spss 15 Data view...

[George Luca]

Hello,

I am a bit confused with a dataset i have, mainly because though i
cannot see ANY Data for a few variables (in dataview mode) in the output
  window I get the needed results, as though data would exist (which
obviously does but i for one reason cannot see it).

Thanks in advance,
George

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

[ajay atluri]

Hi George,

May be your data is not catching up properly.
You may go to    *"File"*   menu and you will get an option as *"Cache
Data..."  *then you will be able to see the data as well in your data
window.

Hope this works for you.


Thanks and Regards,

*Ajay Kumar Atluri l Associate Research Manager*

*market**Rx** India, **a Cognizant Company***

*t: *[phone] extn: 179*  m:*[phone]





On Thu, Mar 27, 2008 at 10:15 PM, George Luca <
[hidden email]> wrote:

> Hello,
>
> I am a bit confused with a dataset i have, mainly because though i
> cannot see ANY Data for a few variables (in dataview mode) in the output
>  window I get the needed results, as though data would exist (which
> obviously does but i for one reason cannot see it).
>
> Thanks in advance,
> George
>
> =====================
> 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

[Nana Nadine]

Hi Guys,
  I have collected some attributte rating data in 7 points scale. The argument around that was that this type of rating scale allow to have a middle. What ar the risk of converting that analysis in 10 points scales. For me it make more sense to report a 7.0 out of 10 mean that reporting a 4.9 out of 7 mean.
  What do you guys thing is more meaningfull?
  Any refrence to back up this would be great.
  Thank you in advance,
  Nadine Nana



---------------------------------
Never miss a thing.   Make Yahoo your homepage.

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

[Richard Ristow]

At 08:59 AM 3/28/2008, Nana Nadine wrote:

>   I have collected some attributte rating data in 7 points scale.
> The argument around that was that this type of rating scale allow
> to have a middle. What ar the risk of converting that analysis in
> 10 points scales. For me it make more sense to report a 7.0 out of
> 10 mean that reporting a 4.9 out of 7 mean.

*  So you want to compute
COMPUTE NewScale = OldScale*10/7 /* like this? */.

I don't suppose there's a lot of risk. Most statistical analyses are
scale-independent.

But I think it's more understandable if you don't transform. Mean 4.9
is on the same scale as the scale values, and will be understood by
your readers as such. The transformed '7/10' value is comparatively
free-floating.

You may hear arguments whether the mean is an appropriate statistic
at all. The rough-and-ready answer is that, technically, it isn't,
but for many real scales, it works plenty well enough.

HOWEVER, your scale is probably not a true interval variable, and
transforming it will encourage a perception that it's closer to
interval than it is.

I wouldn't do it.

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

[Mark.W.Andrews-2]

Nadine,

Some of my purest survey methodology friends would have a heart attack
if I suggested such a thing. They argue that scale data is scale data.
You should not even consider reporting a mean.

I wouldn't go that far, but they have a point. You make a big assumption
about the data you are looking at when you suggest that it can be easily
converted. The assumption is that the difference between a 1 and a 2 is
the same as the difference between a 2 and a 3 and so forth.
Unfortunately, there have countless papers that have concluded that this
is simply not the case.

With that said, a problem with 7 point scales is that they are difficult
to report. If you define it well in your report, you can report top 2
and/or bottom 2 box frequencies. If your data is fairly consistent you
can give them names (e.g. satisfied and dissatisfied). If you have to
have 1 figure, you can report % top 2 boxes/ % bottom 2 boxes. This is a
great measure...but good luck getting your client to understand it.

Mark



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Nana Nadine
Sent: Friday, March 28, 2008 9:59 AM
To: [hidden email]
Subject: Reporting Scales

Hi Guys,
  I have collected some attributte rating data in 7 points scale. The
argument around that was that this type of rating scale allow to have a
middle. What ar the risk of converting that analysis in 10 points
scales. For me it make more sense to report a 7.0 out of 10 mean that
reporting a 4.9 out of 7 mean.
  What do you guys thing is more meaningfull?
  Any refrence to back up this would be great.
  Thank you in advance,
  Nadine Nana



---------------------------------
Never miss a thing.   Make Yahoo your homepage.

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