[thelist] ASP 2.0 Functions

Luther, Ron Ron.Luther at hp.com
Mon Jul 22 07:47:01 CDT 2002


Hi Paul,


It's kinda 'old school'.

It's an attempt to avoid 'systematic bias'.  If you _always_ round your
numbers up (or down) you end up overstating (or understating) your
estimate of the population mean, for example.

{Of course, you'd need a pretty darn precise result set for this
difference to be 'statistically significant'.
But it is theoretically possible.}

I'm not sure when people stopped using the 'x.5 to nearest even integer'
rounding convention. I think somewhere along the line the education
system must have decided it was "too hard" and stopped teaching it.

HTH,

RonL.

-----Original Message-----
From: Paul Cowan [mailto:evolt at funkwit.com]

I'm fairly sure that round(), in VBScript, rounds to the nearest
even integer (e.g. answer mod 2 = 0) when the answer ends exactly in .5.

So:
1.5 -> 2
2.5 -> 2

I don't know why that is.



More information about the thelist mailing list