Contingency table versus raw data

[Myers, Nicholas]

Hello.

I am hoping that somebody can point me to a website and/or relevant text for how to create a dataset in SPSS from contingency table information and not raw (case-level) data.

The data consist of three binary variables:
race (z, white or black),
drug use (x, yes or no),
symptoms (y, yes or no),
and can be summarized as follows:

                                              SYMPTOMS
RACE  DRUG USE    YES   NO
White Yes         14    107
White No          32    113
Black Yes         11    63
Black No          12    55

I want to impose a logistic regression model but I first need to figure out how to create this dataset in SPSS without having access to the raw (case-level) data.

Thank you.


nick myers



-----------------------------------------------------------------------------------------------------------

Nicholas D. Myers, Ph.D.
University of Miami
School of Education
Department of Educational and Psychological Studies

Research, Measurement and Evaluation program



Merrick Building 311E
P.O. Box 248065
Coral Gables, FL 33124-2040



Tel: [phone]
Fax: [phone]
E-mail: [hidden email]
http://www.education.miami.edu/facultyStaff/Faculty_Bio.asp?ID=143
-----------------------------------------------------------------------------------------------------------

________________________________

=====================
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

[ViAnn Beadle]

There are 3 categorical variables--race, drug use and symptom and cell
weight. There are 12 cells in table so you need to create 12 cases in the
data editor or in a begin data--end data block that look like this:

White Yes Yes 14
White Yes No 107
White No Yes 32
...

Where the 4 variables are Race, DrugUse, Symptom, and caseweight.
Then,  weight the file by the caseweight variable.

Then do a quick check with CROSSTABS to make sure your table matches the
source table.


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Myers, Nicholas
Sent: Tuesday, August 19, 2008 6:27 AM
To: [hidden email]
Subject: Contingency table versus raw data

Hello.

I am hoping that somebody can point me to a website and/or relevant text for
how to create a dataset in SPSS from contingency table information and not
raw (case-level) data.

The data consist of three binary variables:
race (z, white or black),
drug use (x, yes or no),
symptoms (y, yes or no),
and can be summarized as follows:

                                              SYMPTOMS
RACE  DRUG USE    YES   NO
White Yes         14    107
White No          32    113
Black Yes         11    63
Black No          12    55

I want to impose a logistic regression model but I first need to figure out
how to create this dataset in SPSS without having access to the raw
(case-level) data.

Thank you.


nick myers



----------------------------------------------------------------------------
-------------------------------

Nicholas D. Myers, Ph.D.
University of Miami
School of Education
Department of Educational and Psychological Studies

Research, Measurement and Evaluation program



Merrick Building 311E
P.O. Box 248065
Coral Gables, FL 33124-2040



Tel: [phone]
Fax: [phone]
E-mail: [hidden email]
http://www.education.miami.edu/facultyStaff/Faculty_Bio.asp?ID=143
----------------------------------------------------------------------------
-------------------------------

________________________________

=====================
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

[Richard Ristow]

At 08:27 AM 8/19/2008, Myers, Nicholas wrote:

>[I want] to create a dataset in SPSS from contingency table information:
>
>The data consist of three binary variables:
>race (z, white or black),
>drug use (x, yes or no),
>symptoms (y, yes or no),
>and can be summarized as follows:
>                   SYMPTOMS
>RACE  DRUG USE    YES   NO
>White Yes         14    107
>White No          32    113
>Black Yes         11    63
>Black No          12    55

At 09:35 AM 8/19/2008, ViAnn Beadle wrote:

>There are 12 cells in table so you need to create 12 cases in the
>data editor or in a begin data--end data block that look like this:
>
>White Yes Yes 14
>White Yes No 107
>White No Yes 32

Exactly right, except that if the contingency table is in SPSS, you
can use VARSTOCASES to get the data in the form she specifies, rather
than hand-reentering:
|-----------------------------|---------------------------|
|Output Created               |19-AUG-2008 22:24:10       |
|-----------------------------|---------------------------|
RACE  DRUG_USE  YES   NO

White Yes        14  107
White No         32  113
Black Yes        11   63
Black No         12   55

Number of cases read:  4    Number of cases listed:  4


VARSTOCASES  /MAKE Instances FROM YES NO
  /INDEX = Symptoms(Instances)
  /KEEP =  RACE DRUG_USE
  /NULL = KEEP.

Variables to Cases
|-----------------------------|---------------------------|
|Output Created               |19-AUG-2008 22:26:26       |
|-----------------------------|---------------------------|
Generated Variables
|---------|------|
|Name     |Label |
|---------|------|
|Symptoms |<none>|
|---------|------|
|Instances|<none>|
|---------|------|

Processing Statistics
|-------------|-|
|Variables In |4|
|-------------|-|
|Variables Out|4|
|-------------|-|

LIST.
List
|-----------------------------|---------------------------|
|Output Created               |19-AUG-2008 22:26:26       |
|-----------------------------|---------------------------|
RACE  DRUG_USE Symptoms Instances

White Yes      YES           14
White Yes      NO           107
White No       YES           32
White No       NO           113
Black Yes      YES           11
Black Yes      NO            63
Black No       YES           12
Black No       NO            55

Number of cases read:  8    Number of cases listed:  8
=============================
APPENDIX: Test data, and code
=============================
DATA LIST LIST/
      RACE  DRUG_USE    YES   NO
      (A5,  A3,         F4,   F4).
BEGIN DATA
      White Yes         14    107
      White No          32    113
      Black Yes         11    63
      Black No          12    55
END DATA.
LIST.
VARSTOCASES  /MAKE Instances FROM YES NO
  /INDEX = Symptoms(Instances)
  /KEEP =  RACE DRUG_USE
  /NULL = KEEP.
LIST.

=====================
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