About matching data

[fevziesen]

Hi all,

I have a 2 million records dataset coded by some symbols. This symbols are repeated continuously.

x     y    
sy   1
df   2
rt   5
sy  6


Also, 2000 symbols are defined and I have a separate symbol dataset too (in another spss file). In this dataset symbols are linked to some variables.

sy     4
df     3
rt     2

I would like to match the second dataset to 1st. Like this:

sy  1   4
df  2   3
rt  5   2
sy 6   4...

sy is repeated two times but still 4 is assigned for sy..

[David Marso]

see SORT CASES
see MATCH FILES (read up on TABLE and BY subcommands).

fevziesen wrote

Hi all,

I have a 2 million records dataset coded by some symbols. This symbols are repeated continuously.

x     y    
sy   1
df   2
rt   5
sy  6


Also, 2000 symbols are defined and I have a separate symbol dataset too (in another spss file). In this dataset symbols are linked to some variables.

sy     4
df     3
rt     2

I would like to match the second dataset to 1st. Like this:

sy  1   4
df  2   3
rt  5   2
sy 6   4...

sy is repeated two times but still 4 is assigned for sy..

[fevziesen]

a warning message : variables are not the same length.

[Bruce Weaver]

If your version of SPSS is new enough, use ALTER TYPE to standardize the formatting of string variables common to the two files before using MATCH FILES.  

fevziesen wrote

a warning message : variables are not the same length.

[fevziesen]

tried but still not working. I want to repeat the problem again.

Basically it is an assignment problem. when it sees `x` in dataset1, it assigns matching value `y` from dataset2 by creating a new column.

[Bruce Weaver]

Please post your syntax and some small sample data sets that illustrate the problem.  (You can use DATA LIST to create the sample data sets -- there are plenty of examples in the archives of this list, and in the Command Syntax Reference manual).

fevziesen wrote

tried but still not working. I want to repeat the problem again.

Basically it is an assignment problem. when it sees `x` in dataset1, it assigns matching value `y` from dataset2 by creating a new column.

[fevziesen]

here is a illustration:

[David Marso]

As previously requested:
Please post SYNTAX and any error messages AFTER REVIEWING the full set of messages previously posted in this thread!!!

[fevziesen]

DATASET ACTIVATE Set1.
SORT CASES BY tick(A).
 
DATASET ACTIVATE Set2.
SORT CASES BY tick(A).
 
DATASET ACTIVATE Set1.
MATCH FILES /FILE=*
  /TABLE='Set2'
  /BY tick.
EXECUTE.

It did not work. Because variables are not the same length and they are strings..

I have illustrated above.

[David Marso]

"variables are not the same length and they are strings..."
----
Someone mentioned ALTER TYPE in the thread (Please review)!
Try again please!
--

fevziesen wrote

DATASET ACTIVATE Set1.
SORT CASES BY tick(A).
 
DATASET ACTIVATE Set2.
SORT CASES BY tick(A).
 
DATASET ACTIVATE Set1.
MATCH FILES /FILE=*
  /TABLE='Set2'
  /BY tick.
EXECUTE.

It did not work. Because variables are not the same length and they are strings..

I have illustrated above.

[Art Kendall]

A David said you might want to use ALTER TYPE.  When the match failed it should have given the kind of message that this syntax does.
{note that lunch was constructed as if it were data filled in by people writing in what they ate.}
Open a new instance of SPSS. Past the syntax below into a syntax window. Run it. Eyeball the output.

data list list /fruit(A10).
begin data
Apple
apple
apples
orange
Orange
Grape
ggrape
Plum
plum
end data.
dataset name lunch.
sort cases by fruit.


data list list /fruit (A12) color (a12).
begin data
plum purple
apple red
grape purple
orange orange
end data.
dataset name palette.
sort cases by fruit.
match files file= lunch /table = palette /by fruit.

dataset activate lunch.
alter type fruit (a=a12).
match files file= lunch /table = palette /by fruit.
dataset name matched.
list.



On 1/22/2013 6:30 PM, David Marso wrote:

"variables are not the same length and they are strings..."
----
Someone mentioned ALTER TYPE in the thread (Please review)!
Try again please!
--

fevziesen wrote

DATASET ACTIVATE Set1.
SORT CASES BY tick(A).

DATASET ACTIVATE Set2.
SORT CASES BY tick(A).

DATASET ACTIVATE Set1.
MATCH FILES /FILE=*
  /TABLE='Set2'
  /BY tick.
EXECUTE.

It did not work. Because variables are not the same length and they are
strings..

I have illustrated above.

-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/About-matching-data-tp5717464p5717599.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