Stats Question

[msherman]

Dear List: I have a data set which provides me with the time it took an
individual to evacuate from the WTCs on 9 /  11. (Time they reached
street level minus time they started to evacuate). Currently the
research team is debating how to calculate the rate of evacuation. One
school of thought is to divide the time of evacuation (End time minus
begin time) by the floor the individual started on. The other school of
thought is to use the time itself as the outcome (End Time minus begin
time)  and use the floor they started on as control variable (in
regression). I would be interested in how other folks would view this
situation. As always I very much appreciate the time and thoughts of
other individuals.  thank you.

martin sherman

Martin F. Sherman, Ph.D.
Professor of Psychology
Director of Masters Education: Thesis Track
Loyola College
222 B Beatty Hall
4501 North Charles Street
Baltimore, MD 21210

410 617-2417 (Office)
410 617-5341 (FAX)
[hidden email]

[Spousta Jan]

Hi Martin,

I am rather with the Second School - I see no reason why the evacuation
time should be directly proportional to the floor number, as the First
School seems to take for granted. I would rather expect a relation in
form

expected evacuation time = mean time for reaching the staircase +
possibly non-linear function of the floor number

...which can be better fitted using a regression in the form suggested
by the Second School of Thought :-)

Greetings

Jan

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Martin Sherman
Sent: Thursday, June 22, 2006 3:07 PM
To: [hidden email]
Subject: Stats Question

Dear List: I have a data set which provides me with the time it took an
individual to evacuate from the WTCs on 9 /  11. (Time they reached
street level minus time they started to evacuate). Currently the
research team is debating how to calculate the rate of evacuation. One
school of thought is to divide the time of evacuation (End time minus
begin time) by the floor the individual started on. The other school of
thought is to use the time itself as the outcome (End Time minus begin
time)  and use the floor they started on as control variable (in
regression). I would be interested in how other folks would view this
situation. As always I very much appreciate the time and thoughts of
other individuals.  thank you.

martin sherman

Martin F. Sherman, Ph.D.
Professor of Psychology
Director of Masters Education: Thesis Track Loyola College
222 B Beatty Hall
4501 North Charles Street
Baltimore, MD 21210

410 617-2417 (Office)
410 617-5341 (FAX)
[hidden email]

[Frederic Villamayor Forcada]

Martin:

I see this as a case of censored data. You are underestimating evacuation
time if those who did not survive are not counted. It the towers had
resisted the attack, the mean evacuation time would have been higher than
that computed only with the data from the surviving people. May be Cox's
regression can help you. About the use of the floor number as a covariate,
I think it's a good idea. Any action you take would be speculative. Why
not take a common start time for all the people on tower 1 and another for
tower 2 (the time when the planes crashed), and all data from the people
who did not survive are censored at the time when each tower fell down?

Greetings


Frederic

%%%%%%%%%%%%%%%%%%%%%%%
Frederic Villamayor
Unitat de Bioestadística
Àrea de Desenvolupament Preclínic
CIDF Ferrer Grupo
Juan de Sada, 32
08028 - Barcelona
Espanya

E-mail: [hidden email]
Tel: [phone]
Fax: [phone]
WWW: www.ferrergrupo.com

%%%%%%%%%%%%%%%%%%%%%%%
"Sanity is not statistical"
1984 (George Orwell)



Martin Sherman <[hidden email]>
Enviado por: "SPSSX(r) Discussion" <[hidden email]>
22/06/2006 15:07
Por favor, responda a
Martin Sherman <[hidden email]>


Para
[hidden email]
cc

Asunto
[SPSSX-L] Stats Question





Dear List: I have a data set which provides me with the time it took an
individual to evacuate from the WTCs on 9 /  11. (Time they reached
street level minus time they started to evacuate). Currently the
research team is debating how to calculate the rate of evacuation. One
school of thought is to divide the time of evacuation (End time minus
begin time) by the floor the individual started on. The other school of
thought is to use the time itself as the outcome (End Time minus begin
time)  and use the floor they started on as control variable (in
regression). I would be interested in how other folks would view this
situation. As always I very much appreciate the time and thoughts of
other individuals.  thank you.

