![]() Springtime girl time? I don’t like the “ Column %” headings, but I can set a different label (I set it to empty below), and different % formats as well. Instead of overriding, it is easier just to set the measurement level once and for all: VARIABLE LEVEL Sex quarter region (NOMINAL). But we can override the dictionary information by adding to make the variable categorical, or to make it scalar: CTABLES That’s just what happened: we got means here, with no decimals as the variable has format F8.0. ![]() The docs point out that you get rows for the categories if your variable is categorical, while if it is scalar it gets summarized. I want to see if the proportions of little boys and little girls are stable through time and space. Of course if we were to use these variables in a regression we would re-code them back to 0-1. Thing is, in pspp the categories of a variable are displayed in the order of the values, and we somehow expect a Yes column (or row) before a No one. See how the SET TNUMBERS trick comes in handy for the following RECODE command? The last two lines of syntax change all the 0-1 variables to 1-2 (or 2-1 actually). VALUE LABELS Twins Cesarean TO Rest 1 "yes" 2 "no". RECODE Twins Cesarean TO Rest (1 = 1) (0 = 2). VALUE LABELS region 1 "France" 2 "other Europe" 3 "other". Presentation 1 "cephalic" 2 "pelvic" 3 "other". Also, I would like to reduce the time scale and the geography. We need to do a little housekeeping, for only Country has value labels. Ok, let’s open the file and see what’s in there: SET PRINTBACK = ON. Once myoutput.html is open in my browser, F5 (refresh) shows me the incremental results. Of course, after the first run the up arrow does the job. A copy of the file (zip) is here.Īs always, I create my pspp syntax incrementally in a text editor: I add one or more commands, save, and run pspp in a terminal: $ pspp -o myoutput.html mysyntax.sps In the end I borrowed the birth dataset from R’s catdata package (saving it as a. Let’s give it a try!įinding public datasets with categorical variables isn’t easy. And you can give your tables the looks you like.īeing an old-timer used to the syntax of SPSS TABLES, I will have to learn the CTABLES syntax just like you. You can make row x column (x layer) tables of categorical variables and summarize scale variables in the table cells too. This man – Ben Pfaff – has brought PSPP from scratch to the full-fledged statistical software it is today, all by himself!Īlmost all my life I have worked with cross-tabulations in SPSS, and I know for sure many (rich) companies and professionals bought SPSS mainly for its TABLES module, of which CTABLES is a newer version.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |