Calculating the difference between cases

[Veena Nambiar]

I have a data file that contains a variable id and score. Id is coded “0” for the target subject and “1” for subjects within the targets group. I would like to create a separate variable diff that takes the difference between the target subject’s score and each of the other subjects in the group. Here is an example below. Is there a way to do this in SPSS?

 

 

id	score	diff
1	82	0
1	75	7
1	82	0
1	61	21
1	66	16
1	92	-10
1	23	59
1	10	72
1	77	5
0	82	.
1	43	39
1	31	51
1	37	45
1	12	70
1	55	27
1	95	-13

 

 

Veena Nambiar

 

[David Marso]

Very difficult to imagine why you want to do this and/or what your full data actually look like since there are multiple subjects.
For the data you present here is a very simple solution.
OTOH you are obviously *NOT* providing adequate information to generate a general solution so this effort may prove to be utterly useless unless you rethink the more general issue (Cartesian product) and redesign your data structure.
data list free / gp score.
begin data
1 82 1 75 1 82 1 61 1 66 1 92 1 23 1 10 1 77 0 82 1 43 1 31 1 37 1 12 1 55 1 95
end data.
COMPUTE @ORDER@=$CASENUM.
SORT CASES BY GP.
DO IF $CASENUM=1.
+  COMPUTE #TARG=score.
ELSE.
+  COMPUTE DIFF=#TARG-SCORE.
END IF.
SORT CASES BY @ORDER@.
LIST.

Veena Nambiar wrote

I have a data file that contains a variable id and score. Id is coded "0" for the target subject and "1" for subjects within the targets group. I would like to create a separate variable diff that takes the difference between the target subject's score and each of the other subjects in the group. Here is an example below. Is there a way to do this in SPSS?


id

score

diff

1

82

0

1

75

7

1

82

0

1

61

21

1

66

16

1

92

-10

1

23

59

1

10

72

1

77

5

0

82

              .

1

43

39

1

31

51

1

37

45

1

12

70

1

55

27

1

95

-13



Veena Nambiar

[David Marso]

Ponder the following while you are at this:
--
DATA LIST FREE / GP SCORE.
BEGIN DATA
1 20 1 40 1 30 1 65 1 23
2 10 2 15 2 30 2 40
END DATA.

CASES TO VARS /ID=GP.
COMPUTE @=$SYSMIS.
COUNT #MAX=score.1 TO @ (LO THRU HI, SYSMIS).
VECTOR DATA=score.1 TO @ .
LOOP SUBID1=1 TO #MAX.
+ LOOP SUBID2=1 TO #MAX.
+   DO IF NOT(SUBID1 EQ SUBID2) AND NOT MISSING (DATA(SUBID1)) AND NOT MISSING (DATA(SUBID2)).
+     COMPUTE SCORE_1=DATA(SUBID1).
+     COMPUTE SCORE_2=DATA(SUBID2).
+     COMPUTE DIFF= SCORE_2-SCORE_1.
+     XSAVE OUTFILE "TEMP.SAV"
           / KEEP GP SUBID1 SUBID2 SCORE_2 SCORE_1 DIFF.
+   END IF.
+ END LOOP.
END LOOP.
EXECUTE.
GET FILE "TEMP.SAV".
LIST.

David Marso wrote

Very difficult to imagine why you want to do this and/or what your full data actually look like since there are multiple subjects.
For the data you present here is a very simple solution.
OTOH you are obviously *NOT* providing adequate information to generate a general solution so this effort may prove to be utterly useless unless you rethink the more general issue (Cartesian product) and redesign your data structure.
data list free / gp score.
begin data
1 82 1 75 1 82 1 61 1 66 1 92 1 23 1 10 1 77 0 82 1 43 1 31 1 37 1 12 1 55 1 95
end data.
COMPUTE @ORDER@=$CASENUM.
SORT CASES BY GP.
DO IF $CASENUM=1.
+  COMPUTE #TARG=score.
ELSE.
+  COMPUTE DIFF=#TARG-SCORE.
END IF.
SORT CASES BY @ORDER@.
LIST.

