calculated variable

[Gene_Clark]

Okay, Kathleen, here^s another one. I have several rows of calculated variables of similar nature. The calculation of interest (problem?) is based on the value of another calculated value and a total field. All but one of the calculated variables is correct, but one is not. I even print on the same line the values of the two variables on which the problem child is dependent (it^s just one subtracted from the other), and the two independent variables are okay, but the dependent variable does not equal the difference of the others, as it should.____The variables are printed on the group band after 4 multiscans. Like I say, all the other 12 variables on the line are okay.____Is there something I need to know about the order of calculation for variables? Or when values are reset?

[Gene_Clark]

In the same problem, when I get a valid value of zero for the (normally miscalculated) variable, the function STR(variable name,5) gives me a blank instead of a zero. All the other similar valid zeroes give me a STR output of "0" like they should.____HELP!____p.s. I just updated my version to 8.1.027, but that made no difference.

[kfleming]

It is definitely because of the multiscan and is probably compounded because you have a calc that references a total.____But to give you a definitive answer on what is happening, I would need to see a copy of your report and data files.______Kathleen__R&R Support

[Gene_Clark]

Problem finally solved. R&R works like a charm, but my data base doesn^t. The problem came from empty fields in my data base. One of the root-level calculations involves the log function, which was trying to take the log of zero. Even though those records were not being printed, because they did not pass other selection criteria, they were getting added to the total field.____Case closed. Embarassed, but happy.