I am running custom tables that need to produce a percentage of time spent to a particular channel as a percentage of total time spent. Both values exist as scale variables. Could you advise on this please?

You can do this by using a computed category. Here is an example using the employee data.sav file shipped with Statistics. If you have a total in the table, you could use that as the denominator.

CTABLES

/PCOMPUTE &cat1 = EXPR(['f'] / (['f'] + ['m']) * 100)

/PPROPERTIES &cat1 LABEL = " percent female" FORMAT=MEAN PCT40.0 HIDESOURCECATS=NO

/TABLE gender > (salary [MEAN] + salbegin [MEAN]) BY minority

/CATEGORIES VARIABLES=gender ['f', 'm', &cat1, OTHERNM] EMPTY=INCLUDE

/CATEGORIES VARIABLES=minority ORDER=A KEY=VALUE EMPTY=INCLUDE MISSING=EXCLUDE.

p.s. Depending on your question, you might choose either MEAN or SUM as the statistic.

Thanks but I can't see how this would work where total time is stored as a scale variable and programme time as a separate scale variable (stored as hours & decimals of hours). I would need to compute e.g. (ProgTime/TotalTime)*100 and I understand pcompute doesn't allow references to 2 separate variables?​

If total time is just the sum of the program times, you could use a total. If not, you could use the STATS TABLE CALC extension command to add this in. If you send me an spv file with the table with both variables in it, I'll show you how to do that calculation. (jkpeckSupport Member)