Any statistics experts here or may be professors: Please help: I'm sick with my quiz - TennisForum.com
Old Sep 22nd, 2003, 03:48 PM Thread Starter
Senior Member

Join Date: Feb 2002
Location: Bangkok
Posts: 719

In my quiz,
this is the question:
The maufacturer claims that the variance in working hours of workers is less than 0.9
And from a random sample size 5, we find that the variance of the sample is 1.1
Test the hypothesis at .05 level of significance.

I really would like to know how we should set up the hypothesis.
Choice 1
Null Hypothesis: variance = 0.9
Alternative Hypothesis : variance > 0.9
Choice 2
Null Hypothesis: variance = 0.9
Alternative Hypothesis : variance < 0.9

In some Thai texts, they just say that because the manufacturer's claim(less than 0.9) does not consist of equal so it can't be the null hypothesis(I mean they just want it to be "does not exceed" more than "less than" if ัyou want to set it to be the null hypothesis.
But I don't think like that I think the 0.9 is just a very small point in a continuous variable system, and we should give the credit to the manufacturer first and set it to be null hypothesis and then try to find enough evidence to reject it

Which one is the real issue? Just the "does not exceed" or " less than"
In the real world we should be able to say less than and it has all rights to be the null hypothesis...

Thai Players, Masha, Lindsay, Ai
Al is offline

Old Sep 22nd, 2003, 05:23 PM
Senior Member

Join Date: May 2003
Location: Hingisville
Posts: 2,541

QUOTE=lakeway11]in the vast majority of cases the researcher sets up the null hypothesis as the one where no significance difference is found thus rejecting the null is what usually what one wants...since the manufacter claims that the variance is less than 0.9 (research hypothesis) it would be set up in this manner:

Ho: the variance of working hours is not less than 0.9
Ha: the variance of working hours < 0.9[/QUOTE]

Mao Asada & Yu-Na Kim figure skating

"Power always thinks...that it is doing God's service when it is violating all his laws." –-John Adams

"Progress should mean that we are always changing the world to fit the vision, instead we are always changing the vision." - Orthodoxy, 1908

"Progress is a comparative of which we have not settled the superlative." - Chapter 2, Heretics, 1905

Last edited by lakeway11; Sep 23rd, 2003 at 02:34 AM.
lakeway11 is offline
Old Sep 23rd, 2003, 08:13 AM Thread Starter
Senior Member

Join Date: Feb 2002
Location: Bangkok
Posts: 719

Quote:
Originally Posted by lakeway11
QUOTE=lakeway11]in the vast majority of cases the researcher sets up the null hypothesis as the one where no significance difference is found thus rejecting the null is what usually what one wants...since the manufacter claims that the variance is less than 0.9 (research hypothesis) it would be set up in this manner:

Ho: the variance of working hours is not less than 0.9
Ha: the variance of working hours < 0.9
[/QUOTE]

Thanks
But the question is we as a person outside the manufacturer's company not we as a manufacturer.
The quiz ask we think whether the manufacturer's claim is right.
Please make it clear for me again.

Thai Players, Masha, Lindsay, Ai
Al is offline

Message:
Options

## Register Now

In order to be able to post messages on the TennisForum.com forums, you must first register.

User Name:
OR

## Log-in

Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.