Analytical Problem

[Rodrigo Guerrero]

Hello all,


I have an analysis problem that I could use some help with, although this is not strictly an SPSS issue. 

I have a sequence of events (T) that happen at specific intervals until a particular response is recorded.  So my data is structure like this:

ID  T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11  Response
1    1  1   1   1  1    0   0   0   0   0   0         1
2    1  1   1   1  1    1   1   1   0   0   0         1
3    1  1   1   1  1    1   1   1   1   1   1         0
4    1  0   0   0  0    0   0   0   0   0   0         1

1 = Yes to Event/Response
0 = No To Event/Response

Everyone was subjected to the sequence of events up to when they responded, so I do not have a control group or different treatment groups. 

I am interested in measuring the effect of each event on the response, but everything I can come up with assumes independence between events, which clearly is not the case here. 

Can anyone point me in the right direction?  I thought about repeated measures, but I only have one group.

Thank you.


Rodrigo Guerrero


--
"Often the difference between the possible and the impossible is a person's determination"

[Maguin, Eugene]

Rodrigo,

 

How about telling us more about this analysis. In particular, 1) how does T1-T11 relate to Response? 2) What is the difference between people who have T1, T2=1; T3-T11=0 and people who have T1-T11=1? 3) Your data suggest that T1-T11 is an ordered sequence in that before T5, for example, can be 1, T1-T4 have to be 1. Is this true? There's more questions but start here.

 

Gene Maguin

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Rodrigo Guerrero
Sent: Monday, August 29, 2011 12:32 PM
To: [hidden email]
Subject: Analytical Problem

Hello all,


I have an analysis problem that I could use some help with, although this is not strictly an SPSS issue. 

I have a sequence of events (T) that happen at specific intervals until a particular response is recorded.  So my data is structure like this:

ID  T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11  Response
1    1  1   1   1  1    0   0   0   0   0   0         1
2    1  1   1   1  1    1   1   1   0   0   0         1
3    1  1   1   1  1    1   1   1   1   1   1         0
4    1  0   0   0  0    0   0   0   0   0   0         1

1 = Yes to Event/Response
0 = No To Event/Response

Everyone was subjected to the sequence of events up to when they responded, so I do not have a control group or different treatment groups. 

I am interested in measuring the effect of each event on the response, but everything I can come up with assumes independence between events, which clearly is not the case here. 

Can anyone point me in the right direction?  I thought about repeated measures, but I only have one group.

Thank you.


Rodrigo Guerrero


--
"Often the difference between the possible and the impossible is a person's determination"

[Bruce Weaver]

Are the events always in the same order?  If so, it looks like life table analysis could be used.  There's an example in the Help:  Analyze - Survival - Life Tables, then click on Help, then click Show Me.

Rodrigo Guerrero wrote

Hello all,


I have an analysis problem that I could use some help with, although this is
not strictly an SPSS issue.

I have a sequence of events (T) that happen at specific intervals until a
particular response is recorded.  So my data is structure like this:

ID  T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11  Response
1    1  1   1   1  1    0   0   0   0   0   0         1
2    1  1   1   1  1    1   1   1   0   0   0         1
3    1  1   1   1  1    1   1   1   1   1   1         0
4    1  0   0   0  0    0   0   0   0   0   0         1

1 = Yes to Event/Response
0 = No To Event/Response

Everyone was subjected to the sequence of events up to when they responded,
so I do not have a control group or different treatment groups.

I am interested in measuring the effect of each event on the response, but
everything I can come up with assumes independence between events, which
clearly is not the case here.

Can anyone point me in the right direction?  I thought about repeated
measures, but I only have one group.

Thank you.


Rodrigo Guerrero


--
"Often the difference between the possible and the impossible is a person's
determination"

[Rodrigo Guerrero]

Hi Bruce,

Thank you.  The events are always in sequential order.  I will look at survival analysis as an option.

 

On Mon, Aug 29, 2011 at 10:35 AM, Bruce Weaver <[hidden email]> wrote:

Are the events always in the same order?  If so, it looks like life table
analysis could be used.  There's an example in the Help:  Analyze - Survival
- Life Tables, then click on Help, then click Show Me.

