DAX STDEV.P giving wrong value

[FergalK]

Hi

I have a formula in my Tabular SSAS model to calc. the Std. Deviation as follows: 

RMAQtyStdDev:=CALCULATE (STDEV.P(drct_rtns[rma_rcpt_qty]  ), ALLSELECTED(drct_rtns))

..but it is giving the wrong value (8) the correct value (when using Excel formula on my pivot table "=STDEV.P(B5:B101)") is 15.

I can't understand why DAX / SSAS Tabular is giving me the wrong value.

Appreciate any/all help!!!

Thanks,

Fergal

below are the values upon which the StdDev is being calculated:

9
8
13
24
30
3
4
18
21
92
10
15
23
10
16
10
12
5
9
6
9
10
4
12
3
2
6
8
3
4
3
3
1
2
3
3
3
1

[v-yuta-msft]

Hi FergalK,

Couldn't reproduce your issue. The STDEV.P works well on my side. 

  

Regards,

Jimmy Tao

[FergalK]

Hi Jimmy

Thanks for the reply, it does seem strange that it does not work correctly for me.

The only real difference is that you are using the formula in PowerBI and I am using it in SSAS tabular model but they should yield the same value.

I am wondering if it is not being calculated correctly in my pivot table because it does not recognise the filters correctly? But note thatr the Stdev value does actually change when I change a filter value in the pivot table, so it does seem like it acknowledges the filter values

Fergal

[FergalK]

Actually I think I may know what the issue is, it seems like the STDEV is including zero values too - how could I alter for formula to remove zeros from the STDEV calc?

thanks!

RMAQtyStdDev:=CALCULATE (STDEV.P(drct_rtns[rma_rcpt_qty]  ), ALLSELECTED()  )

[FergalK · Accepted solution]

I have rewritten my source query to combine both columns into a single column - issue was that zeros in other column were causing inciorrect average to be calculated