.9772 − 0.0228
𝑷(𝟓𝟎 ≤ 𝑿 ≤ 𝟗𝟎) = 𝟎. 𝟗𝟓𝟒𝟒
The probability value 0.9544 is the area between 𝑧 = −2 and 𝑧 = 2 under the standard
normal curve. We can say that the probability of observing this event, that is, the student
scored between 50 and 90 points is 0.9544. Thus, 95.44% of the students are expected to
have a score within this interval [50, 90] points under this model.

Using the Z-table (Table B.1) to find the value for F(2) and F(-2)

𝑃(𝑍 ≤ 2) = 𝐹(2) = 0.9772

 𝑃(𝑍 ≤ −2) = 𝐹(−2) = 0.0228

c) Suppose the passing score for the test is 60 points. What percent of the students
should pass the test?

A student should have a score higher than or equal to 60 points to pass the test. Thus, the
percent of the students who will pass denotes the probability that a randomly chosen student
scored higher than or equal to 60 points.

60−70
For 𝑥 = 60, the corresponding value of 𝑍 is 𝑧 = 10
= −1

Therefore,

𝑃(𝑋 ≥ 60) = 𝑃(𝑍 ≥ −1) = 𝐹(−1)

𝑃(𝑋 ≥ 60) = 𝑃(𝑍 ≥ −1)
= 1 − 𝑃(𝑍 ≤ −1)
= 1 − 0.1587 (value taken from Table B.1)
𝑃(𝑋 ≥ 60) = 0.8413

The probability value 0.8413 corresponds to the area above 𝑧 = −1 under the standard
normal curve. We can say that the probability of this event happening, that is, a student
scored higher than or equal to 60 in the test is 0.8413. We can conclude that 84.13% of the
students should pass the test under this model.

Using the Z-table (Table B.1) to find the value for F(-1)
 𝑃(𝑍 ≤ −1) = 𝐹(−1) = 0.1587

d) What should be the score of a student to be in the top 10% of all takers?

Solution:

The top 10% of all the takers corresponds to the area under the standard normal curve for the
values greater than some value of 𝑧0 :

𝑃(𝑧0 ≤ 𝑍 ≤ +∞) = 0.10

First, 𝑧0 must be determined. From the table the value of 𝑧0 is approximately 1.28. In Table B.1
you can find 𝑃(𝑍 < 1.28) = 0.8997 and the area above it or 𝑃(𝑍 > 1.28) = 1 − 𝑃(𝑍 < 1.28) = 1 −
0.8997 ≅ 0.10. Thus, the value of 𝑧0 is 1.28. Now, 𝑧0 must be transformed to 𝑥0 , on the original
scale using the formula:
𝑥0 − 𝜇
𝑧0 =
𝜎
that is:
𝑥0 = 𝜇 + 𝑧0 (𝜎)
𝑥0 = 70 + (1.28)(10) = 82.8 points
Thus, a student should score higher than or equal to 82.8 points to be in the top 10% of all
the takers.
Using the Z-table (Table B.1) to find the value for 𝑧0

𝑃(𝑍 < 1.28)=0.8997 𝑃(𝑍 < 1.29)=0.9015

In this problem, we are looking for the value of 𝑧0 with area under the standard normal curve
equal 0.10 or with alpha, 𝛼 = 0.10

Note that, the value provided in Table B.1 (Z-table) is the area below the z-score, thus using the
Table B.1, we are going to look for the value of z with area equals to 1 − 0.10 = 0.90 or close to
0.90
In Table B.1, you cannot directly find or locate the exact value of z or z-score with area equals to
or probability equal to 0.90. Based on Table B.1, the exact z-score is between 1.28 and 1.29. You
can compute the exact z-score using interpolation formula but it was not covered in the discussion
in your learning guide (you can just search how to interpolate if you want to know ). To answer
this item, I used 1.28 instead of 1.29 because 𝑃(𝑍 < 1.28)=0.8997 is much closer to 0.90 than
𝑃(𝑍 < 1.29)=0.9015
However, at the bottom part of Table B.1, the z-score of the most commonly used alpha (𝛼) are
provided. As you can see in the figure below, the value of z at alpha = 0.10 is equal to 1.282.
Lesson 3.3: The Binomial and Poisson Distributions
• The Binomial and Poisson distributions are both discrete, in contrast to normal distribution
which is continuous
• Binomial distribution – used to model a behavior of a discrete random variable defined as
the number of “successes” in n trials from a binomial experiment
• PMF of a binomial random variable
𝑛
( ) 𝑝 𝑥 (1 − 𝑝)𝑛−𝑥 for 𝑥 = 0, 1, 2, … , 10
𝑓(𝑥) = 𝑃(𝑋 = 𝑥) = { 𝑥 .
0 for other real numbers .

• Poisson distribution – is a model for the relative frequency of rare events and data defined
as counts and often is used for determination of the probability that some event will happen
in a specific time or space interval, distance, area, volume, or some similar unit

• PMF of a Poisson random variable

Learning Activity

Perform as indicated. In solving probabilities of events associated with a distribution,

always define the random variable of interest and justify why its distribution can be
considered as such (binomial or Poisson).

1.) The expected proportion of cows with more than 4000 kg milk in the standard lactation
is 30%. If we buy 10 cows, knowing nothing about their previous records, what is the
probability:
(Hint: Let 𝑋 =number of cows with more than 4000 kg milk yield in standard lactation.)