Rodrigo Guerrero wrote:
>
> Hello all,
>
>
> I have an analysis problem that I could use some help with, although this
> is
> not strictly an SPSS issue.
>
> I have a sequence of events (T) that happen at specific intervals until a
> particular response is recorded.  So my data is structure like this:
>
> ID  T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11  Response
> 1    1  1   1   1  1    0   0   0   0   0   0         1
> 2    1  1   1   1  1    1   1   1   0   0   0         1
> 3    1  1   1   1  1    1   1   1   1   1   1         0
> 4    1  0   0   0  0    0   0   0   0   0   0         1
>
> 1 = Yes to Event/Response
> 0 = No To Event/Response
>
> Everyone was subjected to the sequence of events up to when they
> responded,
> so I do not have a control group or different treatment groups.
>
> I am interested in measuring the effect of each event on the response, but
> everything I can come up with assumes independence between events, which
> clearly is not the case here.
>
> Can anyone point me in the right direction?  I thought about repeated
> measures, but I only have one group.
>
> Thank you.
>
>
> Rodrigo Guerrero
>
>
> --
> "Often the difference between the possible and the impossible is a
> person's
> determination"
>

-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Analytical-Problem-tp4746799p4747035.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

--
"Often the difference between the possible and the impossible is a person's determination"

[Buhi, Eric]

CONTENTS DELETED

The author has deleted this message.

[Rodrigo Guerrero]

Hi Gene,


Thank you for your questions.  I hope these answers help.

1) how does T1-T11 relate to Response?

     We think that T1-T11 influence response.  Some people would respond regardless, but the hope is that as they are exposed to the T's, their likelihood to respond increases.   

2) What is the difference between people who have T1, T2=1; T3-T11=0 and people who have T1-T11=1?

      Test subjects start at T1 and are sequentially exposed to the events according to a fixed schedule.  When they respond, the get no further T-events which is designated a '0' in the following T-events

3) Your data suggest that T1-T11 is an ordered sequence in that before T5, for example, can be 1, T1-T4 have to be 1. Is this true?

      Yes.  T5 will follow T1-T4 and they have to be 1.

Thank you for your time Gene.


Rodrigo

On Mon, Aug 29, 2011 at 10:29 AM, Gene Maguin <[hidden email]> wrote:

Rodrigo,

 

How about telling us more about this analysis. In particular, 1) how does T1-T11 relate to Response? 2) What is the difference between people who have T1, T2=1; T3-T11=0 and people who have T1-T11=1? 3) Your data suggest that T1-T11 is an ordered sequence in that before T5, for example, can be 1, T1-T4 have to be 1. Is this true? There's more questions but start here.

 

Gene Maguin

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Rodrigo Guerrero
Sent: Monday, August 29, 2011 12:32 PM
To: [hidden email]
Subject: Analytical Problem

Hello all,


I have an analysis problem that I could use some help with, although this is not strictly an SPSS issue. 

I have a sequence of events (T) that happen at specific intervals until a particular response is recorded.  So my data is structure like this:

ID  T1 T2 T3 T4 T5 T6 T7 T8 T9 T10 T11  Response
1    1  1   1   1  1    0   0   0   0   0   0         1
2    1  1   1   1  1    1   1   1   0   0   0         1
3    1  1   1   1  1    1   1   1   1   1   1         0
4    1  0   0   0  0    0   0   0   0   0   0         1

1 = Yes to Event/Response
0 = No To Event/Response

Everyone was subjected to the sequence of events up to when they responded, so I do not have a control group or different treatment groups. 

I am interested in measuring the effect of each event on the response, but everything I can come up with assumes independence between events, which clearly is not the case here. 

Can anyone point me in the right direction?  I thought about repeated measures, but I only have one group.

Thank you.


Rodrigo Guerrero


--
"Often the difference between the possible and the impossible is a person's determination"

--
"Often the difference between the possible and the impossible is a person's determination"