SUM Column but Average Values Based on Duplicates in Another Column

[Anonymous]

A	B
0.10	4945708906
0.86	4945708908
0.17	4945708909
0.29	6041671331
0.22	6041671331
1.51	6041671334
1.67	6041671334
0.22	6041671337

I would like to sum the values in column A, but if there are duplicates in column B, I want to average the values of the duplicates then include that average in the sum.

So instead of adding 0.29 and 0.22 as part of the sum of column A, first gather their average = .255 and use that in the total sum instead. Then do the same for 1.51 and 1.67, etc.

[Anonymous]

A	B	C
0.1	4945708906	30
0.86	4945708908	150
0.17	4945708909	0
0.29	6041671331	90
0.22	6041671331	60
1.51	6041671334	200
1.67	6041671334	220
0.22	6041671337	0

Both solutions work, but what if I have additional columns that I want to SUM

Also this is all being done in a UNION(SELECTCOLUMNS( and if possible would like it achieved under "A"

    SELECTCOLUMNS (
        'Table1',
	"A", 'Table1'[A],
        "D", 'Table1'[D],
        "E", 'Table1'[E],
        "C", 'Table1'[C]/60),

 but maybe thats not feasible, or am setting this up wrong.

Nevermind, the Summarize expression works, I just have to keep adding to it and rebuild my table.

[Ruksuro]

I like to use:

Average Measure = AVERAGEX(Sheet2,Sheet2[A])

Then

Sum Average Measure by B = sumx( DISTINCT(Sheet2[B]), [Average Measure])

[jthomson · Accepted solution]

Should be easy enough with an intermediate summary table, something like:

summarize(yourtablename,[column b],"column a averaged",average([column a])

You can then sum up your new "column a averaged" column