Re: newbee question about selecting a size of sample
From: Art Kendall (Arthur.Kendall_at_verizon.net)
Date: 03/08/05
- Next message: Herman Rubin: "Re: Independent Variables"
- Previous message: slanchogled: "newbee question about selecting a size of sample"
- In reply to: slanchogled: "newbee question about selecting a size of sample"
- Next in thread: Richard Ulrich: "Re: newbee question about selecting a size of sample"
- Messages sorted by: [ date ] [ thread ]
Date: Tue, 08 Mar 2005 18:47:45 GMT
As asked your questions are impossible to answer meaningfully.
What was the design of your experiment? What factors (variables) did
you manipulate? How many levels of each factor did you have?
When you selected a subset of cases did you intend to gather more data
on those cases? How did you select the 5%? Did you use scientific sampling?
Why do you want to analyze only some of them?
Art
Art@DrKendall.org
Social Research Consultants
University Park, MD USA
(301) 864-5570
slanchogled wrote:
> Hi all, this is probably very stupid question, but i'll ask anyway.
>
> I have 500 results back from an experiment (questionnaire) and want to
> analyse only a portion of them. What is the minimum number I should
> select? And is there any strict way of calculating it? What I've done
> is just selected 5% and said that based on 5% of the feedback the
> results are such and such. Can I generalise?
>
> Also, what if I have 40000 results instead of 500? Does the approach
> change?
>
> Thank you!
- Next message: Herman Rubin: "Re: Independent Variables"
- Previous message: slanchogled: "newbee question about selecting a size of sample"
- In reply to: slanchogled: "newbee question about selecting a size of sample"
- Next in thread: Richard Ulrich: "Re: newbee question about selecting a size of sample"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|