a quick survival analysis question (yes/no)

[paulandpen]

Hi all,

My situation is that I have (n=1200 cases) who have "ALL" experienced an event (churned from a service provider).  I have a set of discrete times (i.e. 3 months, 6 months etc) which represents the time each person self-reports being with the provider before they churned.

I have configured the dataset as a retrospective snapshot as follows:

V1 (3 months duration with provider computed retrospectively)
0 = right censored (yet to churn)
1 = event (churned in this period)

V2 (6 months duration with provider computed retrospectively)
0 = right censored (yet to churn)
1 = event (churned in this period)
2 = left censored (churned in previous period)

Is this a legitimate construction of the dataset that enables running of separate survival analyses for each period of discrete time data to look at key predictors that promote different survival rates etc.

Regards Paul

[Brock-15]

Hi Paul,

While I have not formally used SA in my line of work, I will take my shot
at your question.  I am assuming that all of your customers did enter the
study on the same day, therefore, some of your right-censored customers
may have stuck around for a month, 5 months, 2 years, etc.  Therefore, all
you need are two columns (time in months, and an indicator of churn).  For
those who have not churned, the time is simply the length of time they
have been around since your study has 'ended', with the indicator set to
zero for no event (assuming 1 indicates churn).

HTH,

~ Brock

[Melissa Ives]

From the SPSS 14.0.2 electronic syntax guide:
The basic specification requires three subcommands: TABLES, INTERVALS,
and STATUS.
TABLES identifies at least one survival variable from the active
dataset,        TIME indicator
INTERVALS divides the time period into intervals, and
How to break up the TIME indicator
STATUS names a variable that indicates whether the event occurred.
CHURN indicator

Melissa
The bubbling brook would lose its song if you removed the rocks.


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Paul Dickson
Sent: Monday, February 05, 2007 7:36 PM
To: [hidden email]
Subject: [SPSSX-L] a quick survival analysis question (yes/no)

Hi all,

My situation is that I have (n=1200 cases) who have "ALL" experienced an
event (churned from a service provider).  I have a set of discrete times
(i.e. 3 months, 6 months etc) which represents the time each person
self-reports being with the provider before they churned.

I have configured the dataset as a retrospective snapshot as follows:

V1 (3 months duration with provider computed retrospectively) 0 = right
censored (yet to churn)
1 = event (churned in this period)

V2 (6 months duration with provider computed retrospectively) 0 = right
censored (yet to churn)
1 = event (churned in this period)
2 = left censored (churned in previous period)

Is this a legitimate construction of the dataset that enables running of
separate survival analyses for each period of discrete time data to look
at key predictors that promote different survival rates etc.

Regards Paul


PRIVILEGED AND CONFIDENTIAL INFORMATION
This transmittal and any attachments may contain PRIVILEGED AND
CONFIDENTIAL information and is intended only for the use of the
addressee. If you are not the designated recipient, or an employee
or agent authorized to deliver such transmittals to the designated
recipient, you are hereby notified that any dissemination,
copying or publication of this transmittal is strictly prohibited. If
you have received this transmittal in error, please notify us
immediately by replying to the sender and delete this copy from your
system. You may also call us at [phone] for assistance.