Statistics Help. (Pic of nice lady for your trouble)

D

DillingerEscp

Veteran XV
It's some homework I have. Doesn't matter if I get help on it (I'm more honorable than that, goddamnit)

But anyway, heres the question, if anyone could even just point me in the right direction, I'd be greatful.

3 machines fill packages labeled 80 ounces. The 3 filling processes appear to be statistically stable and random samples from each machine are collected:

Machine 1: n=8, mean = 80.2 oz. Standard Deviation = 1.4 oz.

Machine 2: n=8, mean = 81.8 oz. Standard Deviation = 1.8 oz

Machine 3: n=8, mean = 79.9 oz. Standard Deviation = 1.5 oz.

At the 0.05 significance level, test the claim that the 3 machines have the same mean...


I already have the hypothesis down, I got that much Null hypothesis: (mean 1 = mean 2 = mean 3) and the alternative hypothesis (mean 1 /= mean 2 /= mean 3)

Any idea on how to do this? Or at least what test statistic is supposed to be employed?

Appreciate it.




Spoiler
 
i fucking hated statistics.

most worthelss math class i've ever sat through and i've sat through a lot of math...

the only one that comes close is discrete math...
 
To find the bell curve standard deviations of the means you need something that looks like this:

Google Image Result for http://www.1-formula.com/upl/Image/Standart-definition-formula.gif

You have been given the standard deviation scores for the means, and they appear to overlap to .05% significance. I'm not sure what you are asking for besides this. It would be helpful to know the number of samples from the population, as some individual sampling scores will vary, and you need a reasonable sampling to make sure there is little random error. The size of the s.d. seems a little big, an 82 ounce drink varying 2ounces either way. mayny of the samplings fall right into one another's range.

You could put the means on a number line and tic out the s.d.'s to either side and observe the overlap. The first s.d. I think is 34%. The null hypothesis might not be the thing to use here. It's a confidence interval analysis. The ANOVA test will let you compare groups in greater detail.
 
Last edited:
SecretSquirrel said:
To find the bell curve standard deviations of the means you need something that looks like this:

Google Image Result for http://www.1-formula.com/upl/Image/Standart-definition-formula.gif

You have been given the standard deviation scores for the means, and they appear to overlap to .05% significance. I'm not sure what you are asking for besides this. It would be helpful to know the number of samples from the population, as some individual sampling scores will vary, and you need a reasonable sampling to make sure there is little random error. The size of the s.d. seems a little big, an 82 ounce drink varying 2ounces either way. mayny of the samplings fall right into one another's range.

You could put the means on a number line and tic out the s.d.'s to either side and observe the overlap. The first s.d. I think is 34%. The null hypothesis might not be the thing to use here. It's a confidence interval analysis. The ANOVA test will let you compare groups in greater detail.
Click to expand...

I'm fairly certain "testing the claim" implies setting a null and alternative hypothesis.
 
You must log in or register to reply here.
Back
Top