0% found this document useful (0 votes)
23 views7 pages

Binomial Probability Methods Explained

统计学资料

Uploaded by

sichenqi
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
23 views7 pages

Binomial Probability Methods Explained

统计学资料

Uploaded by

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

Statistics Assignment Page 1 of 7

Binomial Problems Using Two Methods

Binomial Situations and Experiments

Which of the following are binomial experiments:

A. How many fives would you get on 10 rolls of a fair die?


B. How many students will pass a calculus course if 36 are enrolled but the course hasn't
started yet?
C. You randomly select 100 people from a population of 1,000 and find the blood type for
each person. How many in your sample will have Rh-negative blood?
D. For each day from December through April, you record whether it snows.

The answers are: A and C. A binomial experiment has three defining qualities:
1. There's a known number of trials (the number of chances to succeed or fail).
2. The chance of success is the same for each trial.
3. Each trial is independent.

In example B, the odds aren't the same for each student. In example D, the odds will be
different at different times of the year: it's less likely to snow in April than in December.

When we record the number of successful outcomes of a binomial event, we refer to that
process as a binomial experiment. In a binomial setting, the variable of interest is discrete:
it's counted. The distribution B(n, p, X) is comprised of the probabilities of getting any X
number of successes in n trials of a binomial event with probability p for each trial.

Recall that there are two ways to find binomial probabilities:


1. Use the exact binomial method, either by hand or using a calculator. This works well
when you're dealing with a relatively small number of trials, but if you have a large n it
can get cumbersome.
2. Use the normal approximation to the binomial. This is often easier because you can use
z-scores to find probabilities.

As you continue to study inferential statistics, you'll find that the normal approximation is
often more convenient than the exact binomial, since it's generally easier to work with z­
scores than binomial calculations.

Copyright © 2021 Apex Learning. See Terms of Use for further information. Images of the TI-84 calculator are used
with the permission of Texas Instruments Incorporated. Copyright © 2011 Texas Instruments Incorporated.
Statistics Assignment Page 2 of 7
Binomial Problems Using Two Methods

Exact Binomial vs. Normal Approximation

This table summarizes each method and type of distribution:

Exact Binomial Normal Approximation


When to use When n is small, usually less than When n is large, and when doing
25 inference
Important In sampling situations, your The expected number of successes and
criteria population must be at least 20 failures must be at least 10 (NOTE:
times larger than your sample. many textbooks say 5).

N > 20n np≥ 10


n(1 - p) ≥ 10
(This sampling situation is referred
to as almost binomial-the trials NOTE: Many textbooks use the letter q
aren't exactly independent, but p in place of (1 - p), so you'll see it as
from one trial to the next is close nq ≥ 10.
enough to approximate the
binomial.)
How to find Calculator or binomial formula Calculator or normal curve table
probabilities
Discrete or Discrete Continuous
continuous
distribution?
Special Because it's a discrete Because it's a continuous distribution,
considerations distribution, you can directly find if you want to find a probability for an
probabilities for exact numbers of exact number (such as P(X = 2)),
outcomes (such as P(X = 2)). you'd need to use the continuity
correction and find the range from .5
When finding ranges of outcomes, below the number to .5 above the
it makes a difference whether the number.
endpoints are included; that is,
whether it's X ≤ 3 or X < 3. When finding probabilities for ranges of
outcomes, it doesn't make a difference
whether the endpoints are included;
that is, X ≤ 3 is the same as X < 3.

Because you're modeling a discrete


distribution with a continuous one, the
continuity correction will make your
calculations more precise.
Expected value np np
for number of
successes
Expected value n(l - p), or nq n(l - p), or nq
for number of
failures
Mean no no
Standard You usually don't need to know it √np(1 - p), or √npq
deviation in this context, but it's √np(1 - p) ,
or √npq.
Copyright © 2021 Apex Learning. See Terms of Use for further information. Images of the TI-84 calculator are used
with the permission of Texas Instruments Incorporated. Copyright © 2011 Texas Instruments Incorporated.
Statistics Assignment Page 3 of 7
Binomial Problems Using Two Methods

Almost Binomial
Remember the definition of a binomial event:
1. There's a set number, n, of identical trials.
2. The outcome of each trial is either a success or a failure.
3. The probability of success on a single trial is p.
4. The trials are independent.

There are times the independence criteria might not be satisfied, but for all practical
purposes the binomial distribution could still be used. The question is whether the total
population is large enough relative to the number of samples. If the population size is at
least 20 times larger than the sample size then it could be considered a binomial event.
(Some textbooks use 10 instead of 20, and 10 is also a valid criterion. The important thing
is that you use a widely accepted standard when justifying that your situation can be
considered binomial.)

