Re: SPSS Proactive Notification - SPSS Statistics 18.0.1 Patch Availability Notice

classic Classic list List threaded Threaded
12 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: SPSS Proactive Notification - SPSS Statistics 18.0.1 Patch Availability Notice

Kornbrot, Diana
Re: SPSS Proactive Notification - SPSS Statistics 18.0.1 Patch Availability Notice Why on earth can’t you give the DIRECT address?
Not 3 clciks away with the links scattered at random

PASW support gets worse by the day
diana


On 13/12/2009 09:09, "SPSS Support" <proactive@...> wrote:

The latest SPSS Statistics patch is now available for download from the SPSS Support Website.  This patch will upgrade your existing SPSS Statistics 18.0.0 to version 18.0.1.  You MUST have PASW Statistics 18.0.0 installed prior to applying this patch.

Note: The 18.0.1 client patches apply to Grad Pack & Student Versions also.

For detailed information on the description of this patch, installation instructions, and downloading information, please browse to: http://support.spss.com and log in to the web site. After entering your login information, select the "Statistics" menu link then click on "Patches".
IDIOCY give the address!!!!

You will need your Support Web site User Id and Password to access this information. If you have forgotten your user id or password you can use the Find Password feature on our Web site. This can be found at: http://support.spss.com/password.html

We are continually striving to improve our communication to our clients. However, if you do not wish to receive further proactive e-mails from SPSS Support, please reply to this message with REMOVE on the subject line.

Thank you,
SPSS, an IBM Company, Technical Support



Professor Diana Kornbrot
email: 
d.e.kornbrot@...    
web:    http://web.me.com/kornbrot/KornbrotHome.html
Work
School of Psychology
 University of Hertfordshire
 College Lane, Hatfield, Hertfordshire AL10 9AB, UK
 voice:   +44 (0) 170 728 4626
   fax:     +44 (0) 170 728 5073
Home
 
19 Elmhurst Avenue
 London N2 0LT, UK
    voice:   +44 (0) 208 883  3657
    mobile: +44 (0)
796 890 2102
   fax:      +44 (0) 870 706 4997





Reply | Threaded
Open this post in threaded view
|

Re: SPSS Proactive Notification - SPSS Statistics 18.0.1 Patch Availability Notice

Bruce Weaver
Administrator
kornbrot wrote
Why on earth can¹t you give the DIRECT address?
Good question.  I know that a fair number of my colleagues would just not bother installing patches, given the number of steps they have to go through to get them.  So when I see there is a patch available, I ask our IT folks to make it available via a direct local link.  But this should not be necessary.

--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.
Reply | Threaded
Open this post in threaded view
|

Re: SPSS Proactive Notification - SPSS Statistics 18.0.1 Patch Availability Notice

Michael Kruger
In reply to this post by Kornbrot, Diana
It would be nice if someone from IBM(SPSS) would reply to some of the
issues and concerns people in the listserv are voicing. I tried to apply
the patch to an 18.0 install on my laptop that has an AMD Dual-Core CPU
(Turon, I believe) and received a message that the 'install was not
supported on this machine's processor' or something to that effect. I
submitted a 'case' to SPSS support over a week ago and have never
received a reply. I realize that there will likely be changes with the
IBM acquisition. But I think IBM should realize that a lot of people use
SPSS not only in their work but in teaching and subsequently the
students in those classes become SPSS users and will help the bottom
line for IBM. As people begin to adopt alternative software packages
(SAS, S-Plus, STATA, R,etc), the company loses potential revenue. These
annual upgrades, often creating more time consuming installation
problems relative to the obtuse improvements in features make the
product less and less attractive to me as a user and educator.


