This post was updated on .
I often need to calculate a share of value among different groups in the dataset.
For this I extract the SUM of the values for each of the different groups and then need to take the percentage of each of these total as a % of the grand total (Excel has this functionality with the use of Pivot tables). But I don't think this can be done directly with CTABLES? Below is an alternative but I have had to WEIGHT the data as a work around. Is there a more direct way of producing this without needing to compute the additional variables and weights? |
Here is the code and picture I indeed to supply before accidentally posting message:
It's the red cells I am trying to derive directly from the data. get file="C:\Program Files\IBM\SPSS\Statistics\22\Samples\English\Employee data.sav". ctables /table (gender)[c] by salary[s][validn'N' sum mean] /categories variables=gender total=yes /titles title="Table 1". aggregate outfile=* mode=addvariables /break=gender /SalaryByGenderTotal=Sum(salary) /GenderSplitBase=N. compute Weight=SalaryByGenderTotal/GenderSplitBase. weight by Weight. ctables /table (gender)[c][ucount comma8.0 ucolpct count comma8.0 colpct ] /categories variables=gender total=yes /titles title="Table 2". spssinc modify tables subtype="custom table" select=-1 dimension= columns level = -1 process = preceding /styles textstyle=bolditalic textcolor=255 0 0. |
In reply to this post by Jignesh Sutar
How about something like this.
CTABLES /TABLE jobcat [C] BY salary [S][SUM, COLPCT.SUM]. Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] phone: 720-342-5621 From: Jignesh Sutar <[hidden email]> To: [hidden email] Date: 01/09/2015 11:58 AM Subject: [SPSSX-L] Percentage of SUM statistics from CTABLES Sent by: "SPSSX(r) Discussion" <[hidden email]> I often need to calculate a share of value among different groups in the dataset. For this I extract the SUM of the values for each of the different groups and then need to take the percentage of each of these total as a % of the grand total (Excel has this functionality with the use of Pivot tables). But I don't think this can be done directly with CTABLES? Below if an alternative but I have had to WEIGHT the data as a work around. Is there a more direct way of producing this without needing to compute the variables and weights? -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Percentage-of-SUM-statistics-from-CTABLES-tp5728399.html 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 |
Free forum by Nabble | Edit this page |