Copyright © 2021 Apex Learning. See Terms of Use for further information. Images of the TI-84 calculator are used
with the permission of Texas Instruments Incorporated. Copyright © 2011 Texas Instruments Incorporated.
Statistics Assignment Page 4 of 7
Binomial Problems Using Two Methods

Assignment Questions

1. Let X be a binomial random variable with n = 20 and p = .3.

A. Write the expression for P(X = 6) in terms of the formula (:}p)x (1- P)n-x , then

evaluate it using your calculator. (1 point)

B. Using probability notation, P(X = x), write out all the probabilities you'd need to add
up in order to calculate P(X < 6), and then evaluate it using your calculator. (1 point)

C. For part B, why did you choose the upper bound that you did? If you were using the
normal approximation could you choose a different upper bound? Why or why not?
(1 point)

2. You're taking samples of size 10 from a population of 20 marbles and recording


the number of blue marbles in each sample. Why isn't this a binomial situation,
and why would it be almost binomial if you did the same experiment with a
population of 200 marbles? (2 points)

3. A psychiatrist believes that 80% of all Americans who visit doctors have
problems of a psychosomatic nature; that is, they have a physical problem with
a psychological cause. She decides to randomly select 25 patients to test her
theory.

A. Is this a binomial experiment? Explain why or why not, using the criteria for a
binomial experiment. Since this is a sampling situation, be sure to mention whether
it meets the criteria of an almost binomial situation. (2 points)

B. Assuming that the psychiatrist's theory is true, what's the expected value of X, the
number of the 25 patients who have psychosomatic problems? (In other words,
what's the expected number of successes.) (1 point)

C. Using the exact binomial calculation, find P(X < 15) on your calculator and show
what you entered. (1 point)

D.

1) Say you gather a sample of 25 and find that X = 13. Using an exact binomial
calculation and assuming that the psychiatrist's theory is true, what's the
probability of finding that in a sample of 25, X = 13 .That is, find P(X= 13).

2) Assuming this is a good sample, what would you conclude about the theory that
80% of all Americans who visit doctors have problems of a psychosomatic
nature? (2 points)

4. Explain the term continuity correction. Why does it work, and how do you use
it? (2 points)

Copyright © 2021 Apex Learning. See Terms of Use for further information. Images of the TI-84 calculator are used
with the permission of Texas Instruments Incorporated. Copyright © 2011 Texas Instruments Incorporated.
Statistics Assignment Page 5 of 7
Binomial Problems Using Two Methods

Copyright © 2021 Apex Learning. See Terms of Use for further information. Images of the TI-84 calculator are used
with the permission of Texas Instruments Incorporated. Copyright © 2011 Texas Instruments Incorporated.
Statistics Assignment Page 6 of 7
Binomial Problems Using Two Methods

5. Suppose that 10% of the fields in a given agricultural area with about 2,500
fields are infested with the sweet potato white fly. One hundred fields in this
area are randomly selected and checked for white fly.

A. Can this be considered an almost binomial situation? Why or why not? (1 point)

B. Using the exact binomial calculation, find the probability that from 4 to 16 randomly
selected fields are infected. (1 point)

C. Using the normal approximation, find the probability that from 4 and 16
randomly selected fields are infected. Be sure to check that np ≥ 10 and n(l - p) ≥
10. Find the probability in two ways: with and without the continuity correction. How
do the answers from the two methods compare, and how do you account for the
difference? (3 points)

D. Sketch the probability distribution and the area you found when using the normal
approximation with the continuity correction. (2 points)

6. A very common dahlia plant with red petals is crossed with another very
common plant having streaky petals. The probability that an offspring from this
cross has red flowers is .75. Let X be the number of plants with red flowers
grown from 100 randomly collected seeds from a very large population (N >
10,000) of this cross.

A. Does the random variable X have a binomial distribution? If not, why not? If so, what
are the values of n and p? What's the mean, or expected number of plants with red
petals, and what's the standard deviation of the distribution? (1 point)

B. Find P(X > 80) using the exact binomial calculation. (Use your calculator.) (1 point)

C. Is it appropriate to use the normal approximation to the binomial probability for this
problem? Why? Support your answer with numbers, and a test. (1 point)

D. Using the normal approximation with the continuity correction, calculate the
probability that you'll observe more than 80 plants with red petals. Indicate the X­
values and z-scores involved. How does your answer compare with the one you got
in part C, and how do you account for the difference? (2 points)

Copyright © 2021 Apex Learning. See Terms of Use for further information. Images of the TI-84 calculator are used
with the permission of Texas Instruments Incorporated. Copyright © 2011 Texas Instruments Incorporated.
Statistics Assignment Page 7 of 7
Binomial Problems Using Two Methods

Acknowledgements
Question 1:
This question is based on question 6.33 from page 234 of Introduction to Probability and Statistics, Tenth Edition,
by W. Mendenhall, R. Beaver, and B. Beaver. Copyright© 1999 by Brooks Cole, division of Thompson Learning
Incorporated. Further reproduction is prohibited without permission of the publisher.