Veena Nambiar wrote

I have a data file that contains a variable id and score. Id is coded "0" for the target subject and "1" for subjects within the targets group. I would like to create a separate variable diff that takes the difference between the target subject's score and each of the other subjects in the group. Here is an example below. Is there a way to do this in SPSS?


id

score

diff

1

82

0

1

75

7

1

82

0

1

61

21

1

66

16

1

92

-10

1

23

59

1

10

72

1

77

5

0

82

              .

1

43

39

1

31

51

1

37

45

1

12

70

1

55

27

1

95

-13



Veena Nambiar

[Bruce Weaver]

Here's another way without the sorting (although it does leave an extra variable in the file--G0score, which has the Group 0 score written on every case).  It could always be deleted with a DELETE VARIABLES command.

data list free / gp score.
begin data
1 82 1 75 1 82 1 61 1 66 1 92 1 23 1 10 1 77 0 82 1 43 1 31 1 37 1 12 1 55 1 95
end data.

IF (GP EQ 0) G0score = score.
AGGREGATE
  /OUTFILE=* MODE=ADDVARIABLES OVERWRITE=YES
  /G0score=FIRST(G0score).
IF (GP EQ 1) DIFF = G0score - score.
LIST.

David Marso wrote

Very difficult to imagine why you want to do this and/or what your full data actually look like since there are multiple subjects.
For the data you present here is a very simple solution.
OTOH you are obviously *NOT* providing adequate information to generate a general solution so this effort may prove to be utterly useless unless you rethink the more general issue (Cartesian product) and redesign your data structure.
data list free / gp score.
begin data
1 82 1 75 1 82 1 61 1 66 1 92 1 23 1 10 1 77 0 82 1 43 1 31 1 37 1 12 1 55 1 95
end data.
COMPUTE @ORDER@=$CASENUM.
SORT CASES BY GP.
DO IF $CASENUM=1.
+  COMPUTE #TARG=score.
ELSE.
+  COMPUTE DIFF=#TARG-SCORE.
END IF.
SORT CASES BY @ORDER@.
LIST.

Veena Nambiar wrote

I have a data file that contains a variable id and score. Id is coded "0" for the target subject and "1" for subjects within the targets group. I would like to create a separate variable diff that takes the difference between the target subject's score and each of the other subjects in the group. Here is an example below. Is there a way to do this in SPSS?


id

score

diff

1

82

0

1

75

7

1

82

0

1

61

21

1

66

16

1

92

-10

1

23

59

1

10

72

1

77

5

0

82

              .

1

43

39

1

31

51

1

37

45

1

12

70

1

55

27

1

95

-13



Veena Nambiar

[Veena Nambiar]

Thanks for the response David. I needed to compute these differences for use in another set of calculations that we do @ the state for reporting.

With the help of another SPSS user, I think I have a solution. Here is a step by step of the solution.

I have a data file with ids and scores. The ids and scores for group 1 are in id1 and score1, group 2 are in id2 and score2, and so on. The zero in the id field represents a target case and the 1 represents cases in the target's group. My file structure is as follows:

row id1 score1  id2 score2
1      1        92      1       41
2      0        56      1       75
3      1         45             0       22

1. Restructure the data file so that ids and scores are stacked on top of each other. Create a group id to differentiate between groups.

varstocases
*create a variable caseid that stacks data from id1 to id1000.
/make caseid from id1 to id1000
*create a variable score that stacks data from score1 to score1000.
/make score from score1 score1000
*create a variable that identifies the group id and score belong to.
  /INDEX=groupid (1000)
*keep other important variables from data file.
  /KEEP=y
*drop null values.
  /NULL=DROP.

This is the resulting file:
groupid  id   score
1                1      92
1                0      56
1                1      45
2                1      41
2                1      75
2                0      22

2. Create a variable tscore, which contains the score of the target cases only.

*sort cases by group id, caseid, and score so that all cases in the group are together and the target subject (coded 0 under id) is the first case in each group.
sort cases by groupid(a) caseid(a).

