Date:         Fri, 14 Jan 2011 16:55:18 -0500
Reply-To:     "Roberts, Michael" <Michael.Roberts@ahca.myflorida.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Roberts, Michael" <Michael.Roberts@ahca.myflorida.com>
Subject:      How to identify blocks of consecutive cases in SPSS?
Content-Type: multipart/alternative;

Good Afternoon All,

I'm hoping someone in here would have a solution for a data manipulation problem I am facing, and one which eludes me:

I have cases, some of which are consecutive in time and the only indicator I have is the month - so for a given case, I have an Identifier, and the month (plus other data which are no help!)  The problem is that for a given identifier, there are breaks between continuous periods, and I need to be able to measure the length of time any given contiguous period - somewhat like a length of stay in a hospital. I have done something like this in the past with regular dates, but cannot for the life of me remember how I did so, and would appreciate some advice.  Data may look like this:

ID   Month
1a  3
1a  6
1a  7
1a  8
1a  11
1a  12
2a  9
2a  11
2a  12
2a  15
2a  16
2a  17
3a  1
3a  3
3a  4
3a  7
3a  8
3a  9

So, 1a has two spans of 3 consecutive months and 2 months, 2a has 1 span of 2 consecutive months, etc.


TIA
Mike

[text/html]