When this posts I should be home from Nepal and mostly recovered. So, back to more details going forward. Take a look, work the problem, solve it, then show your work. Comment with why you chose your response and why you didn’t select one of the others.
- For large populations, the sample size for testing a hypothesis depends on all of the following EXCEPT the
(A) decision risks required (alpha and beta)
(B) population size
(C) size of the smallest difference of interest
(D) variation in characteristic being measured
A good reminder to consider when size matters for hypothesis testing. And, a little math
- A certain electronic component has a constant failure rate of 4 × 10−7/hour. A system requires the use of 64 units of this component, and all of the components must function for the system to work. What is the system failure rate?
(A) 2.56 × 10−5/hour
(B) 3.91 × 10−5/hour
(C) 2.56 × 104hours
(D) 3.91 × 1044hours
I often say one can solve most of the CRE math problems knowing only one formula. If you know my work with NoMTBF.com you know this pains me.
Steve Jones says
Hi Fred, Any chance of the answer to question 2 and how you get there, I’ve been going round in circles all night!
Fred Schenkelberg says
Recall that the reliability of a series system is the product of the reliability values, and when dealing with an exponential distribution (only given failure rate, lambda, here), we can add the failure rates within the R(t) = exp[-lambda t] expression. We have 64 lambdas that are the same, so 64 times the 4 x 10-7 is the system of 64 components failure rate.
I did clean up the presentation of the superscripts (I’ve learned a bit more CSS and HTML since originally writing this article.