Getting multiple identical rows using group by in SQL Server

Question

The code below returns:

DonorCenterDescription  Product         Pledge  
-----------------------------------------------
Avon                    Double Red      14
Avon                    Platelet        6
Avon                    Platelet        36
Avon                    Platelet        18
Avon                    Platelet        40
Avon                    Whole Blood     4
Avon                    Whole Blood     12
Avon                    Whole Blood     192

But I want only one row for each product (Double Red, Platelet, WholeBlood).

I know that I'm getting the extra rows because the table that has the product info has multiple rows for each product:

Select
    L.DonorCenterDescription,
    P.Product,
    Count(I.DonorID) * P.Frequency As Pledge
From
    HemaConnectDailyFiles.dbo.IPledgeCallCodes I 
Inner Join
    DonorCenterCodeAssignedToDonorsMaster C On I.DonorID = C.DonorID 
Inner Join
    MarketingTempTables.dbo.FixedDCDesc L On C.DonorCenterCode = L.DonorCenterCode 
Inner Join
    MarketingTempTables.dbo.PledgeKey2014 P On I.IPICCallCodes = P.Code
Where
    Left(I.IPICCallCodes, 4) = '2014'
Group By 
    L.DonorCenterDescription, P.Product, P.Frequency
Order By
    L.DonorCenterDescription, P.Product

But I don't know how to get to:

Avon    Double Red      14
Avon    Platelet        94
Avon    Whole Blood     208

Answer 1

You need some clarifying logic...these are not duplicate rows, they are rows that are being differentiated when you have P.Frequency in the group by statement.

When you have 5 rows for AVON platelet, it is recieving multiple values for pledge. You need logic to determine what value for p.frequency to use.

Edit : didn't see your end results. 2 alterations:

change pledge line to a full aggregate :

sum(Count(I.DonorID) * P.Frequency)  as pledge 

edit in : this may be what you are looking for...unsure if the sum(count() * field) is valid.

sum(I.DonorID * P.Frequency)

-- (I think this will achieve what you want...if not you'll have to clarify the logic in calculating pledge by the 2 columns DonorCenterDescription and Product

remove frequency from the group by statement

-- in the end, this is why you are getting the duplicate rows...this code is explicitly stating that L.DonorCenterDescription, P.Product, P.Frequency is a unique combination not just L.DonorCenterDescription, P.Product like you want