topic Re: Rolling up Averages in QlikView https://community.qlik.com/t5/QlikView/Rolling-up-Averages/m-p/801384#M282844 <HTML><HEAD></HEAD><BODY><P>Maybe something like this: avg(aggr( count(followers) * count(tweets),party))</P><P>If not, please post a qlikview document with enough data to calculate some correct results.</P></BODY></HTML> Fri, 13 Mar 2015 18:07:51 GMT Gysbert_Wassenaar 2015-03-13T18:07:51Z Rolling up Averages https://community.qlik.com/t5/QlikView/Rolling-up-Averages/m-p/801383#M282843 <HTML><HEAD></HEAD><BODY><P>Hey guys,</P><P></P><P>Got a problem around a two-stage calculation.</P><P></P><P>People and parties. How many followers do they have on Twitter? I get Follower Data at a <EM>Tweet level</EM> (important).</P><P></P><P>SO:</P><P></P><P>maxstring(followers)</P><P></P><P>gives me a correct calculation for <EM>one person's </EM>followers.</P><P></P><P>BUT:</P><P></P><P>How do I roll that up to a party level? At the moment it is being distorted by the number of tweets a user sends. For example, if person A has 100 followers and person B has 10 followers, the average should be 55. </P><P></P><P>BUT if person B tweets 10 times and person A tweets once, the average is 10*(10 followers) + 1*(100 followers) = 200 followers, divided by (10+1)=11 tweets. So the average comes out at 18. </P><P></P><P>Can anyone help?</P></BODY></HTML> Fri, 13 Mar 2015 16:25:50 GMT https://community.qlik.com/t5/QlikView/Rolling-up-Averages/m-p/801383#M282843 2015-03-13T16:25:50Z Re: Rolling up Averages https://community.qlik.com/t5/QlikView/Rolling-up-Averages/m-p/801384#M282844 <HTML><HEAD></HEAD><BODY><P>Maybe something like this: avg(aggr( count(followers) * count(tweets),party))</P><P>If not, please post a qlikview document with enough data to calculate some correct results.</P></BODY></HTML> Fri, 13 Mar 2015 18:07:51 GMT https://community.qlik.com/t5/QlikView/Rolling-up-Averages/m-p/801384#M282844 Gysbert_Wassenaar 2015-03-13T18:07:51Z