Based on your code snippet, using HWTFINL and restricting to one-person per household is appropriate. When pooling data, you will need to divide your sample weights by the number of samples that you are analyzing. If you are analyzing CPS BMS samples from 2023-2025, you would need to divide your weights by 29 (12 samples in 2023, 12 samples in 2024, and 5 samples in 2025). There is some additional information on this forum post regarding weighting CPS pooled data.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Properties of the CPS weights and pooled analyses | 1 | 1036 | June 30, 2020 | |
| I'm using CPS to calculate LFP for a subset of workers. I think I use stata code pweight=wtfinl. | 1 | 873 | June 28, 2017 | |
| Merging Jan-Dec IPUMS CPS files: How to set up weights correctly? | 3 | 783 | April 24, 2020 | |
| How to count people in a specific occupation with wtfinl | 18 | 417 | October 6, 2025 | |
| What weight should I use in IPUMS-CPS | 1 | 1895 | January 20, 2013 |