Michael Kruger
Statistical Analayst & Instructor
Dept. of OB/GYN
Wayne State University School of Medicine

=====================
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
Reply | Threaded
Open this post in threaded view
|

SPSS Statistics 18.0.1 Patch--Not that difficult

Mark A Davenport MADAVENP
In reply to this post by Kornbrot, Diana

I have been patching SPSS products for years.  I just downloaded the 18.0.1 and I don't see any difference in the process post-IBM.  For years I have had to log into the support site, go to 'Patches' on the main menu (left of screen) and click on patch downloads.  Honestly, is that any more cumbersome than patching any other piece of software (don't get me started about Microsoft)?  I will grant that it would be helpful to embed or attach screen captures showing the location of the Main Menu and the 'Patches' hotlink.  Five minutes with 'Snag-It' and Adobe Acrobat is all such a document would require.


***************************************************************************************************************************************************************
Mark A. Davenport Ph.D.
Senior Research Analyst
Office of Institutional Research
The University of North Carolina at Greensboro
336.256.0395
[hidden email]

'An approximate answer to the right question is worth a good deal more than an exact answer to an approximate question.' --a paraphrase of J. W. Tukey (1962)





kornbrot <[hidden email]>
Sent by: "SPSSX(r) Discussion" <[hidden email]>

12/18/2009 07:11 AM

Please respond to
kornbrot <[hidden email]>

To
[hidden email]
cc
Subject
Re: SPSS Proactive Notification - SPSS Statistics 18.0.1 Patch              Availability Notice





Why on earth can’t you give the DIRECT address?
Not 3 clciks away with the links scattered at random

PASW support gets worse by the day
diana


On 13/12/2009 09:09, "SPSS Support" <
proactive@...> wrote:

The latest SPSS Statistics patch is now available for download from the SPSS Support Website.  This patch will upgrade your existing SPSS Statistics 18.0.0 to version 18.0.1.  You MUST have PASW Statistics 18.0.0 installed prior to applying this patch.

Note: The 18.0.1 client patches apply to Grad Pack & Student Versions also.

For detailed information on the description of this patch, installation instructions, and downloading information, please browse to:
http://support.spss.com and log in to the web site. After entering your login information, select the "Statistics" menu link then click on "Patches".
IDIOCY give the address!!!!


You will need your Support Web site User Id and Password to access this information. If you have forgotten your user id or password you can use the Find Password feature on our Web site. This can be found at:
http://support.spss.com/password.html

We are continually striving to improve our communication to our clients. However, if you do not wish to receive further proactive e-mails from SPSS Support, please reply to this message with REMOVE on the subject line.

Thank you,
SPSS, an IBM Company, Technical Support




Professor Diana Kornbrot
email: 
d.e.kornbrot@...    
web:    
http://web.me.com/kornbrot/KornbrotHome.html
Work

School of Psychology
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
voice:   +44 (0) 170 728 4626
  fax:     +44 (0) 170 728 5073
Home
19 Elmhurst Avenue
London N2 0LT, UK
   voice:   +44 (0) 208 883  3657
   mobile: +44 (0) 796 890 2102
  fax:      +44 (0) 870 706 4997





Reply | Threaded
Open this post in threaded view
|

Re: SPSS Proactive Notification - SPSS Statistics 18.0.1 Patch Availability Notice

Salbod
In reply to this post by Bruce Weaver
I agree. A direct link should be sent with the notification. It sounds like they are afraid to have redundancy on their site.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bruce Weaver
Sent: Friday, December 18, 2009 7:27 AM
To: [hidden email]
Subject: Re: SPSS Proactive Notification - SPSS Statistics 18.0.1 Patch Availability Notice

kornbrot wrote:
>
> Why on earth can¹t you give the DIRECT address?
>

Good question.  I know that a fair number of my colleagues would just not
bother installing patches, given the number of steps they have to go through
to get them.  So when I see there is a patch available, I ask our IT folks
to make it available via a direct local link.  But this should not be
necessary.



-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."

