SPSS data recovering

[F. Gabarrot]

Hi list,

My advisor just asked me how to recover data from a mid-80's IBM version of SPSS.
I think his data date from 1984.

He already recovered a .txt file with his data in it, but the file is in a weird format. Data for each item are stuck together (i.e., 1423567436763345321). As the dataset contains more than 700 observations, and as each participant spent about 2 hours on the tasks, we are looking for an automatic way to recover this data.

Thank you.

Regards.

Fabrice.

[ViAnn Beadle]

If you know how the variables map to the positions, just use a data list
specifying position with the text as input. If you are uncomfortable with
syntax you can also use the Text Wizard from the Read Text Data item on the
Data menu to map variables to the data.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of F.
Gabarrot
Sent: Wednesday, October 17, 2007 1:16 AM
To: [hidden email]
Subject: SPSS data recovering

Hi list,

My advisor just asked me how to recover data from a mid-80's IBM version of
SPSS.
I think his data date from 1984.

He already recovered a .txt file with his data in it, but the file is in a
weird format. Data for each item are stuck together (i.e.,
1423567436763345321). As the dataset contains more than 700 observations,
and as each participant spent about 2 hours on the tasks, we are looking for
an automatic way to recover this data.

Thank you.

Regards.

Fabrice.
--
View this message in context:
http://www.nabble.com/SPSS-data-recovering-tf4638594.html#a13248279
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

[Marks, Jim]

You can import this into a spss file as long as you still have the file
layout-- i.e. you can split the numbers into individual variables.

There is no "automatic" solution. SPSS needs the file layout to create
variables. You will need to type out syntax or walk through the text
wizard to crete syntax that defines the layout needed to build the file.

You will have something like this:

GET DATA  /TYPE = TXT
 /FILE = 'yourlocation.txt'
 /DELCASE = LINE
 /ARRANGEMENT = FIXED
 /FIRSTCASE = 1
 /IMPORTCASE = ALL
 /VARIABLES =
 caldate 0-8 A
 id 9-10  F
 var1 11-11  A
 var2 12-16 F .

a couple of notes-- lifted from the help file
1) the variable list contains name startcol-endcol format
2) The columns start at 0 (not 1)
3) a single character still requires a start-end pair (see var1 above)
4) Formats are SPSS formats (F = Fixed (numeric) A = alpha (string)--
the width comes from the column width and other formats are allowed)
5) DELCASE indicates each line is a case-- if not true you need to know
how many variables make a case (and use /DELCASE = VARIABLES n
6) FIRSTCASE = 1 means every row is data-- adjust if you have header
row(s).

Good luck
--jim

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
F. Gabarrot
Sent: Wednesday, October 17, 2007 2:16 AM
To: [hidden email]
Subject: SPSS data recovering

Hi list,

My advisor just asked me how to recover data from a mid-80's IBM version
of SPSS.
I think his data date from 1984.

He already recovered a .txt file with his data in it, but the file is in
a weird format. Data for each item are stuck together (i.e.,
1423567436763345321). As the dataset contains more than 700
observations, and as each participant spent about 2 hours on the tasks,
we are looking for an automatic way to recover this data.

Thank you.

Regards.

Fabrice.
--
View this message in context:
http://www.nabble.com/SPSS-data-recovering-tf4638594.html#a13248279
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

[Antoon Smulders]

Hello Fabrice,

Maybe your advisor can also recover files wich include the syntax that
others suggested. They may bear other extensions then ".sps". I think it was
common these days to use extensions with ".inc" (though I'm not sure).
Also: you may find variable and value lables in there.

Good luck

-----Oorspronkelijk bericht-----
Van: SPSSX(r) Discussion [mailto:[hidden email]] Namens F.
Gabarrot
Verzonden: woensdag 17 oktober 2007 9:16
Aan: [hidden email]
Onderwerp: SPSS data recovering

Hi list,

My advisor just asked me how to recover data from a mid-80's IBM version of
SPSS.
I think his data date from 1984.

He already recovered a .txt file with his data in it, but the file is in a
weird format. Data for each item are stuck together (i.e.,
1423567436763345321). As the dataset contains more than 700 observations,
and as each participant spent about 2 hours on the tasks, we are looking for
an automatic way to recover this data.

Thank you.

Regards.

Fabrice.
--
View this message in context:
http://www.nabble.com/SPSS-data-recovering-tf4638594.html#a13248279
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