T-Distribution and confidence interval)

In summary: The question is asking to cut the confidence interval to 1/3 of its size, while the solution cuts it to 1/2. This could be a mistake in the question or in the solution. It is important to clarify with the instructor or the textbook author to get the correct solution.
  • #1
Willjeezy
29
0
I have a question I've been working on and I've noticed that my solution differs from the textbook solution. I would have considered it a mistake, but the next 4 questions are similar with the same solution "mistake"?

Aside from not converting the % to 0.885, I am wondering about part b. It says to cut the confidence interval to one third; in the solution they cut it to 2, why is that?

(oddly enough, in the next question, they ask to cut the confidence interval to 1/2 but use 1/3 in the solution)

is there something I am not understanding?

question.png
solution.png
 
Physics news on Phys.org
  • #2
That all seems good. It appears that there is a problem in the question itself. Either it is asking to shrink your interval to 1/3 of its size or 1/2 of its size.
The process they outlined seems reasonable.
If you were not using a t test, it would be simpler...
## E = \pm z_{\alpha/2}\frac{\sigma}{\sqrt{N}}##
## E/3 = \pm z_{\alpha/2}\frac{\sigma}{\sqrt{N_2}}##
So N_2 would be 9*N, or an increase of 8*N.
With the t-test, you should need slightly fewer, since you reduce the size of your ##t_{\alpha/2, n-1} ## term as n increases.

Not changing the percent to decimal does not affect the solution, since the units for the standard deviation and the mean were the same (%).
 
  • #3
RUber said:
That all seems good. It appears that there is a problem in the question itself. Either it is asking to shrink your interval to 1/3 of its size or 1/2 of its size.
The process they outlined seems reasonable.
If you were not using a t test, it would be simpler...
## E = \pm z_{\alpha/2}\frac{\sigma}{\sqrt{N}}##
## E/3 = \pm z_{\alpha/2}\frac{\sigma}{\sqrt{N_2}}##
So N_2 would be 9*N, or an increase of 8*N.
With the t-test, you should need slightly fewer, since you reduce the size of your ##t_{\alpha/2, n-1} ## term as n increases.

Not changing the percent to decimal does not affect the solution, since the units for the standard deviation and the mean were the same (%).

Yah the process seemed similar to mine, but I wasnt sure why in the solution the cut the interval in half instead of 1/3

So its safe to say that its wrong, and if i was asked to cut by 1/3, it should be 0.42/3 and not 0.42/2, correct?
 
  • #4
Right. The question does not seem to match te posted solution.
 

What is a T-Distribution?

A T-Distribution, also known as a Student's T-Distribution, is a probability distribution that is used to estimate the mean of a normally distributed population when the sample size is small or the population standard deviation is unknown. It is similar to the normal distribution, but has heavier tails, which allows for better estimation in small sample sizes.

What is a Confidence Interval?

A Confidence Interval is a range of values that is likely to include the true population parameter with a certain level of confidence. It is typically calculated from a sample, and the level of confidence is usually 95%. This means that if the same sample is taken multiple times, 95% of the time the true population parameter will fall within the calculated confidence interval.

How is the T-Distribution used in calculating Confidence Intervals?

The T-Distribution is used in calculating Confidence Intervals by providing critical values that correspond to a specific level of confidence. These critical values are used in conjunction with the sample mean, standard deviation, and sample size to calculate the upper and lower bounds of the confidence interval.

What are the assumptions for using a T-Distribution in calculating Confidence Intervals?

The assumptions for using a T-Distribution in calculating Confidence Intervals are that the population from which the sample is taken follows a normal distribution, the sample size is small (typically less than 30), and the population standard deviation is unknown. If these assumptions are not met, alternative methods such as the z-distribution or bootstrapping may be used.

Can the T-Distribution and Confidence Intervals be used for non-parametric data?

No, the T-Distribution and Confidence Intervals are used for parametric data, meaning that the data follows a specific distribution (such as normal or binomial). For non-parametric data, alternative methods such as bootstrapping or the Wilcoxon signed-rank test should be used.

Similar threads

  • Set Theory, Logic, Probability, Statistics
Replies
3
Views
723
  • Set Theory, Logic, Probability, Statistics
Replies
1
Views
723
  • Set Theory, Logic, Probability, Statistics
Replies
1
Views
661
  • Set Theory, Logic, Probability, Statistics
Replies
3
Views
1K
  • Set Theory, Logic, Probability, Statistics
Replies
3
Views
2K
  • Set Theory, Logic, Probability, Statistics
Replies
4
Views
1K
  • Set Theory, Logic, Probability, Statistics
Replies
1
Views
2K
  • Set Theory, Logic, Probability, Statistics
Replies
6
Views
3K
  • Set Theory, Logic, Probability, Statistics
Replies
7
Views
2K
  • Set Theory, Logic, Probability, Statistics
Replies
26
Views
3K
Back
Top