NOTE:  My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.
--
View this message in context: http://old.nabble.com/Re%3A-SPSS-Proactive-Notification---SPSS-Statistics-18.0.1-Patch-Availability-Notice-tp26842544p26842667.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
Reply | Threaded
Open this post in threaded view
|

Re: SPSS Statistics 18.0.1 Patch--Not that difficult

Art Kendall-3
In reply to this post by Mark A Davenport MADAVENP
<html><head>

<link media="all" type="text/css" href="/webmail/static/deg/css/wysiwyg-3933289048.css" rel="stylesheet">
</head><body>
If SPSS does put direct links into email, users should put their cursor over the link (not click it) and check that the link is in fact to SPSS.&nbsp; Unfortunately, embedding links that go to a bogus address is a common phishing technique. <br><br>Art Kendall<br>Social Research Consultants<br><br><br><br>Dec 18, 2009 08:49:26 AM, [hidden email] wrote:<br><blockquote style="border-left: 3px solid rgb(102, 153, 204);">
<br><font face="sans-serif" size="2">I have been patching SPSS products for
years. &nbsp;I just downloaded the 18.0.1 and I don't see any difference
in the process post-IBM. &nbsp;For years I have had to log into the support
site, go to 'Patches' on the main menu (left of screen) and click on patch
downloads. &nbsp;Honestly, is that any more cumbersome than patching any
other piece of software (don't get me started about Microsoft)? &nbsp;I
will grant that it would be helpful to embed or attach screen captures
showing the location of the Main Menu and the 'Patches' hotlink. &nbsp;Five
minutes with 'Snag-It' and Adobe Acrobat is all such a document would require.</font>
<br>
<br>
<br><font face="sans-serif" size="2">***************************************************************************************************************************************************************<br>
Mark A. Davenport Ph.D.<br>
Senior Research Analyst<br>
Office of Institutional Research<br>
The University of North Carolina at Greensboro<br>
336.256.0395<br>
[hidden email]<br>
<br>
'An approximate answer to the right question is worth a good deal more
than an exact answer to an approximate question.' --a paraphrase of J.
W. Tukey (1962)<br>
<br>
<br>
</font>
<br>
<br>
<br>
<table width="100%">
<tbody><tr valign="top">
<td width="40%"><font face="sans-serif" size="1"><b>kornbrot <[hidden email]></[hidden email]></b>
</font>
<br><font face="sans-serif" size="1">Sent by: "SPSSX(r) Discussion"
<[hidden email]></[hidden email]></font>
<p><font face="sans-serif" size="1">12/18/2009 07:11 AM</font>
<table border="1">
<tbody><tr valign="top">
<td bgcolor="white">
<div align="center"><font face="sans-serif" size="1">Please respond to<br>
kornbrot <[hidden email]></[hidden email]></font></div></td></tr></tbody></table>
<br>
</p></td><td width="59%">
<table width="100%">
<tbody><tr valign="top">
<td>
<div align="right"><font face="sans-serif" size="1">To</font></div>
</td><td><font face="sans-serif" size="1">[hidden email]</font>
</td></tr><tr valign="top">
<td>
<div align="right"><font face="sans-serif" size="1">cc</font></div>
</td><td>
</td></tr><tr valign="top">
<td>
<div align="right"><font face="sans-serif" size="1">Subject</font></div>
</td><td><font face="sans-serif" size="1">Re: SPSS Proactive Notification - SPSS
Statistics 18.0.1 Patch &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;Availability
Notice</font></td></tr></tbody></table>