martin sherman

Martin F. Sherman, Ph.D.
Professor of Psychology
Director of Masters Education: Thesis Track
Loyola College
222 B Beatty Hall
4501 North Charles Street
Baltimore, MD 21210

410 617-2417 (Office)
410 617-5341 (FAX)
[hidden email]

[Hector Maletta]

A "rate of evacuation" may be defined in different ways, depending on
purpose of the exercise. One reasonable way would be to define it as
"velocity of descent", equivalent to time per floor. The rate would be
simply total time (start to exit) divided number of floors. If this average
is called b, you will have b=total time/number of floors and therefore total
time = b(number of floors), irrespective of the starting floor.
The other approach you suggest is computing a regression equation in the
form Total time = b0 + b1(number of floors). This differs from the previous
one in the presence of b0, i.e. a certain constant time that is independent
of the starting floor. There might be some substantive reason to think that
b0 is not zero. For instance, people may have to negotiate one (or a number
of) "difficult" floors requiring more time than the regular cruising speed
of descent for the remaining floors. The relative importance of b0 on total
time (b0/total time) would be inversely proportional to the number of floors
to descend. The first approach, instead, assumes that the number of
"difficult" floors is proportional to total number of floors, and therefore
its relative importance is zero or constant irrespective of the starting
floor.
Trying the two approaches may put these contrasting "schools of thought" to
the test. If b0 is significantly different from zero, the second school is
right, otherwise it is the first one.
What if b0 is negative? It would mean that there were some "easy" floors
that were descended at more than regular speed, such as for instance the
last few floors before reaching ground zero. With a few refinements, the
model may test also this possibility along with the "difficult floors"
hypothesis, because both or any of them may have been present.
All in all, this line of inquiry would require the second approach. To test
the easy and difficult floor hypotheses separately one may need to use a
more complicated functional form, instead of the simple regression mentioned
before. Otherwise what one gets is the net effect of the two (a zero value
for b0 may mean that the delay caused by difficult floors was offset by the
presence of especially easy floors, and thus b0=0 may not mean absence of
difficult floors unequivocally).
Hope this helps.
Hector
-----Mensaje original-----
De: SPSSX(r) Discussion [mailto:[hidden email]] En nombre de
Martin Sherman
Enviado el: Thursday, June 22, 2006 10:07 AM
Para: [hidden email]
Asunto: Stats Question

Dear List: I have a data set which provides me with the time it took an
individual to evacuate from the WTCs on 9 /  11. (Time they reached
street level minus time they started to evacuate). Currently the
research team is debating how to calculate the rate of evacuation. One
school of thought is to divide the time of evacuation (End time minus
begin time) by the floor the individual started on. The other school of
thought is to use the time itself as the outcome (End Time minus begin
time)  and use the floor they started on as control variable (in
regression). I would be interested in how other folks would view this
situation. As always I very much appreciate the time and thoughts of
other individuals.  thank you.

martin sherman

Martin F. Sherman, Ph.D.
Professor of Psychology
Director of Masters Education: Thesis Track
Loyola College
222 B Beatty Hall
4501 North Charles Street
Baltimore, MD 21210

410 617-2417 (Office)
410 617-5341 (FAX)
[hidden email]

[Hector Maletta]

Frederic,
Great idea about using Cox. However, your reasoning applies only to people
that could possibly start descent. All people in or above the floors where
the planes crashed into were not able to do so, and possibly also many or
all the people in the floor/s immediately below, because of the destruction
around that may have killed some of the people instantly, destroyed access
to stairs, or left people otherwise unable to start descending.
The situation is similar to a medical study on survival time after
successful heart surgery. People who died on the operating table without
having their surgery successfully completed should not actually be counted,
I suppose.
Hector Maletta

-----Mensaje original-----
De: SPSSX(r) Discussion [mailto:[hidden email]] En nombre de
Frederic Villamayor Forcada
Enviado el: Thursday, June 22, 2006 11:08 AM
Para: [hidden email]
Asunto: Re: Stats Question

Martin:

I see this as a case of censored data. You are underestimating evacuation
time if those who did not survive are not counted. It the towers had
resisted the attack, the mean evacuation time would have been higher than
that computed only with the data from the surviving people. May be Cox's
regression can help you. About the use of the floor number as a covariate,
I think it's a good idea. Any action you take would be speculative. Why
not take a common start time for all the people on tower 1 and another for
tower 2 (the time when the planes crashed), and all data from the people
who did not survive are censored at the time when each tower fell down?

