MHB Multiple choice test : random variable

AI Thread Summary
A multiple choice test with 10 questions and five possible answers per question leads to the definition of a random variable, X, representing the number of correct answers. Each correct answer has a probability of 1/5, while each incorrect answer has a probability of 4/5, indicating a binomial distribution scenario. The discussion includes calculating various probabilities related to the number of correct answers, such as exactly 4 correct, more than 4, all correct, at least half correct, and between 5 and 8 correct answers. Participants confirm the use of binomial probability formulas to derive these probabilities. The conversation emphasizes understanding the distribution and calculations necessary for solving the test probability questions.
mathmari
Gold Member
MHB
Messages
4,984
Reaction score
7
Hey! 😊

A multiple choice test consists of 10 questions. For every question there are five possible answers, of which exactly one is correct. A test candidate answers all questions by chance.
(a) Give a suitable random variable with value range and probability distribution in order to work on part (b) with it.
(b) Determine (with intermediate steps) the probability that
(i) exactly 4 questions were answered correctly,
(ii) more than 4 questions have been answered correctly,
(iii) all tasks have been answered correctly,
(iv) at least half of the questions were answered correctly,
(v) at least 5 and at most 8 questions have been answered correctly.For (a) :
Let $X$ be a random variable that describes the number of correct answers out of $10$, right?
For each correct answer the probability is equal to $\frac{1}{5}$ and each wrong answer has the probability $\frac{4}{5}$.

Is that correct so far? :unsure:
 
Physics news on Phys.org
mathmari said:
For (a) :
Let $X$ be a random variable that describes the number of correct answers out of $10$, right?
For each correct answer the probability is equal to $\frac{1}{5}$ and each wrong answer has the probability $\frac{4}{5}$.
Hey mathmari!

Yep. (Nod)

We still need to identify the probability distribution function for (a) though. 🤔
 
Klaas van Aarsen said:
Yep. (Nod)

So is the range of the random variable $\{0,1,2,3,4,5,6,7,8,9,10\}$ ? :unsure:
Klaas van Aarsen said:
We still need to identify the probability distribution function for (a) though. 🤔

Do we have to calculate for each value of the random variable the corresponding probability? :unsure:
 
mathmari said:
So is the range of the random variable $\{0,1,2,3,4,5,6,7,8,9,10\}$ ?

Yep. (Nod)
mathmari said:
Do we have to calculate for each value of the random variable the corresponding probability?

It think we should just identify the name of the probability distribution and its parameters. 🤔
If we want to, we can also calculate the corresponding probabilities of the possible outcomes.
 
Klaas van Aarsen said:
It think we should just identify the name of the probability distribution and its parameters. 🤔

What do you mean by the name of probability distribution an its parameters? Is it $P(X=k)$ with $k\in\{0, 1, 2, \ldots , 10\}$? :unsure:
Klaas van Aarsen said:
If we want to, we can also calculate the corresponding probabilities of the possible outcomes.

This is then part (b), or not? :unsure:
 
mathmari said:
What do you mean by the name of probability distribution an its parameters? Is it $P(X=k)$ with $k\in\{0, 1, 2, \ldots , 10\}$?

You wrote:
mathmari said:
For each correct answer the probability is equal to $\frac{1}{5}$ and each wrong answer has the probability $\frac{4}{5}$.
This describes a Bernoulli distribution with parameter $p=\frac 15$, which is for an experiment with a single yes-no question.
However, we don't have a single yes-no question, but we have 10 questions.
We are looking for a distribution for the number of successes in a sequence of $n$ independent experiments, each asking a yes–no question, and each with its own Boolean-valued outcome: success (with probability $p$) or failure (with probability $q = 1 − p$). (Sweating)

mathmari said:
This is then part (b), or not?
More or less. We can use a table with the probabilities of each possible outcome to help us answer each of the questions in (b).
But we can also find the answers for (b) in a different fashion. 🤔
 
Klaas van Aarsen said:
This describes a Bernoulli distribution with parameter $p=\frac 15$, which is for an experiment with a single yes-no question.
However, we don't have a single yes-no question, but we have 10 questions.
We are looking for a distribution for the number of successes in a sequence of $n$ independent experiments, each asking a yes–no question, and each with its own Boolean-valued outcome: success (with probability $p$) or failure (with probability $q = 1 − p$). (Sweating)

So do we have a binomial distribution? :unsure:
Klaas van Aarsen said:
More or less. We can use a table with the probabilities of each possible outcome to help us answer each of the questions in (b).
But we can also find the answers for (b) in a different fashion. 🤔

Do we have the following ?

(i) $p_X(4)=P(X=4)=\binom{10}{4}p^4(1-p)^{10-4}=\binom{10}{4}\left (\frac{1}{5}\right )^4\left (\frac{4}{5}\right )^{6}$
(ii) $P(X>4)=1-P(X\leq 4)=1-\sum_{i=0}^4P(X=i)=1-\sum_{i=0}^4\binom{10}{i}p^i(1-p)^{10-i}=1-\sum_{i=0}^4\binom{10}{i}\left (\frac{1}{5}\right )^i\left (\frac{4}{5}\right )^{10-i}$
(iii) $p_X(10)=P(X=10)=\binom{10}{10}p^{10}(1-p)^{10-10}=\left (\frac{1}{5}\right )^{10}$
(iv) $P(X\geq 5)=P(X>4)=1-\sum_{i=0}^4\binom{10}{i}\left (\frac{1}{5}\right )^i\left (\frac{4}{5}\right )^{10-i}$
(v) $P(5\leq X\leq 8)=P(X\leq 8)-P(X<5)=P(X\leq 8)-P(X\leq 4)=[1-P(X> 8)]-[1-P(X> 4)]=[1-P(X=9)-P(X=10)]-\left [1-1+\sum_{i=0}^4\binom{10}{i}\left (\frac{1}{5}\right )^i\left (\frac{4}{5}\right )^{10-i}\right ]=[1-\binom{10}{9}\left (\frac{1}{5}\right )^9\left (\frac{4}{5}\right )^{10-9}-\left (\frac{1}{5}\right )^{10}]-\left [1-1+\sum_{i=0}^4\binom{10}{i}\left (\frac{1}{5}\right )^i\left (\frac{4}{5}\right )^{10-i}\right ]$

:unsure:
 
Yep. All correct. (Sun)
 

Similar threads

Back
Top