SPSSX Discussion

N on graphs

Classic

List

Threaded

5 messages Options

jason

N on graphs

Been discussing the importance, or lack thereof, of "N" over the past year.
I haven't found SPSS to automatically produce "N" for a graph, and am
thinking that is because its relevance on the display isn't as important as
it might have been in the past. I'm mostly referring to descriptive graphs.

Does anyone have any thoughts on this? Where does "N" fit into your graphs
when using the software, if at all. What reasoning or arguments are there
for or against?

Any help would be great!

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

Re: N on graphs

If one takes the position that there are 3 parts to the expository rhetoric 1) the text 2) table(s) and 3 visualization(s), and that each should "stand alone", then N in each component is often an important part of a complete message. If fewer ways of showing the information are used, then showing N to contextualize the information is even more important.

However, N does not have to be in the body of the graph itself but can be in the titles, or notes.

If you get consensus on this list, perhaps you would want to suggest that patch by emailing [hidden email]

Art Kendall
Social Research Consultants

On 6/10/2011 9:54 AM, Jason wrote:

Been discussing the importance, or lack thereof, of "N" over the past year.
 I haven't found SPSS to automatically produce "N" for a graph, and am
thinking that is because its relevance on the display isn't as important as
it might have been in the past.  I'm mostly referring to descriptive graphs.

Does anyone have any thoughts on this?  Where does "N" fit into your graphs
when using the software, if at all.  What reasoning or arguments are there
for or against?

Any help would be great!

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

Allan Reese (Cefas)

Re: N on graphs

In reply to this post by jason

It's useful to know N, noting that N in this context is the number of points plotted, which may not be the number of points in the dataset or the number of points visible. I would agree the plotting program should report this routinely, whether in the text output or as a component on the plot.

Allan

From: Art Kendall [mailto:[hidden email]]
Sent: 10 June 2011 16:14
Subject: Re: N on graphs

Been discussing the importance, or lack thereof, of "N" over the past year.

 I haven't found SPSS to automatically produce "N" for a graph, and am

thinking that is because its relevance on the display isn't as important as

it might have been in the past.  I'm mostly referring to descriptive graphs.

Does anyone have any thoughts on this?  Where does "N" fit into your graphs

when using the software, if at all.  What reasoning or arguments are there

for or against?

Any help would be great!

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

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

This email and any attachments are intended for the named recipient only. Its unauthorised use, distribution, disclosure, storage or copying is not permitted. If you have received it in error, please destroy all copies and notify the sender. In messages of a non-business nature, the views and opinions expressed are the author's own and do not necessarily reflect those of Cefas. Communications on Cefas’ computer systems may be monitored and/or recorded to secure the effective operation of the system and for other lawful purposes.

ViAnn Beadle

Re: N on graphs

It’s fairly trivial to get N on a graph depending upon what type of graph you have, what the summary stat is and which procedures produce it. Usually this can be automated via a chart template.

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Allan Reese (Cefas)
Sent: Monday, June 13, 2011 2:47 AM
To: [hidden email]
Subject: Re: N on graphs

Allan

From: Art Kendall [mailto:[hidden email]]
Sent: 10 June 2011 16:14
Subject: Re: N on graphs

Been discussing the importance, or lack thereof, of "N" over the past year.

 I haven't found SPSS to automatically produce "N" for a graph, and am

thinking that is because its relevance on the display isn't as important as

it might have been in the past.  I'm mostly referring to descriptive graphs.

Does anyone have any thoughts on this?  Where does "N" fit into your graphs

when using the software, if at all.  What reasoning or arguments are there

for or against?

Any help would be great!

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

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

Margaret MacDougall

Re: N on graphs

In reply to this post by jason

Dear Jason

This is a welcome query. The construction of boxplots is a case in point. One should not assume that PASW (or SPSS) users have the necessary background to appreciate the importance of sample size in explaining differences in variation across categories. I certainly find this with medical students. Leaving out group sizes can lead to misleading responses (for example in the interpretation of why pain scores vary more in one category of patients than another). While, if one takes the 'interactive' route within the menu for constructing boxplots in PASW, it is possible to find a relevant option 'Display count labels', the outcome is far from satisfactory. The counts land immediately beside the median line, which provides the basis for confusion, particularly if the chart appears in a report without explanation of what the numbers stand for. It would be great if these issues could be addressed in future versions of PASW. To complement this, it would be better still if users could hover over options and read brief descriptors of their importance and relevance.

Every best wish

Margaret

From: Jason <[hidden email]>
To: [hidden email]
Sent: Friday, 10 June, 2011 14:54:43
Subject: N on graphs

Been discussing the importance, or lack thereof, of "N" over the past year.
I haven't found SPSS to automatically produce "N" for a graph, and am
thinking that is because its relevance on the display isn't as important as
it might have been in the past. I'm mostly referring to descriptive graphs.

Does anyone have any thoughts on this? Where does "N" fit into your graphs
when using the software, if at all. What reasoning or arguments are there
for or against?

Any help would be great!

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