Date:         Tue, 8 Apr 2008 08:13:04 -0400
Reply-To:     Ellen <statistiekfreak@HOTMAIL.COM>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Ellen <statistiekfreak@HOTMAIL.COM>
Subject:      significant?

Hi all,

My knowledge of statistics is very bad ...

I have a few countries, I know their participation rates in lifelong
learning and I also know the percentages for educational attainment of
those participants.

E.G. :

Country  -  participation %  - Low %  - Middle % - High %
E.g. in the UK are 39.8 % participants, 12.2 % participants with low level
of education, 39.9 % of participants with middle level of education and
60.8 % of participants with high level of education

I did an ANALYZE --> REGRESSION --> LINEAR REGRESSION

Participation is dependent.
Low, Middle and High are independent.

I get the following result :
-1.065 - .056*LOW + .873*MIDDLE + .152*HIGH
The t-test says that only MIDDLE is significant.

I want to know if educational attainment is a significant determinant of
participation, but I have the feeling that I am doing very wrong to know
this.
Can anyone please help me and tell me how I have to check this?
I also have other variables like age, sex and work status.  I want to know
which variables matter most in participation.
If I get a new database of people with information about these socio-
demographic variables.  I want to make a predication if they would be
participants or nonparticipants.

Looking forward to your responses.

I hope to learn from you a lot!

All the best,
Ellen

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@LISTSERV.UGA.EDU (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