Solution:

Let X= number of cows with more than 4000 kg milk yield in standard lactation (success) in the
10 cows bought (trials)

We can view the process of selecting and buying 10 cows as a binomial experiment because of
the following reasons. There are n=10 trials in which we select a cow to buy in each trial. Now,
there are only two (2) possible outcomes in each trial: the cow that you bought yield more than
4000 kg milk in standard lactation (success) and the cow that you bought do not yield more than
4000 kg milk in standard lactation (failure). The probability of success, 𝑝 = 0.30, is the same for
all the trials as stated in the problem. We are taking the cows to be independent in the sense that
the probability of success at a particular trial will not be affected by the outcomes of the previous
trials since we are assuming that the cows are independent in relation to the yield of milk in
standard lactation.

Thus, we can write 𝑋~Bi(𝑛 = 10, 𝑝 = 0.30) and its PMF is

10
( ) 0.30𝑥 (1 − 0.30)10−𝑥 for 𝑥 = 0, 1, 2, … , 10
𝑓(𝑥) = 𝑃(𝑋 = 𝑥) = { 𝑥 .
0 for other real numbers .

a) What is the probability that exactly 5 of them have more than 4000 kg milk yield?

10
𝑃(𝑋 = 5) = 𝑓(5) = ( ) 0.305 (1 − 0.30)10−5
5
= [𝐶(10, 5)]0.305 (0.70)5
= (252)(0.00243)(0.16807)
= (252)(0.0004084101)
= 0.102919345
≅ 𝟎. 𝟏𝟎𝟑

Therefore, the probability that exactly 5 of the 10 cows bought have more than 4000 kg milk
yield in standard lactation is 0.103. With a small probability value, we can say that this event
will less likely to happen.

b) What is the probability that at least two have more than 4000 kg?

𝑃(𝑋 ≥ 2) = 𝑃(𝑋 = 2) + 𝑃(𝑋 = 3) + 𝑃(𝑋 = 4) + 𝑃(𝑋 = 5) + 𝑃 (𝑋 = 6) + 𝑃(𝑋 = 7)

+ 𝑃(𝑋 = 8) + 𝑃(𝑋 = 9) + 𝑃(𝑋 = 10)

or using the complement of the event in finding its probability

𝑃(𝑋 ≥ 2) = 1 − [𝑃(𝑋 = 1) + 𝑃(𝑋 = 0)]

10
𝑃(𝑋 = 1) = 𝑓(1) = ( ) 0.301 (1 − 0.30)10−1
1
= [𝐶(10, 1)]0.301 (0.70)9
= (10)(0.30)(0.040353607)
= (10)(0.012106082)
= 0.121060821

10
𝑃(𝑋 = 0) = 𝑓(0) = ( ) 0.300 (1 − 0.30)10−0
0
= [𝐶(10, 0)]0.300 (0.70)10
= (1)(1)(0.028247524)
= 0.028247524
Thus,
𝑃(𝑋 ≥ 2) = 1 − [0.121060821 + 0.028247524]
= 1 − 0.149308345
= 𝟎. 𝟖𝟓𝟎𝟔𝟗𝟏𝟔𝟓𝟓
≅ 𝟎. 𝟖𝟓𝟏

The probability that at least two of the 10 cows bought have more than 4000 kg milk yield in
standard lactation is 0.851. Based on the probability value which is close to 1, this event will most
likely to happen.

2. A large industrial plant is being planned in a rural area. As a part of the environmental
impact statement, a team of wildlife scientists is surveying the number and types of
small mammals in the region. The number of field mice captured in a trap over a 24-
hour period is the characteristic of interest. It was found that the average number of
field mice captured per trap is 2.3. (Ott and Longnecker, 2016)

a) What is the probability of finding exactly four field mice in a randomly selected trap?

Solution:

Let 𝑌 =number of field mice captured in a trap over a 24-hour period. We can view 𝑌 as a
Poisson random variable with mean 2.3 and variance also 2.3. since the values of 𝑌 can
be seen as being generated by a Poisson experiment with the 24-hour period as the time
interval. Thus, we can find the probability that 𝑌 = 4 as follows:

𝑒 −2.3 2.34 (0.1002588437)(27.9841)

𝑓(4) = 𝑃(𝑌 = 4) = = = 0.1169022295
4! 24
≅ 𝟎. 𝟏𝟏𝟕

Therefore, there is a 0.117 probability of finding exactly four field mice in a randomly
selected trap. This event will less likely to happen.
b) What is the probability of finding more than four field mice in a randomly selected
trap?
(Hint: Use the complement of the event in finding its probability.)
We can find the probability that Y>4 as follows:
𝑃(𝑌 > 4) = 1 − 𝑃(𝑌 ≤ 4)
= 1 − [𝑓(0) + 𝑓(1) + 𝑓(2) + 𝑓(3) + 𝑓(4)]

Now, solving first for the individual probabilities, we have:

𝑒 −2.3 2.30 (0.1002588437)(1)

𝑓(0) = = = 0.10025884
0! 1
𝑒 −2.3 2.31 (0.1002588437)(2.3)
𝑓(1) = = = 0.2305953405
1! 1
𝑒 −2.3 2.32 (0.1002588437)(5.29)
𝑓(2) = = = 0.2651846416
2! 2
𝑒 −2.3 2.33 (0.1002588437)(12.167)
𝑓(3) = = = 0.2033082252
3! 6

𝑓(4) = 0.1169022295, from a)

Thus, we will have

𝑃(𝑌 > 4) = 1 − [𝑓(0) + 𝑓(1) + 𝑓(2) + 𝑓(3) + 𝑓(4)]

= 1 − 0.9162492805
≅ 𝟎. 𝟎𝟖𝟒

Therefore, there is a 0.084 or 8.4% probability of finding more than four field mice in a
randomly selected trap.