0
votes

Looking at a continuous variable under four different categorical coded groups. Attempting to run proc power with a onewayanova test but I can't seem to make it account for multiple standard deviations.

Looking to try and see if this is possible.

Title "Find Power for ANOVA"
proc power; 
  onewayanova test = overall
  groupmeans = 1814120 | 1344300 | 953580 | 1352900
  stddev = 1879922.09 | 969317.15 | 441433.68 | 970670.65
  npergroup = 3 | 4 |5 | 4
   power = .; 
run;

This gives me: 180 ERROR 180-322: Statement is not valid or it is used out of proper order.

1

1 Answers

0
votes

stddev and npergroup use number lists, whereas groupmeans use grouped-number-lists. The syntax between the two are different.

proc power; 
    onewayanova 
        test       = overall
        groupmeans = 1814120 | 1344300 | 953580 | 1352900
        stddev     = 1879922.09 969317.15 441433.68 970670.65
        npergroup  = 3 4 5 4
        power      = .
    ;
run;

enter image description here