<br>
<table>
<tbody><tr valign="top">
<td>
</td><td></td></tr></tbody></table>
<br></td></tr></tbody></table>
<br>
<br>
<br><font face="Arial" size="3">Why on earth can’t you give the DIRECT address?<br>
Not 3 clciks away with the links scattered at random<br>
<br>
PASW support gets worse by the day<br>
diana<br>
<br>
<br>
On 13/12/2009 09:09, "SPSS Support" &lt;</font><a href="http://proactive@..." target="_blank"><font color="blue" face="Arial" size="3"><u>[hidden email]</u></font></a><font face="Arial" size="3">&gt;
wrote:<br>
</font>
<br><font face="Courier New" size="3">The latest SPSS Statistics patch is
now available for download from the SPSS Support Website. &nbsp;This patch
will upgrade your existing SPSS Statistics 18.0.0 to version 18.0.1. &nbsp;You
MUST have PASW Statistics 18.0.0 installed prior to applying this patch.<br>
<br>
Note: The 18.0.1 client patches apply to Grad Pack &amp; Student Versions
also.<br>
<br>
For detailed information on the description of this patch, installation
instructions, and downloading information, please browse to: </font><a href="http://support.spss.com/" target="_blank"><font color="blue" face="Courier New" size="3"><u>http://support.spss.com</u></font></a><font face="Courier New" size="3">
and log in to the web site. After entering your login information, select
the "Statistics" menu link then click on "Patches".</font><font color="red" face="Courier New" size="3"><br>
IDIOCY give the address!!!!</font><font face="Courier New" size="3"><br>
<br>
You will need your Support Web site User Id and Password to access this
information. If you have forgotten your user id or password you can use
the Find Password feature on our Web site. This can be found at: </font><a href="http://support.spss.com/password.html" target="_blank"><font color="blue" face="Courier New" size="3"><u>http://support.spss.com/password.html</u></font></a><font face="Courier New" size="3"><br>
<br>
We are continually striving to improve our communication to our clients.
However, if you do not wish to receive further proactive e-mails from SPSS
Support, please reply to this message with REMOVE on the subject line.<br>
<br>
Thank you,<br>
SPSS, an IBM Company, Technical Support</font>
<br><font face="Verdana" size="1"><b><br>
<br>
</b></font>
<hr><font face="Verdana" size="1"><b>Professor Diana Kornbrot</b><br>
email:&nbsp; </font><a href="http://d.e.kornbrot@..." target="_blank"><font color="blue" face="Times New Roman" size="2"><u>[hidden email]</u></font></a><font face="Times New Roman" size="2">
&nbsp; &nbsp;</font><font face="Verdana" size="1"><br>
web: &nbsp; &nbsp;</font><a href="http://web.me.com/kornbrot/KornbrotHome.html" target="_blank"><font color="blue" face="Times New Roman" size="2"><u>http://web.me.com/kornbrot/KornbrotHome.html</u></font></a><font face="Verdana" size="1"><b><br>
Work </b><br>
School of Psychology<br>
 University of Hertfordshire<br>
 College Lane, Hatfield, Hertfordshire AL10 9AB, UK <br>
 voice: &nbsp; +44 (0) 170 728 4626<br>
 &nbsp; fax: &nbsp; &nbsp; +44 (0) 170 728 5073<b><br>
Home<br>
 </b>19 Elmhurst Avenue<br>
 London N2 0LT, UK<br>
 &nbsp; &nbsp;voice: &nbsp; +44 (0) 208 883 &nbsp;3657<br>
 &nbsp; &nbsp;mobile: +44 (0) 796 890 2102<br>
 &nbsp; fax: &nbsp; &nbsp; &nbsp;+44 (0) 870 706 4997<br>
<br>
<br>
</font><font face="Verdana" size="2"><br>
</font><font face="Arial" size="3"><br>
</font>
<br></blockquote></body></html>

