> total sample population is 10000
Please clarify. are you saying that you drew a sample of 300 cases from
a population of 10,000 cases?
or Are you saying you drew a sample of 10,000 cases from a larger
population and then drew as subsample of 300cases from the 10,000cases?
You appear to be standing the decision making process on its head.
Typically I recommend sticking with .05 as alpha and .80 as beta.
Then you would ask what size effect would make a difference in
theorizing, practice, or policy.
Then use statistical power software to see what sample size would make a
difference.
Please describe what your population is. What data do you have on the
whole population?
On the 10,000 cases?
What questions are you using the data to answer?
What is the substantive meaning of your dependent value?
Art Kendall
Social Research Consultants
On 1/17/2013 10:31 PM, k.h.naik wrote:
> dear all,i am going to use multinomial logistic regression with 360 samples.
> is it enough to decide the significance of the model, confidence interval
> and confidence level?
> total sample population is 10000. please help me
>
>
>
> --
> View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/sample-size-for-multinomial-logistic-regression-tp5717491.html> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
>
[hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Art Kendall
Social Research Consultants
Locked
