Frequency and Graph Options

[<B.Y.>]

Hi there,

I am trying to run simple frequencies and graphs on a few variables but am
having a small issue. Lets say my variable has 4 options (it was a
survey), 1 = Workgroup#1 2 = Workgroup#2 and so on... When I try to graph
and run frequencies on this variable (lets call it WorkGroup), the
workgroup with no respondents obviously doesn't appear.

My question is that can you somehow change that and make it appear, even
if there is no one from that workgroup who responded on the survey? I need
to make it consistent with how the item actually looked. So for example,
one sample might have 0 in workgroup 1, 10 in workgroup 2, 7 in 3, 5 in 4,
and another sample might have 0 in workgroup 3 but have all other
workgroups represented... how can I make the graphs look the same for both?

I'm using PASW 17 and I have the value labels defined and it still isn't
working.

Thanks so much!

Boris

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

[ViAnn Beadle]

What commands do you use to create your graphs?

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
<B.Y.>
Sent: Thursday, September 24, 2009 6:36 AM
To: [hidden email]
Subject: Frequency and Graph Options

Hi there,

I am trying to run simple frequencies and graphs on a few variables but am
having a small issue. Lets say my variable has 4 options (it was a
survey), 1 = Workgroup#1 2 = Workgroup#2 and so on... When I try to graph
and run frequencies on this variable (lets call it WorkGroup), the
workgroup with no respondents obviously doesn't appear.

My question is that can you somehow change that and make it appear, even
if there is no one from that workgroup who responded on the survey? I need
to make it consistent with how the item actually looked. So for example,
one sample might have 0 in workgroup 1, 10 in workgroup 2, 7 in 3, 5 in 4,
and another sample might have 0 in workgroup 3 but have all other
workgroups represented... how can I make the graphs look the same for both?

I'm using PASW 17 and I have the value labels defined and it still isn't
working.

Thanks so much!

Boris

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

[John F Hall]

Not sure quite what you want, but I think you can force cells to display 0 counts. 

 

I encountered a similar problem when writing a tutorial for SPSS 11 using a short attitude scale to measure negative attitudes to women ("Sexism") among teenagers.   Scores for boys ranged from 1 to 7 and for girls from 0 to 6 on the same scale.  In separate frequency counts and histograms for boys and girls. blank cells weren't displayed. 

 

The full tutorial (with comments on the logic of analysis and SPSS displays) is on Derived variables: COUNT but this is quite old and is about to be replaced by a new one using SPSS 15.  The problem persists, but the following will show you what I mean.

 

My note in the text says:

 

Note also that SPSS has not included, either in the tables or the charts, value 0 for the boys or value 7 for the girls.  This can be a bit confusing, but there are no ways round it that the author can find, other than to produce a crosstab of sex by sexism or histogram with boys and girls displayed separately on the same chart. 

 

Boys:

 

Q.33 Sexism [count]

		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	1	5	10.2	10.2	10.2
	2	12	24.5	24.5	34.7
	3	9	18.4	18.4	53.1
	4	2	4.1	4.1	57.1
	5	8	16.3	16.3	73.5
	6	8	16.3	16.3	89.8
	7	5	10.2	10.2	100.0
	Total	49	100.0	100.0

 

 

 

Girls:

 

Q.33 Sexism [count]

		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	0	5	10.9	10.9	10.9
	1	20	43.5	43.5	54.3
	2	9	19.6	19.6	73.9
	3	5	10.9	10.9	84.8
	4	5	10.9	10.9	95.7
	5	1	2.2	2.2	97.8
	6	1	2.2	2.2	100.0
	Total	46	100.0	100.0

The same happens with the separate charts.

A later note says:

You can also display boys and girls together on the same chart<A title="" style="mso-footnote-id: ftn1" href="mhtml:mid://00000414/#_ftn1" name=_ftnref1>[1] as on Fig 4 below, (which, incidentally, is the only way of forcing all values from 0 to 7 to be displayed, but it could still do with a spot of red for the boys in the 0 column and green for the girls in the 7 column to help you to compare the correct pairs of columns.)