Greetings


Frederic

%%%%%%%%%%%%%%%%%%%%%%%
Frederic Villamayor
Unitat de Bioestadística
Àrea de Desenvolupament Preclínic
CIDF Ferrer Grupo
Juan de Sada, 32
08028 - Barcelona
Espanya

E-mail: [hidden email]
Tel: [phone]
Fax: [phone]
WWW: www.ferrergrupo.com

%%%%%%%%%%%%%%%%%%%%%%%
"Sanity is not statistical"
1984 (George Orwell)



Martin Sherman <[hidden email]>
Enviado por: "SPSSX(r) Discussion" <[hidden email]>
22/06/2006 15:07
Por favor, responda a
Martin Sherman <[hidden email]>


Para
[hidden email]
cc

Asunto
[SPSSX-L] Stats Question





Dear List: I have a data set which provides me with the time it took an
individual to evacuate from the WTCs on 9 /  11. (Time they reached
street level minus time they started to evacuate). Currently the
research team is debating how to calculate the rate of evacuation. One
school of thought is to divide the time of evacuation (End time minus
begin time) by the floor the individual started on. The other school of
thought is to use the time itself as the outcome (End Time minus begin
time)  and use the floor they started on as control variable (in
regression). I would be interested in how other folks would view this
situation. As always I very much appreciate the time and thoughts of
other individuals.  thank you.

martin sherman

Martin F. Sherman, Ph.D.
Professor of Psychology
Director of Masters Education: Thesis Track
Loyola College
222 B Beatty Hall
4501 North Charles Street
Baltimore, MD 21210

410 617-2417 (Office)
410 617-5341 (FAX)
[hidden email]

[Hector Maletta]

Martin,
Apparently what we are arriving at is: you apply Cox regression to estimate
risk factors determining the event of evacuation (and the time spent in
achieving it), i.e. the time to arrival to ground zero, using as predictors
the floor of origin plus all your other predictors.
Cox regression estimate the impact of risk factors on odds ratios on the
assumption of proportional hazards. This means that any risk factors
increases the odds of the event in a certain fixed proportion, irrespective
of the time elapsed. For instance, if knowledge of the building reduces the
time by a half, the expected reduction (for people of given characteristics
and given floor of origin) would be a half of the baseline or reference
case, regardless of the time elapsed until exiting the building. A
refinement is introducing non proportional hazards, for instance a
time-dependent covariate: one could imagine that time from a certain floor
to the ground is dependent on tiredness, which would be a function of the
number of floors already descended. In other words, the pace of descent
would be decelerating because people become more tired as they go down. If
so, one possible time-dependent covariate may be time itself.
Then for your second problem. If you already know the time taken to descend,
and the floor of origin, you know the descent velocity (minutes/floors),and
this could be used as a predictor of other things like injuries. As you will
be used the observed velocity as predictor, this second analysis is
independent of the previous one, unless you want to use the PREDICTED
velocity instead of using the observed one, but why would you want that?
You sent me this question off list, but I think you would not object my
sharing it and my comments with everybody, would you?
Hector


-----Mensaje original-----
De: Martin Sherman [mailto:[hidden email]]
Enviado el: Thursday, June 22, 2006 12:09 PM
Para: Hector Maletta
Asunto: Re: Stats Question

Hector:  What would you suggest then.  There are two things we are trying to
do with the data. First, we want to see what variables are predictive of
rate of evacuation (prior evacuations, knowledge of the building, gender,
age, emergency preparedness, etc).  Second we want to see whether rate of
evacuation is predictive of injury during evacuation.   Your thoughts much
appreciated.  martin