The file looks like this:

groupid  id   score
1               0      56
1               1      92
1               1      45
2               0      22
2               1      75
2               1      41

*select only target subjects identified as 0 under caseid.
DO IF (caseid=0).
*compute a new variable with the scores of target cases.
COMPUTE tscore=score.
*select cases in each target's group, identified as 1 under caseid.
ELSE IF (caseid=1).
*compute the lag of the target score, which populates the variable tscore with the target's score for all cases in the target's group.
COMPUTE tscore = LAG(tscore).
END IF.
execute.

The above syntax yields this:
groupid  id   score      tscore
1        0              56              56
1        1              92              56
1        1              45              56
2        0             22       22
2        1             75               22
2        1             41               22

3. Calculate the difference between the target's score and the score of each person in the group.
*select only those cases in each target's group, coded 1 under caseid.
do if caseid=1.
*compute the difference between the target's score and the score of each case in the target's group.
compute diff=tscore-score.
end if.
execute.

The final file looks like this:
groupid  id   score      tscore      diff
1        0              56              56         .
1        1              92              56         -36
1        1              45              56          11
2        0             22       22          .
2        1             75               22         -53
2        1             41               22        -19


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of David Marso
Sent: Tuesday, April 03, 2012 5:38 AM
To: [hidden email]
Subject: Re: Calculating the difference between cases

Very difficult to imagine why you want to do this and/or what your full data actually look like since there are multiple subjects.
For the data you present here is a very simple solution.
OTOH you are obviously *NOT* providing adequate information to generate a general solution so this effort may prove to be utterly useless unless you rethink the more general issue (Cartesian product) and redesign your data structure.
data list free / gp score.
begin data
1 82 1 75 1 82 1 61 1 66 1 92 1 23 1 10 1 77 0 82 1 43 1 31 1 37 1 12 1 55 1
95
end data.
COMPUTE @ORDER@=$CASENUM.
SORT CASES BY GP.
DO IF $CASENUM=1.
+  COMPUTE #TARG=score.
ELSE.
+  COMPUTE DIFF=#TARG-SCORE.
END IF.
SORT CASES BY @ORDER@.
LIST.

Veena Nambiar wrote

>
> I have a data file that contains a variable id and score. Id is coded "0"
> for the target subject and "1" for subjects within the targets group.
> I would like to create a separate variable diff that takes the
> difference between the target subject's score and each of the other
> subjects in the group. Here is an example below. Is there a way to do this in SPSS?
>
>
> id
>
> score
>
> diff
>
> 1
>
> 82
>
> 0
>
> 1
>
> 75
>
> 7
>
> 1
>
> 82
>
> 0
>
> 1
>
> 61
>
> 21
>
> 1
>
> 66
>
> 16
>
> 1
>
> 92
>
> -10
>
> 1
>
> 23
>
> 59
>
> 1
>
> 10
>
> 72
>
> 1
>
> 77
>
> 5
>
> 0
>
> 82
>
>               .
>
> 1
>
> 43
>
> 39
>
> 1
>
> 31
>
> 51
>
> 1
>
> 37
>
> 45
>
> 1
>
> 12
>
> 70
>
> 1
>
> 55
>
> 27
>
> 1
>
> 95
>
> -13
>
>
>
> Veena Nambiar
>

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Calculating-the-difference-between-cases-tp5601947p5614983.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

=====================
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]

I think this gets you to the same result, but again, without the sorting.

* Read in sample data.
data list list / row id1 score1  id2 score2 .
begin data
1      1        92      1       41
2      0        56      1       75
3      1         45     0       22
end data.

* Given the order of the variables, "id1 TO id2"
* and "score1 TO score2" will not work--the variables
* must be listed individually. Perhaps the variables
* are ordered differently in the actual file with
* 1000 variables.

varstocases
 /make caseid from id1 id2
 /make score from score1 score2
 /INDEX=groupid (2)
 /NULL=DROP
.

IF (caseid EQ 0) tscore = score.
AGGREGATE
  /OUTFILE=* MODE=ADDVARIABLES OVERWRITE=YES
  /BREAK = groupid
  /tscore=FIRST(tscore).
IF (caseid EQ 1) DIFF = tscore - score.
LIST.

OUTPUT:
     row groupid   caseid    score   tscore     DIFF
 
    1.00      1      1.00    92.00    56.00   -36.00
    1.00      2      1.00    41.00    22.00   -19.00
    2.00      1       .00    56.00    56.00      .
    2.00      2      1.00    75.00    22.00   -53.00
    3.00      1      1.00    45.00    56.00    11.00
    3.00      2       .00    22.00    22.00      .
 
