You have two different populations (for example from: two welders, two candy coa

Question

You have two different populations (for example from: two welders, two candy coaters, two electro polishing devices, etc.) and one has a large variance in some performance parameter compared to the other. You would like to estimate the mean of the performance parameter. For you to have the same confidence in your estimate of the mean, which one do you need more samples from? If you have two different spray coaters and you are interested in the mean thickness of coatings that they produce what type of test would you run and what would be the null and alternative hypotheses for the test? In the question above what is symbol for the probability that you reject the null hypothesis when it was actually true? What are typical values used for that probability? If you are trying to determine whether or not workers perform better at a certain task after they've taken a training class what type of test would you use? What is another option for a test you could use and what are the problems with it? Below is a graph of the confidence interval for the mean of some performance parameter for two machines. With only this information can you conclude that the means are different? What other conclusions if any can you make? Below is a probability plot for a particular distribution. Do you think that the distribution is normal? Why or why not?

Explanation / Answer

1. take more samples from which the variance is low. because less variance one is effective.

2. for testing the mean thick ness we use test for difference of means( for small sample t-test and for large sample Z-test)

null hypothesis : there is no difference between mean thickness coating in two different spray coaters

alternative hypothesis : there is a difference between mean thickness coating in two different spray coaters

3. alpha symbol is used to reject the null hypothesis when it is actually true. the values are 0.05,0.01

4. here the performance is dependent on training class. here we take the values in pariwaise so we use paired t-test .

5. based on the given information we are not able to caluculate the difference in the mean performance .

Related Questions

Navigate

• Browse (All)
• Browse Y
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.