>>> Hector Maletta <[hidden email]> 06/22/06 10:27 AM >>>
Frederic,
Great idea about using Cox. However, your reasoning applies only to people
that could possibly start descent. All people in or above the floors where
the planes crashed into were not able to do so, and possibly also many or
all the people in the floor/s immediately below, because of the destruction
around that may have killed some of the people instantly, destroyed access
to stairs, or left people otherwise unable to start descending.
The situation is similar to a medical study on survival time after
successful heart surgery. People who died on the operating table without
having their surgery successfully completed should not actually be counted,
I suppose.
Hector Maletta

-----Mensaje original-----
De: SPSSX(r) Discussion [mailto:[hidden email]] En nombre de
Frederic Villamayor Forcada
Enviado el: Thursday, June 22, 2006 11:08 AM
Para: [hidden email]
Asunto: Re: Stats Question

Martin:

I see this as a case of censored data. You are underestimating evacuation
time if those who did not survive are not counted. It the towers had
resisted the attack, the mean evacuation time would have been higher than
that computed only with the data from the surviving people. May be Cox's
regression can help you. About the use of the floor number as a covariate,
I think it's a good idea. Any action you take would be speculative. Why
not take a common start time for all the people on tower 1 and another for
tower 2 (the time when the planes crashed), and all data from the people
who did not survive are censored at the time when each tower fell down?

Greetings


Frederic

%%%%%%%%%%%%%%%%%%%%%%%
Frederic Villamayor
Unitat de Bioestadística
Àrea de Desenvolupament Preclínic
CIDF Ferrer Grupo
Juan de Sada, 32
08028 - Barcelona
Espanya

E-mail: [hidden email]
Tel: [phone]
Fax: [phone]
WWW: www.ferrergrupo.com

%%%%%%%%%%%%%%%%%%%%%%%
"Sanity is not statistical"
1984 (George Orwell)



Martin Sherman <[hidden email]>
Enviado por: "SPSSX(r) Discussion" <[hidden email]>
22/06/2006 15:07
Por favor, responda a
Martin Sherman <[hidden email]>


Para
[hidden email]
cc

Asunto
[SPSSX-L] Stats Question





Dear List: I have a data set which provides me with the time it took an
individual to evacuate from the WTCs on 9 /  11. (Time they reached
street level minus time they started to evacuate). Currently the
research team is debating how to calculate the rate of evacuation. One
school of thought is to divide the time of evacuation (End time minus
begin time) by the floor the individual started on. The other school of
thought is to use the time itself as the outcome (End Time minus begin
time)  and use the floor they started on as control variable (in
regression). I would be interested in how other folks would view this
situation. As always I very much appreciate the time and thoughts of
other individuals.  thank you.

martin sherman

Martin F. Sherman, Ph.D.
Professor of Psychology
Director of Masters Education: Thesis Track
Loyola College
222 B Beatty Hall
4501 North Charles Street
Baltimore, MD 21210

410 617-2417 (Office)
410 617-5341 (FAX)
[hidden email]

[Spousta Jan]

Hi Hector and Martin,

Perhaps I understand it wrong, but as far as I know, for Cox regression you would need data even about the dead persons - from which floor they started to evacuate, how much time they spent during  the evacuation before the structure collapsed, and all other predictors known about the survivors. I fear that this will be a big problem here - you cannot interview these people. :-(

Jan

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Hector Maletta
Sent: Friday, June 23, 2006 5:13 AM
To: [hidden email]
Subject: Re: Stats Question

Martin,
Apparently what we are arriving at is: you apply Cox regression to estimate risk factors determining the event of evacuation (and the time spent in achieving it), i.e. the time to arrival to ground zero, using as predictors the floor of origin plus all your other predictors.
Cox regression estimate the impact of risk factors on odds ratios on the assumption of proportional hazards. This means that any risk factors increases the odds of the event in a certain fixed proportion, irrespective of the time elapsed. For instance, if knowledge of the building reduces the time by a half, the expected reduction (for people of given characteristics and given floor of origin) would be a half of the baseline or reference case, regardless of the time elapsed until exiting the building. A refinement is introducing non proportional hazards, for instance a time-dependent covariate: one could imagine that time from a certain floor to the ground is dependent on tiredness, which would be a function of the number of floors already descended. In other words, the pace of descent would be decelerating because people become more tired as they go down. If so, one possible time-dependent covariate may be time itself.
Then for your second problem. If you already know the time taken to descend, and the floor of origin, you know the descent velocity (minutes/floors),and this could be used as a predictor of other things like injuries. As you will be used the observed velocity as predictor, this second analysis is independent of the previous one, unless you want to use the PREDICTED velocity instead of using the observed one, but why would you want that?
You sent me this question off list, but I think you would not object my sharing it and my comments with everybody, would you?
Hector


