Probability question -- A test to see if a coin is fair....

In summary, the conversation discusses how to determine the probability of obtaining 4 heads in 5 flips of a fair coin, and the correct method for calculating the p-value in this scenario. It is determined that the p-value is equal to the probability of getting 4 or more heads, which is 3/16.
  • #1
Mr Davis 97
1,462
44

Homework Statement


[/B]
You want to see if a coin is fair. You flip it 5 times and count the number of heads. If H is the number of heads obtained in five flips of the coin, what is the P-value of the test when H equals 4?

Homework Equations


None

The Attempt at a Solution



To solve this problem, I thought that it would be correct to use the binomial PDF, to answer the question "If the probability of getting heads is .5, then what are the chances of getting 4 heads in 5 flips?" This gives .15625, which is not the right P-value. The correct answer is 3/16, but how do I get this value? What probability distribution to I use to obtain this P-value?
 
Physics news on Phys.org
  • #2
The p-value is not the probability of landing on the exact outcome. It is the probability of obtaining that or a more extreme outcome. I would also disagree on how the "correct" answer has chosen to define an outcome as extreme (getting 1 head is as extreme as getting 4, getting 0 is as extreme as getting 5 - you would typically not design a test which broke the symmetry).
 
  • #3
Mr Davis 97 said:

Homework Statement


[/B]
You want to see if a coin is fair. You flip it 5 times and count the number of heads. If H is the number of heads obtained in five flips of the coin, what is the P-value of the test when H equals 4?

Homework Equations


None

The Attempt at a Solution



To solve this problem, I thought that it would be correct to use the binomial PDF, to answer the question "If the probability of getting heads is .5, then what are the chances of getting 4 heads in 5 flips?" This gives .15625, which is not the right P-value. The correct answer is 3/16, but how do I get this value? What probability distribution to I use to obtain this P-value?

For binomial(5, 1/2), they take p-value = P(4) + P(5) = (5/32) + (1/32) = 3/16, so they take p-value = P(4 or more heads).
 

FAQ: Probability question -- A test to see if a coin is fair....

What is the purpose of a probability test for a coin?

The purpose of a probability test for a coin is to determine if the coin is fair or biased. This means that the test will determine if the coin has equal chances of landing on heads or tails, or if it consistently lands on one side more than the other.

How is a probability test for a coin conducted?

A probability test for a coin typically involves flipping the coin a certain number of times and recording the results. The more times the coin is flipped, the more accurate the results will be. The results are then compared to what would be expected for a fair coin, and any significant deviations indicate bias.

What is the significance of a probability test for a coin?

A probability test for a coin is significant because it can provide valuable information about the fairness of the coin. If a coin is found to be biased, it can affect the outcome of any decisions or games that involve the use of that coin. It can also be an indicator of potential flaws in the manufacturing or handling of the coin.

How is the probability calculated in a coin test?

The probability in a coin test is calculated by dividing the number of times the coin landed on the desired outcome (heads or tails) by the total number of flips. For example, if a coin was flipped 100 times and landed on heads 60 times, the probability of getting heads would be 60/100 or 0.6.

Are there any limitations to a probability test for a coin?

Yes, there are limitations to a probability test for a coin. The accuracy of the results can be affected by external factors such as the method of flipping the coin, the surface it is being flipped on, and any potential biases of the person conducting the test. It is also important to note that even a fair coin can have streaks of heads or tails due to random chance.

Back
Top