I'd like to add a value and a label to every variable in my dataset. It's my understanding that the VALUE LABELS command erases values+labels already in the variables (and not included in the command), so I'd like not to do that.
Any ideas? Thank you!
Once you've prepared a table for variable labels and a table for value labels, you can use them to create a syntax that will put them in the dataset.
read the following text to get an example of how your variable label list should look like:
data list list/vr (a10) lb (a50).
begin data
"myvar1" "this is the label for var1"
"myvar2" "this is the label for var2"
"myvar3" "this is the label for var3"
end data.
Now you can run the following syntax:
string cmd1 (a100).
compute cmd1=concat("variable labels ", rtrim(vr), " '", rtrim(lb), "'.").
write out = "yourpath\do variable labels.sps" /cmd1.
exe.
This will create a new syntax called do variable labels.sps which contains the variable labels commands.
Now for value labels - start with the following example of how your table should look:
data list list/vr (a10) vl (f3) lb (a50).
begin data
"myvar1" 1 "cat"
"myvar1" 2 "dog"
"myvar1" 3 "horse"
"myvar2" 1 "bed"
"myvar2" 2 "table"
"myvar2" 3 "chair"
end data.
Once the table is prepared, use the following syntax:
string cmd2 (a100).
compute cmd2=concat("add value labels ", rtrim(vr), " ", string(vl,f3), " '", rtrim(lb), "'.").
write out = "yourpath\do value labels.sps" /cmd2.
exe.
Now you can use the new syntax created called do value labels.sps.
You can also look into insert command to run the two new syntaxes from your main syntax without opening them.
If you'd like to label your variables (for ex: v1 v2 v3) you can use:
VARIABLE LABELS
v1 'label for v1'
v2 'label for v2'
v3 'label for v3' .
If you'd like to add labels to the values associated with your variables, but NOT lose any existing value labels you can use:
ADD VALUE LABELS
v1
1 'label for v1=1'
2 'label for v1=2' /
v2
1 'label for v2=1'
2 'label for v2=2' /
v3
1 'label for v3=1'
2 'label for v3=2' .
If there were any previously existing labels, they will still persist (unless explicitly overwritten in your ADD VALUES command).
As user #user45392 pointed out you can use the ADD VALUE LABELS command.
In order to add the same value und label to every variable in your data set you can simply write the keyword ALL at the place of your variable list.
ADD VALUE LABELS ALL
N "new label".
N stands for the new added value.
Values are added only for numeric variables.
Related
can someone help me figure out how to select cases containing number 1, for example I coded nausea as side effect as 1 and was then noting it with other side effects as 1234 and now i wan to select all combinations with 1 but i cant figure out how. Or at least how to properly name what I am trying to achieve, since I am fairly new to spss so I can try to further search online.
I have tried variable = 1 and variable = 1 and neither worked and a few random commands that did not work either. I have put the variable as string and did not change anything either.
Once you change the variable into text you can use text search commands to find "1" within the text, like this for example:
compute nausea=(char.index(YourVariable,"1")>0).
char.index command searches for "1" in YourVariable - if it is there, it will output it's position in the text. If it isn't there, the output is 0. So nausea will get a value of 1 in all cases that contain "1" and will gat a value of 0 in all cases that don't.
NOTE - if you get as high as 10 in your numbers, this method will fail, as "10" contains "1". In order for any method to work here, you'd need to add a delimiter between the numbers when you record them, e.g. "1,3,8,17,22".
I have to select group of cases starting with specific numbers in multiple variables.
I am using this
CHAR.SUBSTR(variable1,1,x) ="y" | CHAR.SUBSTR(variable2,1,x) ="y" .............| CHAR.SUBSTR(variable40,1,x) ="y".
(x is number of character,y is characters I am choosing) the variables are named similar with just the number 1 to 40 being different
it works but problem is there are 40 variables and code is very length.
any elegant way to write it? like variable1 THRU variable 40?
You can loop through the variables and then select. Like this:
do repeat vr=variable1 to variable40.
if CHAR.SUBSTR(vr,1,1)="y" keep_this=1.
end repeat.
select if keep_this=1.
after running the loop, if any of the variables starts with "y" then the line will be marked with 1 in the variable keep_this. Now you can select only cases where keep_this=1.
If i have a variable in SPSS, with name (My_Variable), label (My Variable), values(1: Yes, 2: No) etc but without data (the column in data view is empty), i want to add data using syntax! For example, i want to add a participant in 1st row, who answered "Yes", so i want 1 to be added!!! How can i do it???
I found similar questions, but the solutions refers to creating A NEW SPSS window and add the values there! But i dont want this! I want to add data in an existing variable, without creating new SPSS file!
Apparently there is no way to directly add cases to an SPSS dataset through syntax.
But the following seems to me pretty close - you don't create new files but you create a new dataset and add it to your original.
