To make this work in SQL 2000

Change the first Query to:

[Code]

-- Calculate prevalence of column name

SELECT COLUMN_NAME, CONVERT(DECIMAL(12,2), Count(*)* 100.0/t.total) as [%]

INTO #Prevalence

FROM INFORMATION_SCHEMA.COLUMNS

CROSS JOIN

(SELECT Count(*) as total FROM INFORMATION_SCHEMA.COLUMNS) t

GROUP BY COLUMN_NAME, t.total

-- Do the columns differ on datatype across the schemas and tables?

[/code]

--

JimFive