Question 3:
This question is based on question 5.61 from page 209 of Introduction to Probability and Statistics, Tenth Edition,
by W. Mendenhall, R. Beaver, and B. Beaver. Copyright© 1999 by Brooks Cole, division of Thompson Learning
Incorporated. Further reproduction is prohibited without permission of the publisher.

Question 5:
This question is based on question 5.26 from page 192 of Introduction to Probability and Statistics, Tenth Edition,
by W. Mendenhall, R. Beaver, and B. Beaver. Copyright© 1999 by Brooks Cole, division of Thompson Learning
Incorporated. Further reproduction is prohibited without permission of the publisher.

Question 6:
This question is based on question 5.67 from page 210 of Introduction to Probability and Statistics, Tenth Edition,
by W. Mendenhall, R. Beaver, and B. Beaver. Copyright© 1999 by Brooks Cole, division of Thompson Learning
Incorporated. Further reproduction is prohibited without permission of the publisher.

Copyright © 2021 Apex Learning. See Terms of Use for further information. Images of the TI-84 calculator are used
with the permission of Texas Instruments Incorporated. Copyright © 2011 Texas Instruments Incorporated.

Common questions

Powered by AI

Applying the continuity correction is appropriate when using a continuous normal distribution to approximate a discrete binomial distribution. It involves adjusting the probability range by 0.5 units for each boundary (e.g., finding probability for X = 5 entails calculating between 4.5 and 5.5) to correct for discreteness, improving accuracy by aligning bins more closely with the continuous curve .

The size of the population relative to the sample size influences whether a scenario is treated as binomial because it affects trial independence. A large population (at least 20 times the sample size) implies minimal changes in probability from one trial to the next, justifying the binomial model despite finite sampling. If the population is small, probability changes between trials can violate the independence condition, necessitating alternative statistical methods .

The normal approximation to the binomial distribution can be advantageous because it's generally easier to work with z-scores associated with the normal distribution than with exact binomial calculations, especially for large sample sizes. The normal approximation simplifies the calculation process when n is large, but it's necessary to ensure that the sampling distribution meets requirements such as np ≥ 10 and n(1-p) ≥ 10 .

For the normal approximation to be valid in the context of binomial probabilities, two key conditions must be met: the expected number of successes, np, and the expected number of failures, n(1-p), must both be at least 10. This ensures the binomial distribution's shape is close enough to normal for the approximation to be accurate .

To determine if marble sampling is a binomial experiment, evaluate if sampling meets the four criteria: fixed number of trials, dichotomous outcome (e.g., blue marble vs. others), constant probability of success per trial, and independence of trials. If sampling extends over a small population, trial independence is violated, but if the population is sufficiently large (e.g., 20 times the sample size), it approximates a binomial situation .

The exact binomial calculation is preferable for small sample sizes, usually 25 or fewer trials, since the normal approximation may not be precise when the number of trials is small and the probability distribution far from normal. In such cases, the exact calculation provides more accurate probabilities for specific outcomes, emphasizing its importance in small-scale sampling .

The expected value of a binomial distribution is calculated as np, which represents the average number of successes over multiple trials, while the standard deviation is √np(1-p), quantifying the variation around this expected value. These metrics collectively describe the distribution's central tendency and variability, facilitating statistical predictions about binomial experiment outcomes .

The continuity correction is applied when using the normal approximation for a binomial distribution, adjusting the range of outcomes by 0.5 units to make them more accurate. For example, to approximate P(X = 2) in a continuous model, the range is adjusted to include values from 1.5 to 2.5. This correction accounts for the difference between a discrete (binomial) and a continuous (normal) distribution, improving the precision of probability estimates .

A binomial experiment has three defining qualities: there is a known number of trials (the number of chances to succeed or fail), the chance of success is the same for each trial, and each trial is independent. For example, in the scenario 'How many fives would you get on 10 rolls of a fair die,' each roll is an independent trial with the same probability of rolling a five, making it a binomial experiment .

A sampling situation might be considered almost binomial when the population size is at least 20 times larger than the sample size because this reduces the dependence between trials, approximating binomial conditions. This approximation allows for the use of binomial formulas even when trials aren't completely independent, which is crucial for practicing sound statistical inference in real-world scenarios where populations are finite .

You might also like