The short answer is no... The problem that you run up against with this type
of rounding is that there is no definitieve way to assess the least
significant digit. In your example with 1200 you indicate that the result
should be rounded to the nearest 100. Your assumption is that 1200 is only
accurate to the nearest 100 units of measure. It could actually be accurate
to the nearest single unit. 1201 would be accurate to the nearest single
unit. To be truely accurate you either need to know the the accuracy of the
least significant digit, or if you have a large enough data set of repeated
mesurements that you can use stats to determine the least significant digit.
--
HTH...

Jim Thomlinson


"bgarey" wrote:

When adding measurements, the final answer needs to be rounded off to the
least significant place value of the measurements:

The answer to 12.3 + 312.56 should be rounded off to the .10's place.
The answer to 1200 + 346 should be rounded off to the 100's place.

When multiply and dividing, the final answer should be rounded off to the
least number of significant digits.

The answer to 245 x 1.2 should be rounded to 2 digits and adding zero's for
place value if necessasry.


"Rick Rothstein" wrote:

I am not finding your question to be clear.. can you provide a little more
detail as to what you are looking for along with 3 or 4 examples of what you
have and what you expect them to look like after being rounded?

--
Rick (MVP - Excel)


"bgarey" wrote in message
...
Can I use excel to round operations with measurements to significant
figures