-----Mensaje original-----
De: Martin Sherman [mailto:[hidden email]] Enviado el: Thursday, June 22, 2006 12:09 PM
Para: Hector Maletta
Asunto: Re: Stats Question

Hector:  What would you suggest then.  There are two things we are trying to do with the data. First, we want to see what variables are predictive of rate of evacuation (prior evacuations, knowledge of the building, gender, age, emergency preparedness, etc).  Second we want to see whether rate of
evacuation is predictive of injury during evacuation.   Your thoughts much
appreciated.  martin

>>> Hector Maletta <[hidden email]> 06/22/06 10:27 AM >>>
Frederic,
Great idea about using Cox. However, your reasoning applies only to people that could possibly start descent. All people in or above the floors where the planes crashed into were not able to do so, and possibly also many or all the people in the floor/s immediately below, because of the destruction around that may have killed some of the people instantly, destroyed access to stairs, or left people otherwise unable to start descending.
The situation is similar to a medical study on survival time after successful heart surgery. People who died on the operating table without having their surgery successfully completed should not actually be counted, I suppose.
Hector Maletta

-----Mensaje original-----
De: SPSSX(r) Discussion [mailto:[hidden email]] En nombre de Frederic Villamayor Forcada Enviado el: Thursday, June 22, 2006 11:08 AM
Para: [hidden email]
Asunto: Re: Stats Question

Martin:

I see this as a case of censored data. You are underestimating evacuation time if those who did not survive are not counted. It the towers had resisted the attack, the mean evacuation time would have been higher than that computed only with the data from the surviving people. May be Cox's regression can help you. About the use of the floor number as a covariate, I think it's a good idea. Any action you take would be speculative. Why not take a common start time for all the people on tower 1 and another for tower 2 (the time when the planes crashed), and all data from the people who did not survive are censored at the time when each tower fell down?

Greetings


Frederic

%%%%%%%%%%%%%%%%%%%%%%%
Frederic Villamayor
Unitat de Bioestadística
Àrea de Desenvolupament Preclínic
CIDF Ferrer Grupo
Juan de Sada, 32
08028 - Barcelona
Espanya

E-mail: [hidden email]
Tel: [phone]
Fax: [phone]
WWW: www.ferrergrupo.com

%%%%%%%%%%%%%%%%%%%%%%%
"Sanity is not statistical"
1984 (George Orwell)



Martin Sherman <[hidden email]>
Enviado por: "SPSSX(r) Discussion" <[hidden email]>
22/06/2006 15:07
Por favor, responda a
Martin Sherman <[hidden email]>


Para
[hidden email]
cc

Asunto
[SPSSX-L] Stats Question





Dear List: I have a data set which provides me with the time it took an individual to evacuate from the WTCs on 9 /  11. (Time they reached street level minus time they started to evacuate). Currently the research team is debating how to calculate the rate of evacuation. One school of thought is to divide the time of evacuation (End time minus begin time) by the floor the individual started on. The other school of thought is to use the time itself as the outcome (End Time minus begin
time)  and use the floor they started on as control variable (in regression). I would be interested in how other folks would view this situation. As always I very much appreciate the time and thoughts of other individuals.  thank you.

martin sherman

Martin F. Sherman, Ph.D.
Professor of Psychology
Director of Masters Education: Thesis Track Loyola College
222 B Beatty Hall
4501 North Charles Street
Baltimore, MD 21210

410 617-2417 (Office)
410 617-5341 (FAX)
[hidden email]

[Frederic Villamayor Forcada]

Hector,

I'd just want to remark that what you are defining of "velocity of
descent" is not a velocity, but a period (time elapsing between the
beginning and the end of a cycle). Velocity is inversely proportional to
period, and period is defined in time units.
I'd like to think in a rate of evacuation as something that increases as
evacuation is quicker. Otherwise it is confusing.



