|
Hi,
is it possible to create a ctable with subtotals and show the significance for subtotals within the normal table for significances? I can use CATEGORIES=ALLVISIBLE or CATEGORIES=SUBTOTALS but is threre a possibility to get the result in one table? Thanks in advance! -- Sent from: http://spssx-discussion.1045642.n5.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 |
|
Yes (as of V24). You specify this on the COMPARETEST subcommand. Here is an example using employee data.sav. CTABLES /TABLE educ BY jobcat /CATEGORIES VARIABLES=educ [8, 12, SUBTOTAL='Subtotal1', 14, 15, 16, SUBTOTAL='Subtotal2', 17, 18, 19, 20, 21, SUBTOTAL='Subtotal3', OTHERNM] EMPTY=INCLUDE POSITION=AFTER /CATEGORIES VARIABLES=jobcat ORDER=A KEY=VALUE EMPTY=INCLUDE MISSING=EXCLUDE /COMPARETEST TYPE=PROP ALPHA=0.05 ADJUST=BONFERRONI ORIGIN=COLUMN INCLUDEMRSETS=YES CATEGORIES=SUBTOTALS MERGE=YES STYLE=SIMPLE SHOWSIG=NO. On Thu, May 17, 2018 at 2:26 AM, emma78 <[hidden email]> wrote: Hi, |
|
Hi Jon,
thanks for your answer. I know this option but now I have only the significances for the subtotals, what I want is all significances together: for the subtotals and the 'normal' values, is that posible too? -- Sent from: http://spssx-discussion.1045642.n5.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 |
|
You cannot directly get the test statistic both for the subtotal and the categories that are subtotaled. The subtotal significance calculation replaces the calculation for its individual categories. However, you can stack a copy of the variable above or below the table with subtotals and exclude from it the categories that are in the subtotal and thus get both tests. You might think of creating the subtotal as a computed category, but computed categories do not participate in significance tests. On Fri, May 18, 2018 at 1:47 AM, emma78 <[hidden email]> wrote: Hi Jon, |
| Free forum by Nabble | Edit this page |