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: SPSS Proactive Notification - SPSS Statistics 18.0.1 Patch Availability Notice

SPSS Support-2
In reply to this post by Kornbrot, Diana

Hi Michael,

I believe this is the first note we've seen come across the list reporting a problem installing the patch. I checked our case tracking system and there was no case in the system for you regarding this issue, so I've opened a case. We'll need to get specifics on the error you're getting and your machine specifications to track down what might be happening here.

David Nichols
SPSS, an IBM Company

From: Michael Kruger <[hidden email]>
To: [hidden email]
Date: 12/18/2009 06:32 AM
Subject: Re: SPSS Proactive Notification - SPSS Statistics 18.0.1 Patch              Availability Notice
Sent by: "SPSSX(r) Discussion" <[hidden email]>





It would be nice if someone from IBM(SPSS) would reply to some of the
issues and concerns people in the listserv are voicing. I tried to apply
the patch to an 18.0 install on my laptop that has an AMD Dual-Core CPU
(Turon, I believe) and received a message that the 'install was not
supported on this machine's processor' or something to that effect. I
submitted a 'case' to SPSS support over a week ago and have never
received a reply. I realize that there will likely be changes with the
IBM acquisition. But I think IBM should realize that a lot of people use
SPSS not only in their work but in teaching and subsequently the
students in those classes become SPSS users and will help the bottom
line for IBM. As people begin to adopt alternative software packages
(SAS, S-Plus, STATA, R,etc), the company loses potential revenue. These
annual upgrades, often creating more time consuming installation
problems relative to the obtuse improvements in features make the
product less and less attractive to me as a user and educator.


Michael Kruger
Statistical Analayst & Instructor
Dept. of OB/GYN
Wayne State University School of Medicine

=====================
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


Reply | Threaded
Open this post in threaded view
|

Patehces and updates

Kornbrot, Diana
In reply to this post by Salbod
Patehces and updates The easiest way to stay up to date is to check for updates each time you launch pasw/spss

Help > check for updates

Much more efficient than going through links in the update email, and less prone to scams
The email, intended to be helpful, simply led me to frrustration and away from normal good practice – check for updates regularly, and always if you meet a bug
People who say they haven’t noticed  a decline since ibm, are probably right – it was ever thus

Happy holidays

Best

Diana

Reply | Threaded
Open this post in threaded view
|

ANOVA or Correlation

Humphrey Paulie
In reply to this post by SPSS Support-2
Dear all,
I'm investigating the relationship between EQ and job perfomance or success..
I want to correlate scores on the EQ test and job performnace ratings.
A colleagues says it's better to divide the subjects into three groups of low EQ, mid EQ and high EQ and then run one-way ANOVA and compare the three means on the job performance ratings.
 What is the difference between the two approaches?
 Which one is better?
Regards
Humphrey




Reply | Threaded
Open this post in threaded view
|

Re: ANOVA or Correlation

Mike
Off the top of my head, consider the following points:
 
(1)  Whether you use a Pearson r between EQ and Job Performance (JP)
or ANOVA is dependent upon whether the correlation between EQ
and JP is linear.  A scatterplot of JP against EQ will give you an idea
of the degree to which the relationship is linear.
 
(2)  If the correlation between EQ and JP is strongly linear, then the
conclusions you reach will be pretty much the same regardless of
whether you use a  correlational or ANOVA analysis.  However,
with ANOVA, if significant you would still have to do post hoc
test to determine whether all 3 means are different.
 
(3) If the correlation between EQ and JP is weakly linear or is nonlinear,
then the Pearson is likely to be nonsignificant.  This is why examining
the scatterplot is so important.  Imagine the JP is low for both low EQ
and high EQ but high for Moderate EQ (i.e., an upside U function).
The Pearson r will be close to zero because it is only measure linear
relationships not curvilinear relationship.  The ANOVA could show
a significant differences among means and post hoc testing may show
that low EQ and high EQ groups differ significantly from the moderate EQ
group on JP.  If the relationship is weakly linear, it is still possible for
the means to be different but not in the pattern shownin (2) above.
 
I hope that I was clear.
 
Mike Palij
New York University
 
----- Original Message -----
Sent: Saturday, December 19, 2009 5:33 AM
Subject: ANOVA or Correlation

Dear all,
I'm investigating the relationship between EQ and job perfomance or success..
I want to correlate scores on the EQ test and job performnace ratings.
A colleagues says it's better to divide the subjects into three groups of low EQ, mid EQ and high EQ and then run one-way ANOVA and compare the three means on the job performance ratings.
 What is the difference between the two approaches?
 Which one is better?
Regards
Humphrey




Reply | Threaded
Open this post in threaded view
|

Re: ANOVA or Correlation

Art Kendall
very good advice. 

