Modelling multivariate outcomes of the same variable

[Luca Meyer-3]

Here's a more mathematical/statistical issue than an SPSS's one.

I have got about 10 months data describing the distribution of one
categorical variable related to incoming calls into a call-centre. The
variable can take the following values:

[a] call answered immediately
[b] call answered after some wainting in cue
[c] hang-ups while waiting in cue

I would like to try modellng these data with the objective to study the
relationship between [a] and [c]. More specifically I would like to identify
some sort of function relating the 2 values - that is for instance I would
like to establish that if the proportion of calls answered immediately
decreases by 5%, that hungs-ups while waiting increases by 8%.

Is there some method/procedure I could use with SPSS (v 16)?

I was thinking to multivariate logit functions but that would imply I have
some predictors which I not necessarily want to use, wouldn't it?

Thanks in advance for the assistance you can provide.

Luca

Mr. Luca MEYER
Market research, data analysis & more
www.lucameyer.com <http://www.lucameyer.com/>  - Tel: [phone]

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