Question 435:
1Answer:
No answer provided yet.First, I'm not sure there is a clear distinction between "parametric" and "non-parametric" as the questions suggest. One can use either classical parametric statistics or non-parametric statistics on all sorts of data. In fact this is a hotly debated topic in statistics and there is rarely full agreement. My answers are from the perspective of one who practices statistics, not just teaches them or waxes idealistic about the mathematical properties.
I've included my answers and what I think the exercise is asking for.
- P (although
cycle time is usually not-normal, and some use NP which is overly
cautious). We’d use the t-test or Mann-Whitney test here.
- NP is what your
instructor is looking for, but it is usually just fine to compute P on survey
data, I do it all the time. We’d use Mann-Whitney or the t-test here.
- P (This is a
terrible question: I'm not even sure what the parameter would be here but
this is what will likely get you the correct answer as I believe the
calibration in degrees is our clue)
- P (We can use ANOVA
here and our key is the use of the word mean)
- NP You’d probably use
the CHI-square test if the data were categorized as counts by day of the
week.
- NP (just like #2)
- NP (Using the
Chi-Square again here is probably the way to go).
- P (comparing means)