I would just like to suggest that you should impose a linear and a loess fit on the scatterplot.
You can do this by editing the output file.

Art Kendall
Social Research Consultants

Mike Palij wrote:
Off the top of my head, consider the following points:
 
(1)  Whether you use a Pearson r between EQ and Job Performance (JP)
or ANOVA is dependent upon whether the correlation between EQ
and JP is linear.  A scatterplot of JP against EQ will give you an idea
of the degree to which the relationship is linear.
 
(2)  If the correlation between EQ and JP is strongly linear, then the
conclusions you reach will be pretty much the same regardless of
whether you use a  correlational or ANOVA analysis.  However,
with ANOVA, if significant you would still have to do post hoc
test to determine whether all 3 means are different.
 
(3) If the correlation between EQ and JP is weakly linear or is nonlinear,
then the Pearson is likely to be nonsignificant.  This is why examining
the scatterplot is so important.  Imagine the JP is low for both low EQ
and high EQ but high for Moderate EQ (i.e., an upside U function).
The Pearson r will be close to zero because it is only measure linear
relationships not curvilinear relationship.  The ANOVA could show
a significant differences among means and post hoc testing may show
that low EQ and high EQ groups differ significantly from the moderate EQ
group on JP.  If the relationship is weakly linear, it is still possible for
the means to be different but not in the pattern shownin (2) above.
 
I hope that I was clear.
 
Mike Palij
New York University
 
----- Original Message -----
Sent: Saturday, December 19, 2009 5:33 AM
Subject: ANOVA or Correlation

Dear all,
I'm investigating the relationship between EQ and job perfomance or success..
I want to correlate scores on the EQ test and job performnace ratings.
A colleagues says it's better to divide the subjects into three groups of low EQ, mid EQ and high EQ and then run one-way ANOVA and compare the three means on the job performance ratings.
 What is the difference between the two approaches?
 Which one is better?
Regards
Humphrey




===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: ANOVA or Correlation

Bruce Weaver
Administrator
In reply to this post by Humphrey Paulie
Humphrey-6 wrote
Dear all,

I'm investigating the relationship between EQ and job perfomance or success..

I want to correlate scores on the EQ test and job performnace ratings.

A colleagues says it's better to divide the subjects into three groups
of low EQ, mid EQ and high EQ and then run one-way ANOVA and compare
the three means on the job performance ratings.

 What is the difference between the two approaches?

 Which one is better?

Regards

Humphrey
Carving continuous variables into categories unnecessarily is usually a bad idea.  (See David Streiner's article "Breaking Up is Hard to Do: The Heartbreak of Dichotomizing Continuous Data" -- you can probably find it via Google Scholar.)  As others have suggested, start by plotting the data.  If the functional relationship is linear, run a linear regression with EQ as a continuous explanatory variable.  

If the functional relationship is not linear, you still have plenty of options that don't require you to carve EQ into groups.  E.g., sometimes a quadratic function (or some other polynomial function) provides a good fit.  A quadratic model would include both EQ and EQ-squared as predictors.  Another possibility is to transform the variables.  I think Mosteller & Tukey discuss this approach in their 1977 book on regression.  Another book you could look at is Aiken & West's book on interactions in linear regression models.  IIRC, it also discusses fitting polynomial functions.  Finally, regression splines are another possibility, but I don't have a lot of experience with them, so can't say a lot more about it.

I'll say just a bit more about why carving EQ into 3 groups is not a great idea.  If you carve EQ into 3 groups, the fitted (or predicted) value of Y will be the mean of the group a subject belongs to.  So two subjects who fall just on either side of the cut-point between two groups could have VERY different fitted values, despite having nearly identical values of EQ.  Likewise, take a subject who falls just above the lower cut-point for a group, and another who falls just the upper cut-point for the same group.  They will have exactly the same fitted value, despite having very different values of EQ.  Is this really what you want to do?  Why would you throw away the information contained in the continuous variable like that?  

HTH.
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.