Let's first create a small data to demonstrate on:
Data list list/ID (a5) var1 var2 var3 (3f2).
begin data
"first" 1 17 7
"secnd" 5 5 12
"third" 34 11 91
end data.
dataset name originalDataset.
So this is your original data. Now imaging that you want to add a new case to the data, with the ID value of "hello" and the number 42 in all the columns. This is what you do:
* creating the new case in a separate dataset.
Data list list/ID (a5) var1 var2 var3 (3f2).
begin data
"hello" 42 42 42
end data.
dataset name addition.
* going back to original dataset and adding the new case.
dataset activate originalDataset.
add files /file=* /file=addition.
exe.
dataset close addition.
You don't have to create data in the first data set. Just create the variables and define them however you want.
DATASET CLOSE ALL.
INPUT PROGRAM.
NUMERIC My_Variable (F1).
VARIABLE LABELS My_Variable "I want this!".
VALUE LABELS My_Variable 1 "Yes" 2 "No".
END FILE.
END INPUT PROGRAM.
DATASET NAME Empty.
DATA LIST FREE /My_Variable.
BEGIN DATA.
1 2
END DATA.
APPLY DICTIONARY /FROM Empty
/SOURCE VARIABLES=My_Variable
/TARGET VARIABLES=My_Variable
/VARINFO VALLABELS=REPLACE VARLABEL.
DATASET CLOSE Empty.
FREQUENCIES VARIABLES ALL.
I used DATASET but you could have save the empty file to disk.
See the APPLY DICTIONARY command for more details about how it works.
Using python you can add data with the cases.append() method
begin program.
import spss
spss.StartDataStep()
dataset = spss.Dataset()
dataset.cases.append([1])
spss.EndDataStep()
end program.
Say you have 3 variables, you can assign values to each by appending the list passed to the method
begin program.
spss.StartDataStep()
dataset = spss.Dataset()
dataset.cases.append([1,2,3])
spss.EndDataStep()
end program.
Would add a case wit value 1 in the first variable, value 2 in the second variable, 3 in the third variable.
Note: the method will only work within an open datastep.
Check out the ADD FILES command. You can also add cases with Python code.
I'm using SPSS modeler and I have a variable that the software recognizes as numeric. So the missing values are $null$. I want that the missing values of the variable are selectionable with '', as character.
So I would: or trasform the format of the variable from numeric to character or change only the missing values from $null$ to ''.
How can I fix?
thanks in advance
The best way to select null values in a numeric field is to use the #NULL() function from the Blanks and Null section of the Expression Builder.
For example, if you wanted to keep only the null values so that you could inspect them, you might use a Select node. Leave the radio button set as Include. Press the Expression Builder (calculator) button. Change the filter in the drop menu on the left side from General Functions to show Blanks and Null (press B 2 or 3 times). Double-click on #NULL(ITEM). Go to the right side and double-click on your numeric field name. Put a Table node at the end and run it.
Using Select #NULL in IBM SPSS Modeler
Another way to view just the null rows is to enter the #NULL(varname) function into the "Highlight records where" section of the Table dialog box.
"Highlight records where" dialog
When you run the table, any row that is true for this condition will be shown in red.
If you really need the variable to be a string, then use a Compute node to create a copy of this field under a new name and use the to_string() function in the Conversion section of the Expression Builder to change the type of the variable. Now you will be able to use the the Select node to grab "" as the missing value. Or you could use the Filler node to replace the column, but then you would not be able to compare before and after.
The dialog examples shown in this answer use this sample stream that is installed with your IBM SPSS Modeler software:
C:\Program Files\IBM\SPSS\Modeler\18.0\Demos\streams\featureselection.str
The easiest way to do it it's using the Fill node with the configurations:
A) FIELD
B) Condition = #NULL(#FIELD)
C) Replace by = ' '
This node will replace all $null$ for ' ' at the same variable chosen in option a.
I don't think you can customize how you visualize $nulls. (I know it's possible in SQL db though)
So I'd suggest that you work with the numbers and when you want to visualize or export the results, then turn the field to a string one then replace nulls:
Filled node > to_string(#FIELD)
Filler node > blank and nulls > #FIELD = ''
I have a data set in SPSS containing a sequence of six variables from which I have to create a new variable which should contain the last value present in the sequence. Let's say the data look like this: (the second row contains all missing values but represents a case to which I'll merge some other variables later, so I need this too.)
DATA LIST /V1 TO V6 1-6.
BEGIN DATA
423451
73453
929
0257
END DATA.
Now if I wish to generate a variable named lastscr which should have values 1, ., 3, 9, 7. Can anyone help me on how should I do it in SPSS? I could not find any clue about it. Thank you in advance for any help.
This can easily be done with the DO REPEAT command:
DO REPEAT Var = V1 TO V6.
IF NOT(SYSMIS(Var)) lastscr = Var.
END REPEAT.