Number of cases read:  6    Number of cases listed:  6

Veena Nambiar wrote

Thanks for the response David. I needed to compute these differences for use in another set of calculations that we do @ the state for reporting.

With the help of another SPSS user, I think I have a solution. Here is a step by step of the solution.

I have a data file with ids and scores. The ids and scores for group 1 are in id1 and score1, group 2 are in id2 and score2, and so on. The zero in the id field represents a target case and the 1 represents cases in the target's group. My file structure is as follows:

row id1 score1  id2 score2
1      1        92      1       41
2      0        56      1       75
3      1         45             0       22

1. Restructure the data file so that ids and scores are stacked on top of each other. Create a group id to differentiate between groups.

varstocases
*create a variable caseid that stacks data from id1 to id1000.
/make caseid from id1 to id1000
*create a variable score that stacks data from score1 to score1000.
/make score from score1 score1000
*create a variable that identifies the group id and score belong to.
  /INDEX=groupid (1000)
*keep other important variables from data file.
  /KEEP=y
*drop null values.
  /NULL=DROP.

This is the resulting file:
groupid  id   score
1                1      92
1                0      56
1                1      45
2                1      41
2                1      75
2                0      22

2. Create a variable tscore, which contains the score of the target cases only.

*sort cases by group id, caseid, and score so that all cases in the group are together and the target subject (coded 0 under id) is the first case in each group.
sort cases by groupid(a) caseid(a).

The file looks like this:

groupid  id   score
1               0      56
1               1      92
1               1      45
2               0      22
2               1      75
2               1      41

*select only target subjects identified as 0 under caseid.
DO IF (caseid=0).
*compute a new variable with the scores of target cases.
COMPUTE tscore=score.
*select cases in each target's group, identified as 1 under caseid.
ELSE IF (caseid=1).
*compute the lag of the target score, which populates the variable tscore with the target's score for all cases in the target's group.
COMPUTE tscore = LAG(tscore).
END IF.
execute.

The above syntax yields this:
groupid  id   score      tscore
1        0              56              56
1        1              92              56
1        1              45              56
2        0             22       22
2        1             75               22
2        1             41               22

3. Calculate the difference between the target's score and the score of each person in the group.
*select only those cases in each target's group, coded 1 under caseid.
do if caseid=1.
*compute the difference between the target's score and the score of each case in the target's group.
compute diff=tscore-score.
end if.
execute.

The final file looks like this:
groupid  id   score      tscore      diff
1        0              56              56         .
1        1              92              56         -36
1        1              45              56          11
2        0             22       22          .
2        1             75               22         -53
2        1             41               22        -19


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of David Marso
Sent: Tuesday, April 03, 2012 5:38 AM
To: [hidden email]
Subject: Re: Calculating the difference between cases

Very difficult to imagine why you want to do this and/or what your full data actually look like since there are multiple subjects.
For the data you present here is a very simple solution.
OTOH you are obviously *NOT* providing adequate information to generate a general solution so this effort may prove to be utterly useless unless you rethink the more general issue (Cartesian product) and redesign your data structure.
data list free / gp score.
begin data
1 82 1 75 1 82 1 61 1 66 1 92 1 23 1 10 1 77 0 82 1 43 1 31 1 37 1 12 1 55 1
95
end data.
COMPUTE @ORDER@=$CASENUM.
SORT CASES BY GP.
DO IF $CASENUM=1.
+  COMPUTE #TARG=score.
ELSE.
+  COMPUTE DIFF=#TARG-SCORE.
END IF.
SORT CASES BY @ORDER@.
LIST.