Greetings


Frederic

%%%%%%%%%%%%%%%%%%%%%%%
Frederic Villamayor
Unitat de Bioestadística
Àrea de Desenvolupament Preclínic
CIDF Ferrer Grupo
Juan de Sada, 32
08028 - Barcelona
Espanya

E-mail: [hidden email]
Tel: [phone]
Fax: [phone]
WWW: www.ferrergrupo.com

%%%%%%%%%%%%%%%%%%%%%%%
"Sanity is not statistical"
1984 (George Orwell)



Hector Maletta <[hidden email]>
Enviado por: "SPSSX(r) Discussion" <[hidden email]>
22/06/2006 16:10
Por favor, responda a
Hector Maletta <[hidden email]>


Para
[hidden email]
cc

Asunto
Re: [SPSSX-L] Stats Question





A "rate of evacuation" may be defined in different ways, depending on
purpose of the exercise. One reasonable way would be to define it as
"velocity of descent", equivalent to time per floor. The rate would be
simply total time (start to exit) divided number of floors. If this
average
is called b, you will have b=total time/number of floors and therefore
total
time = b(number of floors), irrespective of the starting floor.
The other approach you suggest is computing a regression equation in the
form Total time = b0 + b1(number of floors). This differs from the
previous
one in the presence of b0, i.e. a certain constant time that is
independent
of the starting floor. There might be some substantive reason to think
that
b0 is not zero. For instance, people may have to negotiate one (or a
number
of) "difficult" floors requiring more time than the regular cruising speed
of descent for the remaining floors. The relative importance of b0 on
total
time (b0/total time) would be inversely proportional to the number of
floors
to descend. The first approach, instead, assumes that the number of
"difficult" floors is proportional to total number of floors, and
therefore
its relative importance is zero or constant irrespective of the starting
floor.
Trying the two approaches may put these contrasting "schools of thought"
to
the test. If b0 is significantly different from zero, the second school is
right, otherwise it is the first one.
What if b0 is negative? It would mean that there were some "easy" floors
that were descended at more than regular speed, such as for instance the
last few floors before reaching ground zero. With a few refinements, the
model may test also this possibility along with the "difficult floors"
hypothesis, because both or any of them may have been present.
All in all, this line of inquiry would require the second approach. To
test
the easy and difficult floor hypotheses separately one may need to use a
more complicated functional form, instead of the simple regression
mentioned
before. Otherwise what one gets is the net effect of the two (a zero value
for b0 may mean that the delay caused by difficult floors was offset by
the
presence of especially easy floors, and thus b0=0 may not mean absence of
difficult floors unequivocally).
Hope this helps.
Hector
-----Mensaje original-----
De: SPSSX(r) Discussion [mailto:[hidden email]] En nombre de
Martin Sherman
Enviado el: Thursday, June 22, 2006 10:07 AM
Para: [hidden email]
Asunto: Stats Question

Dear List: I have a data set which provides me with the time it took an
individual to evacuate from the WTCs on 9 /  11. (Time they reached
street level minus time they started to evacuate). Currently the
research team is debating how to calculate the rate of evacuation. One
school of thought is to divide the time of evacuation (End time minus
begin time) by the floor the individual started on. The other school of
thought is to use the time itself as the outcome (End Time minus begin
time)  and use the floor they started on as control variable (in
regression). I would be interested in how other folks would view this
situation. As always I very much appreciate the time and thoughts of
other individuals.  thank you.

martin sherman

Martin F. Sherman, Ph.D.
Professor of Psychology
Director of Masters Education: Thesis Track
Loyola College
222 B Beatty Hall
4501 North Charles Street
Baltimore, MD 21210

410 617-2417 (Office)
410 617-5341 (FAX)
[hidden email]

[Marta García-Granero]

Hi

SJ> Perhaps I understand it wrong, but as far as I know, for Cox
SJ> regression you would need data even about the dead persons - from
SJ> which floor they started to evacuate, how much time they spent
SJ> during  the evacuation before the structure collapsed, and all
SJ> other predictors known about the survivors. I fear that this will
SJ> be a big problem here - you cannot interview these people. :-(

I didn't follow the thread until now. Jan is absolutely right, you are
missing the information about "uncensored" cases.

