Line graph for multi response variables

[Nicole]

Hello to all of you,

I am using SPSS 17.0 and I am trying to create a line graph for a set of multi response variables, but even after numerous attempts all I seem to be able to produce are seperate graphs for the concerning variables :(

The variables indicate various types of aggressive behaviour. Since in the answers (N = 750) to my research question several aggressive acts may be mentioned in various combinations, I decided to note them separately -> variable A = Offensive behaviour / B = Threatening behaviour / C = Physical aggression against a person / D = Physical aggression against an object, etc. with value 0 = Non-respons / 1 = Yes / 2 = No.

In the line graph that I am aspiring, the X-axis is supposed to denote the time range (from 01/01/2000 - 31/12/2010) and I would like the lines to denote the types of aggression (with value 1 as the only one that counts).

I really hope someone can tell me how to do this (either by using SPSS or Excel)...

Many thanks in advance!!!

[ViAnn Beadle]

So tell us what you've done so far.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Nicole
Sent: Wednesday, May 18, 2011 9:59 AM
To: [hidden email]
Subject: Line graph for multi response variables

Hello to all of you,

I am using SPSS 17.0 and I am trying to create a line graph for a set of
multi response variables, but even after numerous attempts all I seem to be
able to produce are seperate graphs for the concerning variables :(

The variables indicate various types of aggressive behaviour. Since in the
answers (N = 750) to my research question several aggressive acts may be
mentioned in various combinations, I decided to note them separately ->
variable A = Offensive behaviour / B = Threatening behaviour / C = Physical
aggression against a person / D = Physical aggression against an object,
etc. with value 0 = Non-respons / 1 = Yes / 2 = No.

In the line graph that I am aspiring, the X-axis is supposed to denote the
time range (from 01/01/2000 - 31/12/2010) and I would like the lines to
denote the types of aggression (with value 1 as the only one that counts).

I really hope someone can tell me how to do this (either by using SPSS or
Excel)...

Many thanks in advance!!!

--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Line-graph-for-multi-response-
variables-tp4406894p4406894.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

[Nicole]

I for example defined them as a multiresponse set and I tried to use this 'new' variable to make a line graph, but this did not work.

Furthermore I have selected them simultaneously and I moved them together into the axis-box of the 'legacy dialogs - line graph' / 'interactive - line graph'. Also without success.

I then tried to adapt the formula for generating a line graph so that the programme would include seperate variables at the same time, but this only led to error notifications...

[ViAnn Beadle]

What summary are you trying to apply to these variables? What is their level
of measurement?

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Nicole
Sent: Wednesday, May 18, 2011 1:04 PM
To: [hidden email]
Subject: Re: Line graph for multi response variables

I for example defined them as a multiresponse set and I tried to use this
'new' variable to make a line graph, but this did not work.

Furthermore I have selected them simultaneously and I moved them together
into the axis-box of the 'legacy dialogs - line graph' / 'interactive - line
graph'. Also without success.

I then tried to adapt the formula for generating a line graph so that the
programme would include seperate variables at the same time, but this only
led to error notifications...

--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Line-graph-for-multi-response-
variables-tp4406894p4407414.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

[Nicole]

These are nominal data, so a mean or a mode would not make any sense. Answer possibilities are YES (= 1) or NO (= 2). For my graph I am interested in the number of cases that the answer was YES on a particular date / in a particular month.

[ViAnn Beadle]

Because the variables are categorical is why you cannot grab multiple
variables and put them on the Y-Axis. You could do this if you recoded your
variables into new variables with values 0 or No and 1 for yes. Specify
these as interval and then specify sum as your summary statistic.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Nicole
Sent: Thursday, May 19, 2011 12:59 AM
To: [hidden email]
Subject: Re: Line graph for multi response variables

These are nominal data, so a mean or a mode would not make any sense. Answer
possibilities are YES (= 1) or NO (= 2). For my graph I am interested in the
number of cases that the answer was YES on a particular date / in a
particular month.

--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Line-graph-for-multi-response-
variables-tp4406894p4408914.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

[Nicole]

Yesssss, it works -> You are my hero!!! If you could see me now you would see a big smile...

I changed the measure of the variables from nominal to ordinal and the values to 0 = No / 1 = Yes.

Then I selected graphs - legacy dialogs - line - multiple summaries of separate variables. I have put the variables simultaneously in the box 'lines represent' and I chose 'sum of values'. The variable indicating the time sequence I put in the box 'category axis'.

This produced exactly the graph I was longing for :) Thank you very, very much!