Thank you for the answer, Claire.

I came to a similar conclusion after a bit of googling, but you were faster.

Just wondering what the grand-masters will say about a checksum solution?

From what I know about checksums they are pretty reliable way to tell one set from the other... Any...

I wonder if checksum could be an alternative to a two-step process suggested by the article?

Something to the effect:

; WITH CTE

as (

SELECT worker_id,

CHECKSUM_AGG(CHECKSUM(rate_code)) as chk

FROM #worker_rate

GROUP BY worker_id...

Here's a good reference article.

Group Islands of Contiguous Dates (SQL Spackle)

http://www.sqlservercentral.com/articles/T-SQL/71550/