Regards,

Marta

[Frederic Villamayor Forcada]

Hi all.

As Hector noted yesterday, people in the crash floors and higher floors
could not escape to its fate. They did not evacuate, so they are not
included in the study. But people from the lower flors had the possibility
to evacuate. I do not know if there are data available, but I'd for each
of these persons it should be known:
        Which was the floor where they were at the moment of the crash?
        Did they reach the ground floor?
        When?
As I suggested yesterday, evacuation time could be approximated by the
time from the crash to the arrival at the ground floor. Censoring time
would be the time until the final building falling down. As deads cannot
be interviewed nothing more than approximations can be done.
With this data, Cox's regression can be applied, and the different
covariates can be assessed. I've never used Cox's regression. Please, you
wiser co-listers let us know if my reasoning is wrong.


HTH

Frederic


%%%%%%%%%%%%%%%%%%%%%%%
Frederic Villamayor
Unitat de Bioestadística
Àrea de Desenvolupament Preclínic
CIDF Ferrer Grupo
Juan de Sada, 32
08028 - Barcelona
Espanya

E-mail: [hidden email]
Tel: [phone]
Fax: [phone]
WWW: www.ferrergrupo.com

%%%%%%%%%%%%%%%%%%%%%%%
"Sanity is not statistical"
1984 (George Orwell)



Marta García-Granero <[hidden email]>
Enviado por: "SPSSX(r) Discussion" <[hidden email]>
23/06/2006 09:53
Por favor, responda a
Marta García-Granero              <[hidden email]>


Para
[hidden email]
cc

Asunto
Re: [SPSSX-L] Stats Question





Hi

SJ> Perhaps I understand it wrong, but as far as I know, for Cox
SJ> regression you would need data even about the dead persons - from
SJ> which floor they started to evacuate, how much time they spent
SJ> during  the evacuation before the structure collapsed, and all
SJ> other predictors known about the survivors. I fear that this will
SJ> be a big problem here - you cannot interview these people. :-(

I didn't follow the thread until now. Jan is absolutely right, you are
missing the information about "uncensored" cases.

Regards,

Marta

[Hector Maletta]

I do not remember whether I expressed it right. My idea was about time
elapsed divided number of floors, i.e. time per floor, which is the
reciprocal of velocity (floors per unit of time).
Hector

-----Mensaje original-----
De: SPSSX(r) Discussion [mailto:[hidden email]] En nombre de
Frederic Villamayor Forcada
Enviado el: Friday, June 23, 2006 4:31 AM
Para: [hidden email]
Asunto: Re: Stats Question

Hector,

I'd just want to remark that what you are defining of "velocity of
descent" is not a velocity, but a period (time elapsing between the
beginning and the end of a cycle). Velocity is inversely proportional to
period, and period is defined in time units.
I'd like to think in a rate of evacuation as something that increases as
evacuation is quicker. Otherwise it is confusing.



Greetings


Frederic

%%%%%%%%%%%%%%%%%%%%%%%
Frederic Villamayor
Unitat de Bioestadística
Àrea de Desenvolupament Preclínic
CIDF Ferrer Grupo
Juan de Sada, 32
08028 - Barcelona
Espanya

E-mail: [hidden email]
Tel: [phone]
Fax: [phone]
WWW: www.ferrergrupo.com

%%%%%%%%%%%%%%%%%%%%%%%
"Sanity is not statistical"
1984 (George Orwell)



Hector Maletta <[hidden email]>
Enviado por: "SPSSX(r) Discussion" <[hidden email]>
22/06/2006 16:10
Por favor, responda a
Hector Maletta <[hidden email]>


