question

Every programming language except VB uses 1 byte to store bit information, setting the rightmost bit to 1 and other 7 bits to 0 for true, and all 8 bits to 0 for false. This results in evaluation of true to 1 and false to 0. In VB (or VBA), the way the boolean values (close equivalent of bit in SQL Server) are handled is different. In these languages, 2 bytes are used, setting all 16 bits (2 times 8 = 16) to 1 for true, and all 16 bits to 0 for false. If you take 2 bytes and set every bit to 1 then the resulting number (in hexadecimal notation) is 0xFFFF. This number evaluates to (surprise) -1. You can try this in T-SQL to check it: select cast(0xffff as smallint) TrueInVBA, cast(0x0000 as smallint) FalseInVBA; -- the above returns this: TrueInVBA FalseInVBA --------- ---------- -1 0 The reason I chose smallint was because it takes 2 bytes, so the script above mimics the VBA behaviour. One way to address your issue is to replace the bit type column with something like this: cast(ThatBitColumn as int) as ThatBitColumn In other words, do the conversion part before, not after. This way, the column will appear as a whole number (integer) to Excel and you can map it. Hope this helps. Oleg

Thank you for that excellent explanation Oleg, I understand my own database a little better now.