pspp-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

PSPP: means and sums, by group, using the menus


From: someone
Subject: PSPP: means and sums, by group, using the menus
Date: Sat, 7 May 2022 16:14:33 +0000 (UTC)

Thanks very much again for responding.

I wanted to check to make sure I understand. Are you saying that for my data set, I'd have to use command line syntax, and not the menus in order to get means by group, for my particular data set?

I also made a very simple data set, and in this data set, in the menu system, when I open analyze / compare means / means. I also don't see the grouping variable in the means menu. This is the simple data set. Two columns (group and variable) and the columns separated by comma. 

group,variable
one,1
two,3
one,5
two,3
three,2
four,5
three,3
four,1
one,5


Sorry for causing trouble.

Thanks

Gene


-----Original Message-----
From: Dr. Oliver Walter <o.walter@psychometrie-online.de>
To: pspp-users@gnu.org <pspp-users@gnu.org>
Sent: Sat, May 7, 2022 11:44 am
Subject: Re: PSPP: means and sums, by group, using the menus

You have to use the command with another variable and  change the variable name in the PSPP command. It works. I know it, because I did it.
If you need training on how to analyze your data with PSPP, then I recommend you to take a course on data analysis with PSPP or IBM SPSS Statistics or you can search for PSPP tutorials e.g. on youtube.

Kind regards,

Dr. Oliver Walter

Am 07.05.2022 um 14:33 schrieb someone:
Hi Dr. Walter.

Thanks very much for responding. 

Unfortunately, two issues:

I saw that method, however for some reason it doesn't work in my data set. When I bring up the menu for mean, Arearegion doesn't appear in the list of variables. It -is- in the data set, but it doesn't appear in the list in the means box.

Also, I -can- get means when I do one way anova. However, in the descriptive box, two regions East AsiaSoutheast Asia and South Asia appear in scientific notation. When I save this as html and bring it into a spreadsheet, I get the first two digits and then a bunch of zeros. 

My data set seems to have a lot of things that  make it difficult.  :(  


Thanks

Gene


-----Original Message-----
From: Dr. Oliver Walter <o.walter@psychometrie-online.de>
To: pspp-users@gnu.org
Sent: Sat, May 7, 2022 1:43 am
Subject: Re: means and sums, by group, using the menus

Hello,
you can get sums by the command that compare "means" in the "compare means" menu
Analyze > compare means > means .
How you use this command is shown in this video: https://www.youtube.com/watch?v=XdC8vWiEKb4
Of course, if you calculated means you only have to multiply the mean of a group times the sample size N in this group to get the sum. You know: mean = sum/N. Hence, sum = mean * N.
But it can also be done by PSPP. If you click on "paste" in the "compare means" menu, then another window will open which contains the command:
MEANS TABLES =  population
    BY AreaRegion .
Then you delete the last full stop and add
    / CELLS = COUNT MEAN STDDEV SUM .
as the last line. The whole command will then look like
MEANS TABLES =  population
    BY AreaRegion
    / CELLS = COUNT MEAN STDDEV SUM .
You select this whole command and execute it.
If you need training on how to analyze your data with PSPP, then I recommend you to take a course on data analysis with PSPP or IBM SPSS Statistics or you can search for PSPP tutorials e.g. on youtube.

Kind regards,

Dr. Oliver Walter
Am 07.05.2022 um 01:15 schrieb someone:
Hi all

Another question, again using the data set below. Well, a version after I removed all non letter and number characters.

shortdataset.csv

I want to get means and sums of population by arearegion. I can't quite figure out how to do that.

I can get means if I do analyze / compare means / one way anova. But that doesn't give me sums. And it seems like there should be a simpler way to get means of a variable by group. Is there, and I'm just not seeing it? 

I want to do this in the drop down menu system, not by command line. 

Thanks

Gene


reply via email to

[Prev in Thread] Current Thread [Next in Thread]