## Support #1494

### Weights

100%

**Description**

Dear USoC team,

I am doing an analysis on young adults which comprises wave 12,13,17 of BHPS and 2,3,5,7,9 of UKHLS.

All the boosts are considered. I would like to know if I should use cross-sectional weights (which,I think,

is the case) and if I should rescale weights to consider the bigger sample size of UKHLS. In

the case, this is necessary, how could it be performed?

Thank you very much,

Lydia

#### Updated by Understanding Society User Support Team 8 months ago

**Status**changed from*New*to*In Progress***% Done**changed from*0*to*10***Private**changed from*Yes*to*No*

Dear Lydia,

Many thanks for your enquiry. The Understanding Society team is looking into it and we will get back to you as soon as we can.

We aim to respond to simple queries within 48 hours and more complex issues within 7 working days. While we will aim to keep to this response times due to the current coronavirus (COVID-19) related situation it may take us longer to respond.

Best wishes,

Understanding Society User Support Team

#### Updated by Understanding Society User Support Team 8 months ago

**Category**set to*Weights***Assignee**set to*Olena Kaminska*

#### Updated by Olena Kaminska 8 months ago

Lydia,

Thank you. Are you pooling all the information together and using it in one model?

Are you studying a cross-sectional concept - so are you not using any information from previous wave per person?

Thanks,

Olena

#### Updated by Lydia Palumbo 8 months ago

Dear Olena,

thank you for the answer. Yes - I am pooling observations. Individuals might be observed more than once,

but I am treating them as cross-sectional units. So, no, there is no information from the previous wave.

#### Updated by Olena Kaminska 8 months ago

Lydia,

Then it's yes to both questions: use cross-sectional weights, and rescale them to take into account larger sample size in UKHLS.

Best,

Olena

#### Updated by Lydia Palumbo 8 months ago

Thank you for your answer. I would like to check if I follow the correct procedure.

Following the instructions you sent me some time ago, the appropriate procedure to rescale weights is the following:

1) Calculate the weighted sample size for each wave; Sum of the cross-sectional weights of the waves which I am interested in (waves 12,13,17 of BHPS and 2,3,5,7,9 of UKHLS), limited to the sample of young adults;

2) Take the average of weighted sample sizes across the waves you use;

3) divide the average by the weighted sample size for each year to get the scaling factor;

4) multiply the scaling factor for each wave by its cross-sectional weight.

5) Use this product as a new weight for pooled data.

Thank you and best,

Lydia

#### Updated by Lydia Palumbo 8 months ago

Sorry, Olena.

Let me just add a question to the previous post re: 1).

1) Calculate the weighted sample size for each wave; Sum of the cross-sectional weights of the waves which I am interested in (waves 12,13,17 of BHPS and 2,3,5,7,9 of UKHLS), *limited to the sample of young adults;*

Is it necessary that I compute the rescaling factor only on young adults? If I remember correctly, I can also compute the rescaling factor on the entire sample (in the appropriate waves) and, then, analyse the sub-sample I am interested in.

Thank you again.

Best,

Lydia

#### Updated by Olena Kaminska 8 months ago

Lydia,

Yes, all the steps described above are correct.

The scaling fractions on the whole population or just youth should be very similar - it's ok to use either. I would personally recommend to use sum of weights of the people in your analysis (youth or its subgroup). But this is practical if this is constant across everything you analyse. If you conduct many analyses you may want to create one general scaling factor - and general weights, and go from there.

Hope this helps,

Olena

#### Updated by Lydia Palumbo 8 months ago

Yes, Olena. That indeed helped.

Thanks very much.

Best,

Lydia

#### Updated by Understanding Society User Support Team 8 months ago

**Status**changed from*In Progress*to*Resolved***Assignee**deleted ()*Olena Kaminska***% Done**changed from*10*to*100*