Para
[hidden email]
cc

Asunto
Re: [SPSSX-L] Stats Question





A "rate of evacuation" may be defined in different ways, depending on
purpose of the exercise. One reasonable way would be to define it as
"velocity of descent", equivalent to time per floor. The rate would be
simply total time (start to exit) divided number of floors. If this
average
is called b, you will have b=total time/number of floors and therefore
total
time = b(number of floors), irrespective of the starting floor.
The other approach you suggest is computing a regression equation in the
form Total time = b0 + b1(number of floors). This differs from the
previous
one in the presence of b0, i.e. a certain constant time that is
independent
of the starting floor. There might be some substantive reason to think
that
b0 is not zero. For instance, people may have to negotiate one (or a
number
of) "difficult" floors requiring more time than the regular cruising speed
of descent for the remaining floors. The relative importance of b0 on
total
time (b0/total time) would be inversely proportional to the number of
floors
to descend. The first approach, instead, assumes that the number of
"difficult" floors is proportional to total number of floors, and
therefore
its relative importance is zero or constant irrespective of the starting
floor.
Trying the two approaches may put these contrasting "schools of thought"
to
the test. If b0 is significantly different from zero, the second school is
right, otherwise it is the first one.
What if b0 is negative? It would mean that there were some "easy" floors
that were descended at more than regular speed, such as for instance the
last few floors before reaching ground zero. With a few refinements, the
model may test also this possibility along with the "difficult floors"
hypothesis, because both or any of them may have been present.
All in all, this line of inquiry would require the second approach. To
test
the easy and difficult floor hypotheses separately one may need to use a
more complicated functional form, instead of the simple regression
mentioned
before. Otherwise what one gets is the net effect of the two (a zero value
for b0 may mean that the delay caused by difficult floors was offset by
the
presence of especially easy floors, and thus b0=0 may not mean absence of
difficult floors unequivocally).
Hope this helps.
Hector
-----Mensaje original-----
De: SPSSX(r) Discussion [mailto:[hidden email]] En nombre de
Martin Sherman
Enviado el: Thursday, June 22, 2006 10:07 AM
Para: [hidden email]
Asunto: Stats Question

Dear List: I have a data set which provides me with the time it took an
individual to evacuate from the WTCs on 9 /  11. (Time they reached
street level minus time they started to evacuate). Currently the
research team is debating how to calculate the rate of evacuation. One
school of thought is to divide the time of evacuation (End time minus
begin time) by the floor the individual started on. The other school of
thought is to use the time itself as the outcome (End Time minus begin
time)  and use the floor they started on as control variable (in
regression). I would be interested in how other folks would view this
situation. As always I very much appreciate the time and thoughts of
other individuals.  thank you.

martin sherman

Martin F. Sherman, Ph.D.
Professor of Psychology
Director of Masters Education: Thesis Track
Loyola College
222 B Beatty Hall
4501 North Charles Street
Baltimore, MD 21210

410 617-2417 (Office)
410 617-5341 (FAX)
[hidden email]

[Maguin, Eugene]

I saw the original posting and then watched the exchanges about how to work
this problem. I'd like to ask a really basic question, more out of curiosity
than anything else. The question is where did these data come from? I read
the popularly published account of the attack (the name of which I forget)
and find it kind of hard to believe that people could recall very accurately
when they started down and when they hit the 'ground floor' however you
define that. I doubt also that there were 'official' records of people
coming down. To the extent that the accounts in the book are accurate, I'd
also think that one's velocity down the stairs depended on when one started
down. Velocity would have slowed as volume increased and as rescue people
started up. But, mainly, I'm curious of the circumstances of the data's
collection.

Gene Maguin

[Mark A Davenport MADAVENP]

Correct me if I am wrong, but I seem to recall that people were initially
asked to stay inside the building for fear of falling debris.  If this is
so, the time of escape for those that followed this request might be
artificially long.


***************************************************************************************************************************************************************
Mark A. Davenport Ph.D.
Asst. to the Vice Chancellor for Student Affairs
Office of Student Affairs Research and Evaluation
The University of North Carolina at Greensboro
336.334.5582
[hidden email]

'An approximate answer to the right question is worth a good deal more
than an exact answer to an approximate question.' --a paraphrase of J. W.
Tukey (1962)