Veena Nambiar wrote
>
> I have a data file that contains a variable id and score. Id is coded "0"
> for the target subject and "1" for subjects within the targets group.
> I would like to create a separate variable diff that takes the
> difference between the target subject's score and each of the other
> subjects in the group. Here is an example below. Is there a way to do this in SPSS?
>
>
> id
>
> score
>
> diff
>
> 1
>
> 82
>
> 0
>
> 1
>
> 75
>
> 7
>
> 1
>
> 82
>
> 0
>
> 1
>
> 61
>
> 21
>
> 1
>
> 66
>
> 16
>
> 1
>
> 92
>
> -10
>
> 1
>
> 23
>
> 59
>
> 1
>
> 10
>
> 72
>
> 1
>
> 77
>
> 5
>
> 0
>
> 82
>
>               .
>
> 1
>
> 43
>
> 39
>
> 1
>
> 31
>
> 51
>
> 1
>
> 37
>
> 45
>
> 1
>
> 12
>
> 70
>
> 1
>
> 55
>
> 27
>
> 1
>
> 95
>
> -13
>
>
>
> Veena Nambiar
>


--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Calculating-the-difference-between-cases-tp5601947p5614983.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

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

[Veena Nambiar]

Yes, you are correct Bruce - the syntax wouldn't work given the order of the sample - my actual file has the variables in the appropriate order so I can use "TO".

AGGREGATE works perfectly - a much cleaner solution than what I was using. I changed my syntax to use AGGREGATE.
Veena

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bruce Weaver
Sent: Tuesday, April 03, 2012 11:43 AM
To: [hidden email]
Subject: Re: Calculating the difference between cases

I think this gets you to the same result, but again, without the sorting.

* Read in sample data.
data list list / row id1 score1  id2 score2 .
begin data
1      1        92      1       41
2      0        56      1       75
3      1         45     0       22
end data.

* Given the order of the variables, "id1 TO id2"
* and "score1 TO score2" will not work--the variables
* must be listed individually. Perhaps the variables
* are ordered differently in the actual file with
* 1000 variables.

varstocases
 /make caseid from id1 id2
 /make score from score1 score2
 /INDEX=groupid (2)
 /NULL=DROP
.

IF (caseid EQ 0) tscore = score.
AGGREGATE
  /OUTFILE=* MODE=ADDVARIABLES OVERWRITE=YES
  /BREAK = groupid
  /tscore=FIRST(tscore).
IF (caseid EQ 1) DIFF = tscore - score.
LIST.

OUTPUT:
     row groupid   caseid    score   tscore     DIFF

    1.00      1      1.00    92.00    56.00   -36.00
    1.00      2      1.00    41.00    22.00   -19.00
    2.00      1       .00    56.00    56.00      .
    2.00      2      1.00    75.00    22.00   -53.00
    3.00      1      1.00    45.00    56.00    11.00
    3.00      2       .00    22.00    22.00      .

