cancel
Showing results for
Did you mean:
Helper I

Calculate the average of values across several columns in same row?

For each observation (row), I am trying to add a calculated column that displays the average across seven columns and I am stuck. I guess what I am looking for is a "row average" that will ignore null values. Here's the structure:

For the first observation, PCODE 8411, the average of these seven columns, ignoring null values, is [(1+4+1+1)/4] = 1.75.

There are approximately 30 columns, so I only want to average these seven.

Any help would be great!

Thank you!

1 ACCEPTED SOLUTION

@Rymatt830

You could do two calculated columns, one for total between all 7 columns, and one that calculates the number of non blank columns.

Total Across Columns:=([a]+[b]+[c]+[d]+[e]+[f]+[g])

Non Blank Columns =IF(ISBLANK([a]),0,1)+IF(ISBLANK([b]),0,1)+IF(ISBLANK([c]),0,1)+IF(ISBLANK([d]),0,1)+IF(ISBLANK([e]),0,1)+IF(ISBLANK([f]),0,1)+IF(ISBLANK([g]),0,1)

Avg=[Total across columns]/[Non Blank Columns]

Please mark it as a solution or give a kudo if it works for you, otherwise let me know if you run into an issue and I'll do my best to assist.

Thanks,

Ryan Durkin

10 REPLIES 10
New Member

Greetings!

How can you average two columns generating random numbers at the same time. The average result is not correct in output. Is it possible or not?

Query

Field:           ITR                     PRIMER DFT MIN: Int((90-75+1)*Rnd([ITR])+75)      PRIMER DFT MAX: Int((90-5+1)*Rnd([ITR])+75)        AVE: ([PRIMER DFT MIN]+[PRIMER DFT MAX])/2

Run Output

ITR         PRIMER DFT MIN         PRIMER DFT MAX       AVE

1                     82                                   87                      86

2                     75                                   78                      76

3                     77                                   85                      81

Your response will be highly appreciated.

Thanks,

Alex

Community Champion

@Rymatt830 easy way to do is go to query editor after importing your data, make sure data type of seven columns is whole number.

Then select all seven columns, under Transform tab click Unpivot Columns. This will give you table as below.

Then under Transform tab click Group By and fill details as below and click OK.

Close&Apply and everytime you refresh you will get these steps applied.

Helper I

Thank you for the suggestion, and this solution would be ideal if I didn't have 30 additional columns in my query. Also, I have a lot of records in the query so duplicating it may be a bad idea.

Super User II

You can add a column in the query editor like this:

`Table.AddColumn(#"PreviousStep", "NNCount", each List.NonNullCount({[A],[B]}))`

Imke Feldmann (The BIccountant)

If you liked my solution, please give it a thumbs up. And if I did answer your question, please mark this post as a solution. Thanks!

Helper I

Hi Imke,

Thank you for the response. This looks like a great solution; but, can you explain it in more detail, please? Do the "A" and "B" reference column names? So, would the formula for my data be,

`Table.AddColumn(#"PreviousStep", "NNCount", each List.NonNullCount({[Bin (DSCR)],[Bin (NCF)], [Bin (OpEx)] ...}))`

Super User II

That's correct: List.NonNullCount counts the number of items in your list that are not null. So you don't have to write a condition for every column separately.

Imke Feldmann (The BIccountant)

If you liked my solution, please give it a thumbs up. And if I did answer your question, please mark this post as a solution. Thanks!

Helper I

I have the same question for averaging over 3 columns, some cells having blank values.  I am unable to replicate this formula, can you help?

Super User II

Hi @swragg ,

I believe you need a bit of a different approach here. Add a column like so:

List.Average ( List.Select ( Record.FieldValues ( [FirstColumn] & [SecondColumn] & [ThirdColumn] ), (x) => x <> null ) )

Imke Feldmann (The BIccountant)

If you liked my solution, please give it a thumbs up. And if I did answer your question, please mark this post as a solution. Thanks!

@Rymatt830

You could do two calculated columns, one for total between all 7 columns, and one that calculates the number of non blank columns.

Total Across Columns:=([a]+[b]+[c]+[d]+[e]+[f]+[g])

Non Blank Columns =IF(ISBLANK([a]),0,1)+IF(ISBLANK([b]),0,1)+IF(ISBLANK([c]),0,1)+IF(ISBLANK([d]),0,1)+IF(ISBLANK([e]),0,1)+IF(ISBLANK([f]),0,1)+IF(ISBLANK([g]),0,1)

Avg=[Total across columns]/[Non Blank Columns]

Please mark it as a solution or give a kudo if it works for you, otherwise let me know if you run into an issue and I'll do my best to assist.

Thanks,

Ryan Durkin

Helper I

Thanks, rdurkin, this is a great solution.

Announcements

Happy New Year from Power BI

This is a must watch for a message from Power BI!