How can I apply different weighting to dashboard with multiple datasets | XM Community
Skip to main content
Solved

How can I apply different weighting to dashboard with multiple datasets


Forum|alt.badge.img+3
  • Level 3 ●●●
  • 67 replies

Hello guys, 

Happy new year to everyone. 

I have a question, I am currently working on creating a dashboard with 2 different datasets (Survey 2022 and Survey 2023). I have applied a weighting to my 2022 dataset and I would like to apply another weighting scheme to my 2023 dataset. 

 

To give you context in this surveu I have multiple embedded data on which I apply the weighting :

  • Q_Language (FR, IT, PT, ES or PT)
  • Type (buyer or visitor)
  • And I also want to apply certain weighting to respondents ages.

It would give something like that within the same weighting scheme  :

 

Can someone please help with this ? 

Best answer by InessaG

The way I do it is create a Cell-based weighting scheme

and select “Data needs to be weighted as waves” checkbox that appears once you select that option.

You will then selecte the Date field you want to use - typically it’s the “Survey Metadata - Date Ended”, and you’ll want to break it out by year.

 

 

It will then allow you to define the weights for each year separately. Just add all the variables you want to be included in the weighting, and define the separate weights for each year.

 

View original

13 replies

InessaG
Level 5 ●●●●●
Forum|alt.badge.img+19
  • Level 5 ●●●●●
  • 410 replies
  • January 11, 2024

Just add a year to your model, and you can then define the weight model for each year.


Forum|alt.badge.img+3
  • Author
  • Level 3 ●●●
  • 67 replies
  • January 12, 2024

Hello @InessaG thanks for your reply.

How can I do this in my dashboard please 

What do you mean by model ? Is it dataset source ? 


InessaG
Level 5 ●●●●●
Forum|alt.badge.img+19
  • Level 5 ●●●●●
  • 410 replies
  • Answer
  • January 14, 2024

The way I do it is create a Cell-based weighting scheme

and select “Data needs to be weighted as waves” checkbox that appears once you select that option.

You will then selecte the Date field you want to use - typically it’s the “Survey Metadata - Date Ended”, and you’ll want to break it out by year.

 

 

It will then allow you to define the weights for each year separately. Just add all the variables you want to be included in the weighting, and define the separate weights for each year.

 


Forum|alt.badge.img+32
  • Level 6 ●●●●●●
  • 241 replies
  • January 14, 2024

I would also do this as per InessaG comment.


Forum|alt.badge.img+3
  • Author
  • Level 3 ●●●
  • 67 replies
  • January 15, 2024

Hello @InessaG 

Thanks for your answer. 

I’m trying to do this but I don’t see no “Data Ended” in my options :

 

 


Forum|alt.badge.img+3
  • Author
  • Level 3 ●●●
  • 67 replies
  • January 15, 2024

Hello @InessaG 

Oh sorry I finally saw it thanks !

I wonder now how do I create multiple waves within the same Weighting scheme. I could only create one. 

Thanks in advance for your help. 


InessaG
Level 5 ●●●●●
Forum|alt.badge.img+19
  • Level 5 ●●●●●
  • 410 replies
  • January 15, 2024

It should automatically create a wave for each year, you’ll just need to adjust the weights for each year as you see fit.


Forum|alt.badge.img+3
  • Author
  • Level 3 ●●●
  • 67 replies
  • January 15, 2024

@InessaG 

Oh ok I see what you mean now. 

The thing is I would like to apply a certain weighting to some variables (Age x Country) within the different sources, not the data source itself. Do you think it’s possible ? 

My sentences might not be clear enough let me know !


InessaG
Level 5 ●●●●●
Forum|alt.badge.img+19
  • Level 5 ●●●●●
  • 410 replies
  • January 15, 2024

You can add as many variables as you need to to the Cell Based weighting scheme. 


Forum|alt.badge.img+3
  • Author
  • Level 3 ●●●
  • 67 replies
  • January 16, 2024

Hello @InessaG 

Thanks ! Yes, I can see that I can apply weighting on different variables, but what I would like to do is cross my variables a sort of mix between cell based and raked weighting. To be more specific, I would like to apply a weighting on my italiens respondants that have between 30 and 39 year so two different variables. Is there any way I can do this ? 

 

Regards, 

Tata


InessaG
Level 5 ●●●●●
Forum|alt.badge.img+19
  • Level 5 ●●●●●
  • 410 replies
  • January 16, 2024

As I mentioned above, you can add as many variables as you want. So you can add both the language and age into your weighting scheme and assign weights appropriately.


Forum|alt.badge.img+3
  • Author
  • Level 3 ●●●
  • 67 replies
  • January 17, 2024

Hello @InessaG for your help.

I am trying to apply weightings on variables that I would cross (example : age & country) and I am trying to do this for each country but it seems like in your solution if I add any variables they’ll all crossed together, it that right ? 


InessaG
Level 5 ●●●●●
Forum|alt.badge.img+19
  • Level 5 ●●●●●
  • 410 replies
  • January 17, 2024

Yes, just apply a weight of 0 where it’s not applicable.


Leave a Reply