Number of cases read:  6    Number of cases listed:  6



Veena Nambiar wrote

>
> Thanks for the response David. I needed to compute these differences
> for use in another set of calculations that we do @ the state for reporting.
>
> With the help of another SPSS user, I think I have a solution. Here is
> a step by step of the solution.
>
> I have a data file with ids and scores. The ids and scores for group 1
> are in id1 and score1, group 2 are in id2 and score2, and so on. The
> zero in the id field represents a target case and the 1 represents
> cases in the target's group. My file structure is as follows:
>
> row id1 score1  id2 score2
> 1      1        92      1       41
> 2      0        56      1       75
> 3      1         45             0       22
>
> 1. Restructure the data file so that ids and scores are stacked on top
> of each other. Create a group id to differentiate between groups.
>
> varstocases
> *create a variable caseid that stacks data from id1 to id1000.
> /make caseid from id1 to id1000
> *create a variable score that stacks data from score1 to score1000.
> /make score from score1 score1000
> *create a variable that identifies the group id and score belong to.
>   /INDEX=groupid (1000)
> *keep other important variables from data file.
>   /KEEP=y
> *drop null values.
>   /NULL=DROP.
>
> This is the resulting file:
> groupid  id   score
> 1                1      92
> 1                0      56
> 1                1      45
> 2                1      41
> 2                1      75
> 2                0      22
>
> 2. Create a variable tscore, which contains the score of the target
> cases only.
>
> *sort cases by group id, caseid, and score so that all cases in the
> group are together and the target subject (coded 0 under id) is the
> first case in each group.
> sort cases by groupid(a) caseid(a).
>
> The file looks like this:
>
> groupid  id   score
> 1               0      56
> 1               1      92
> 1               1      45
> 2               0      22
> 2               1      75
> 2               1      41
>
> *select only target subjects identified as 0 under caseid.
> DO IF (caseid=0).
> *compute a new variable with the scores of target cases.
> COMPUTE tscore=score.
> *select cases in each target's group, identified as 1 under caseid.
> ELSE IF (caseid=1).
> *compute the lag of the target score, which populates the variable
> tscore with the target's score for all cases in the target's group.
> COMPUTE tscore = LAG(tscore).
> END IF.
> execute.
>
> The above syntax yields this:
> groupid  id   score      tscore
> 1        0              56              56
> 1        1              92              56
> 1        1              45              56
> 2        0             22       22
> 2        1             75               22
> 2        1             41               22
>
> 3. Calculate the difference between the target's score and the score
> of each person in the group.
> *select only those cases in each target's group, coded 1 under caseid.
> do if caseid=1.
> *compute the difference between the target's score and the score of
> each case in the target's group.
> compute diff=tscore-score.
> end if.
> execute.
>
> The final file looks like this:
> groupid  id   score      tscore      diff
> 1        0              56              56         .
> 1        1              92              56         -36
> 1        1              45              56          11
> 2        0             22       22          .
> 2        1             75               22         -53
> 2        1             41               22        -19
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:SPSSX-L@.UGA] On Behalf Of David
> Marso
> Sent: Tuesday, April 03, 2012 5:38 AM
> To: SPSSX-L@.UGA
> Subject: Re: Calculating the difference between cases
>
> Very difficult to imagine why you want to do this and/or what your
> full data actually look like since there are multiple subjects.
> For the data you present here is a very simple solution.
> OTOH you are obviously *NOT* providing adequate information to
> generate a general solution so this effort may prove to be utterly
> useless unless you rethink the more general issue (Cartesian product)
> and redesign your data structure.
> data list free / gp score.
> begin data
> 1 82 1 75 1 82 1 61 1 66 1 92 1 23 1 10 1 77 0 82 1 43 1 31 1 37 1 12
> 1 55
> 1
> 95
> end data.
> COMPUTE @ORDER@=$CASENUM.
> SORT CASES BY GP.
> DO IF $CASENUM=1.
> +  COMPUTE #TARG=score.
> ELSE.
> +  COMPUTE DIFF=#TARG-SCORE.
> END IF.
> SORT CASES BY @ORDER@.
> LIST.
>
> Veena Nambiar wrote
>>
>> I have a data file that contains a variable id and score. Id is coded "0"
>> for the target subject and "1" for subjects within the targets group.
>> I would like to create a separate variable diff that takes the
>> difference between the target subject's score and each of the other
>> subjects in the group. Here is an example below. Is there a way to do
>> this in SPSS?
>>
>>
>> id
>>
>> score
>>
>> diff
>>
>> 1
>>
>> 82
>>
>> 0
>>
>> 1
>>
>> 75
>>
>> 7
>>
>> 1
>>
>> 82
>>
>> 0
>>
>> 1
>>
>> 61
>>
>> 21
>>
>> 1
>>
>> 66
>>
>> 16
>>
>> 1
>>
>> 92
>>
>> -10
>>
>> 1
>>
>> 23
>>
>> 59
>>
>> 1
>>
>> 10
>>
>> 72
>>
>> 1
>>
>> 77
>>
>> 5
>>
>> 0
>>
>> 82
>>
>>               .
>>
>> 1
>>
>> 43
>>
>> 39
>>
>> 1
>>
>> 31
>>
>> 51
>>
>> 1
>>
>> 37
>>
>> 45
>>
>> 1
>>
>> 12
>>
>> 70
>>
>> 1
>>
>> 55
>>
>> 27
>>
>> 1
>>
>> 95
>>
>> -13
>>
>>
>>
>> Veena Nambiar
>>
>
>
> --
> View this message in context:
> http://spssx-discussion.1045642.n5.nabble.com/Calculating-the-differen
> ce-between-cases-tp5601947p5614983.html
> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> 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
>
> =====================
> 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.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Calculating-the-difference-between-cases-tp5601947p5616079.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

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