Re: New with pspp

[Alan Mead]

Hi Ellen, My answers and inline below.  -Alan

On 5/25/2016 10:34 AM, Ellen Bowers wrote:

I have Windows 10.  I have been able to upload a file of 117 subjects with 112 Columns from Excel 14 saved in .csv  -- there will be nearly 300 subjects ultimately.  For the first part of the columns, there are 37 statements re feelings of the subjects' mother image and father image expressed in a 5-point Likert Scale.  I have not attempted any statistics.  I did not save anything in pspp yet.

You probably want to save the file as a SPSS "system file" (which is the default).  The filename would end in .SAV (Windows 10 probably hides the 'SAV extension, but it will refer to it as a "PSPP file" or something like that).

First -- Once I upload and save in pspp, do I have to begin a new Excel file?

What you should have done is (1) save the Excel file as an Excel file, (2) then created a copy as a CSV file.  But if you skipped step 1, Excel will read the CSV file... So you haven't "lost" the ability to have the original file in Excel.

However, PSPP doesn't care about what you do in Excel ... Once you have your data in a PSPP/SPSS system file, it doesn't matter what happens to the Excel file.

Can I add the new Subjects to the open file ?

Yes. Like SPSS, PSPP has the capability to add new subjects and new variables. You said that 37 of 112 columns are Likert responses, but if any of the remaining 75 columns are character/string data then you might run into a situation where you have to manually adjust all the string variables in both the old and new datasets to have the same widths.  For example, if there is an EMAIL column and it happens to be 42 characters in the current dataset but it happens to be 36 characters in the next wave of data, then you will need to manually adjust the width of the new EMAIL column from 36 to 42. You do this manual adjustment in PSPP the same way as SPSS (or ask and we'll explain).

Also, I hope you picked good names for the columns and note that the names must match across any old and new dataset that you want to combine (or you'll have to rename one set).

And it's a really good idea to label the variables and the values. You can do this manually in PSPP in the same way that you do it in SPSS, or you can use syntax. For example, if I'm going to perform item analyses or factor analyze Likert data (or if I just want to be sure that the dataset will be clear in the future when I've forgotten which item is X23), then I label each item with the text of the Likert statement:

variable value x1 'I feel that my mother was warm and nurturing'
  /x1 "To this day, I don't get along with my mother."
  / ... put the rest of the items here... you can use double quotes if an item has an embeded single quote
 /X37 "My father was there when I needed him."
.
execute.

(Notice that there is a single period after the last label for the last variable and then the "execute." command.)

value labels sex 1 'male' 2 'female'
  /employment 1 'student' 2 'employed part-time' 3 'employed full time' 4 'retired'
.
execute.

(Again, a single period ends the value labels command and an execute is needed to make it take effect.)

To use syntax (which is occasionally easier and not that hard), you first open up your dataset.  Then in that window, choose "File > New... > Syntax" and a new syntax window opens up.  Then paste or type commands like those above.  When you're ready, choose "Run > All" to run the syntax and apply the labels. Either you'll see the labels (in the data window, click "variable view" on the bottom left) or else you'll get error in the output window.

I hope this reaches the proper area.

Ellen Bowers

_______________________________________________
Pspp-users mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/pspp-users

-- 

Alan D. Mead, Ph.D.
President, Talent Algorithms Inc.

science + technology = better workers

+815.588.3846 (Office)
+267.334.4143 (Mobile)

http://www.alanmead.org

I've... seen things you people wouldn't believe...
functions on fire in a copy of Orion.
I watched C-Sharp glitter in the dark near a programmable gate.
All those moments will be lost in time, like Ruby... on... Rails... Time for Pi.

          --"The Register" user Alister, applying the famous 
            "Blade Runner" speech to software development