 

 <IMG height=353 src="file://C:\DOCUME~1\Owner\LOCALS~1\Temp\msohtml1\01\clip_image002.gif" width=441 v:shapes="_x0000_i1025">

---

<A title="" style="mso-footnote-id: ftn1" href="mhtml:mid://00000414/#_ftnref1" name=_ftn1>[1] SPSS instructions for this are:

 

 GRAPH

  /BAR(GROUPED) = PCT BY sexism BY v348.

 

generated using paste from SPSS, hence the capital letters.  SPSS is case insensitive for  everything except text in primes, so lower case works just as well.

 

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Thursday, September 24, 2009 2:36 PM

Subject: Frequency and Graph Options

Hi there,

I am trying to run simple frequencies and graphs on a few variables but am
having a small issue. Lets say my variable has 4 options (it was a
survey), 1 = Workgroup#1 2 = Workgroup#2 and so on... When I try to graph
and run frequencies on this variable (lets call it WorkGroup), the
workgroup with no respondents obviously doesn't appear.

My question is that can you somehow change that and make it appear, even
if there is no one from that workgroup who responded on the survey? I need
to make it consistent with how the item actually looked. So for example,
one sample might have 0 in workgroup 1, 10 in workgroup 2, 7 in 3, 5 in 4,
and another sample might have 0 in workgroup 3 but have all other
workgroups represented... how can I make the graphs look the same for both?

I'm using PASW 17 and I have the value labels defined and it still isn't
working.

Thanks so much!

Boris

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

[John F Hall]

Here's what should have been in my previous reply.  SPSS, Word and Outlook Express don't seem to like each other for copying and pasting.

 

You can also display boys and girls together on the same chart as below, (which, incidentally, is the only way of forcing all values from 0 to 7 to be displayed, but it could still do with a spot of red for the boys in the 0 column and green for the girls in the 7 column to help you to compare the correct pairs of columns.)

 

 

 

 

<IMG height=353 src="file://C:\DOCUME~1\Owner\LOCALS~1\Temp\msohtml1\01\clip_image002.gif" width=441 v:shapes="_x0000_i1025">

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Thursday, September 24, 2009 2:36 PM

Subject: Frequency and Graph Options

Hi there,

I am trying to run simple frequencies and graphs on a few variables but am
having a small issue. Lets say my variable has 4 options (it was a
survey), 1 = Workgroup#1 2 = Workgroup#2 and so on... When I try to graph
and run frequencies on this variable (lets call it WorkGroup), the
workgroup with no respondents obviously doesn't appear.

My question is that can you somehow change that and make it appear, even
if there is no one from that workgroup who responded on the survey? I need
to make it consistent with how the item actually looked. So for example,
one sample might have 0 in workgroup 1, 10 in workgroup 2, 7 in 3, 5 in 4,
and another sample might have 0 in workgroup 3 but have all other
workgroups represented... how can I make the graphs look the same for both?

I'm using PASW 17 and I have the value labels defined and it still isn't
working.

Thanks so much!

Boris

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

[Peck, Jon]

With V17, you can do a bar chart of frequencies with or without empty categories.  Use the Chart Builder, put your variable on the x axis, and on the X-axis bar properties, choose the option to show empty categories.

 

If you have the Custom Tables option, you also have the choice of whether or not to show unlabelled categories in the table (made on the Categories and Totals subdialog.)

 

HTH,

Jon Peck

From: [hidden email]

To: [hidden email]

Sent: Thursday, September 24, 2009 2:36 PM

Subject: Frequency and Graph Options

 

Hi there,

I am trying to run simple frequencies and graphs on a few variables but am
having a small issue. Lets say my variable has 4 options (it was a
survey), 1 = Workgroup#1 2 = Workgroup#2 and so on... When I try to graph
and run frequencies on this variable (lets call it WorkGroup), the
workgroup with no respondents obviously doesn't appear.

My question is that can you somehow change that and make it appear, even
if there is no one from that workgroup who responded on the survey? I need
to make it consistent with how the item actually looked. So for example,
one sample might have 0 in workgroup 1, 10 in workgroup 2, 7 in 3, 5 in 4,
and another sample might have 0 in workgroup 3 but have all other
workgroups represented... how can I make the graphs look the same for both?

I'm using PASW 17 and I have the value labels defined and it still isn't
working.

Thanks so much!

Boris