Estimating Group Effects and Differences

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Estimating Group Effects and Differences

Chao yawo-2
Hello, I am estimating a regression model predicting the odds of openness to PREP among foreign born and American born persons at two settings: clinic and community. Foreign born were approached only in the community, while Americans in the clinic.

so the population is stratified into two groups: nativity and location.

i want to be able to account for these differences - specifically whether location or nativity makes a difference in the outcome variable. I can combine all the data and then include dummy variables for each group (foreign vs native) and location (clinic vs community).

Is this the most optimal approach or do we have to do an HLM?

thanks, Yy--
Emmanuel F. Koku, PhD Associate Professor Department of Sociology Drexel University Philadelphia, PA 19104 [hidden email] 215 - 895 - 6144
===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Estimating Group Effects and Differences

Bruce Weaver
Administrator
Perhaps you should post a small sample of what the data looks like.  I say
that, because what you've said about it suggests that nativity and location
are completely confounded. Specifically, you said, "Foreign born were
approached only in the community, while Americans in the clinic."  You
appear to be saying that the Foreign+Clinic and American+Community cells are
empty.  If that is not what you meant, please try again, and please post a
small sample of the data (e.g., with a DATA LIST command).  

HTH.


Chao yawo-2 wrote

> Hello, I am estimating a regression model predicting the odds of openness
> to PREP among foreign born and American born persons at two settings:
> clinic and community. Foreign born were approached only in the community,
> while Americans in the clinic.
>
> so the population is stratified into two groups: nativity and location.
>
> i want to be able to account for these differences - specifically whether
> location or nativity makes a difference in the outcome variable. I can
> combine all the data and then include dummy variables for each group
> (foreign vs native) and location (clinic vs community).
>
> Is this the most optimal approach or do we have to do an HLM?
>
> thanks, Yy--
> Emmanuel F. Koku, PhD Associate Professor Department of Sociology Drexel
> University Philadelphia, PA 19104

> emmanuelkoku@

>  215 - 895 - 6144
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (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





-----
--
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.

--
Sent from: http://spssx-discussion.1045642.n5.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
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

PLEASE NOTE THE FOLLOWING: 
1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above.
2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/).
Reply | Threaded
Open this post in threaded view
|

Re: Estimating Group Effects and Differences

Chao yawo-2
Bruce et al:   Thanks for your response

Here is an example of the data requested.   As you can see, foreign-born women are only recruited in the community and not clinics, and native/american born women are recruited and exposed in the clinic and not the community. 

DATA LIST /ID 1-3 SEX 5 (A) AGE 7-8 nativity(10) site (11) prep(12).
BEGIN DATA
001 m 28 110
002 f 29 221
003 f 45 111
 ...
128 m 17 220
END DATA.

Where:
  • nativity(1--American; 2=foreignborn)
  • Site: (1=clinic; 2-community)
  • prep exposure: (0=no; 1=yes)

Hope this helps - thanks - Yy


On Thu, May 16, 2019 at 8:20 PM Bruce Weaver <[hidden email]> wrote:
Perhaps you should post a small sample of what the data looks like.  I say
that, because what you've said about it suggests that nativity and location
are completely confounded. Specifically, you said, "Foreign born were
approached only in the community, while Americans in the clinic."  You
appear to be saying that the Foreign+Clinic and American+Community cells are
empty.  If that is not what you meant, please try again, and please post a
small sample of the data (e.g., with a DATA LIST command). 

HTH.


Chao yawo-2 wrote
> Hello, I am estimating a regression model predicting the odds of openness
> to PREP among foreign born and American born persons at two settings:
> clinic and community. Foreign born were approached only in the community,
> while Americans in the clinic.
>
> so the population is stratified into two groups: nativity and location.
>
> i want to be able to account for these differences - specifically whether
> location or nativity makes a difference in the outcome variable. I can
> combine all the data and then include dummy variables for each group
> (foreign vs native) and location (clinic vs community).
>
> Is this the most optimal approach or do we have to do an HLM?
>
> thanks, Yy--
> Emmanuel F. Koku, PhD Associate Professor Department of Sociology Drexel
> University Philadelphia, PA 19104

> emmanuelkoku@

>  215 - 895 - 6144
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (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





-----
--
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.

--
Sent from: http://spssx-discussion.1045642.n5.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
===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Estimating Group Effects and Differences

Bruce Weaver
Administrator
If you cross-tabulate nativity, site and sex using your entire data set, what
do you see?  Try this:

CROSSTABS nativity BY site BY SEX.

In the output, right-click the cross-tabulation table and Copy As Plain
Text, and paste the plain text table into a reply.  It may not line up
perfectly in Nabble or in people's e-mail, but readers can copy it and paste
it into a text editor that uses a fixed font and insert tabs where needed.
E.g., here's the cross-tabulation I get using the 4 records you provided:

nativity * site * SEX Crosstabulation
Count
SEX site Total
                        1 2
f nativity 1 1 0 1
                2 0 1 1
        Total 1 1 2
m nativity 1 1 0 1
                2 0 1 1
        Total 1 1 2
Total nativity 1 2 0 2
                2 0 2 2
        Total 2 2 4


If nativity and site are indeed completely confounded in your entire data
set, you can't include both of them in your model.  And you will be unable
to determine how much of any difference between the two cells is due to
nativity, how much is due to site, or how much is due to the interaction
between them.  





Chao yawo-2 wrote

> Bruce et al:   Thanks for your response
>
> Here is an example of the data requested.   As you can see, foreign-born
> women are only recruited in the community and not clinics, and
> native/american born women are recruited and exposed in the clinic and not
> the community.
>
> DATA LIST /ID 1-3 SEX 5 (A) AGE 7-8 nativity(10) site (11) prep(12).
> BEGIN DATA
> 001 m 28 110
> 002 f 29 221
> 003 f 45 111
>  ...
> 128 m 17 220
> END DATA.
>
> Where:
>
>
>    - nativity(1--American; 2=foreignborn)
>    - Site: (1=clinic; 2-community)
>    - prep exposure: (0=no; 1=yes)
>
>
> Hope this helps - thanks - Yy





-----
--
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.

--
Sent from: http://spssx-discussion.1045642.n5.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
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

PLEASE NOTE THE FOLLOWING: 
1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above.
2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/).