Haiti High Frequency Survey 2025: Sampling Design and Weighting
Summary: How the 2025 Haiti High Frequency Survey was sampled and weighted. The sample was drawn by random digit dialling across all active mobile numbers, and the note sets out the stratification, the response rates and the post-stratification weights used to bring a phone-based sample back to the population. It matters beyond this survey: with face-to-face fieldwork constrained by insecurity, phone surveys carry much of Haiti's recent household evidence, and this is the document that says how far they can be pushed.
Key Findings
- The sample was generated by random digit dialing across all active cell phone numbers at the time of selection in May 2025.
- Estimates therefore represent households with at least one cell phone, and individuals aged 18 or older living in them, which is the coverage limit any user of the data has to carry.
- Derives household and individual selection probabilities from the inclusion probabilities of the cell phone numbers.
- Collected data on one randomly selected child aged 0 to 5 or adolescent aged 6 to 17 in each interviewed household, so child-level estimates rest on a within-household random selection.
How to cite
World Bank, Haiti High Frequency Survey 2025: Sampling Design and Weighting, https://microdata.worldbank.org/index.php/catalog/8516