# Helpdesk

**URL:** https://community.worldpop.org/c/helpdesk/19.md

[Latest](https://community.worldpop.org/latest.md) · [Categories](https://community.worldpop.org/categories.md) · [Tags](https://community.worldpop.org/tags.md)

---

## [About the Helpdesk category](https://community.worldpop.org/t/about-the-helpdesk-category/26)

<div class="topic-metadata">

**Author:** [@Graeme\_worldpop](https://community.worldpop.org/u/Graeme_worldpop)\
**Replies:** 0\
**Last updated:** [July 14, 2026, 10:44pm UTC](https://community.worldpop.org/t/about-the-helpdesk-category/26 "2026-07-14T22:44:23Z")

</div>

Get help using WorldPop data and tools, from troubleshooting to technical how-to’s. Once you’ve signed up, you can reply to this message, or click New Post :memo: to raise your own query. Alternatively, if you have que…

---

## [Zonal statistics giving unexpected population totals](https://community.worldpop.org/t/zonal-statistics-giving-unexpected-population-totals/64)

<div class="topic-metadata">

**Author:** [@Graeme\_test](https://community.worldpop.org/u/Graeme_test)\
**Replies:** 2\
**Last updated:** [November 17, 2026, 12:07pm UTC](https://community.worldpop.org/t/zonal-statistics-giving-unexpected-population-totals/64 "2026-11-17T12:07:09Z")

</div>

I ran zonal statistics in QGIS to get population totals per district, but my numbers are way higher than official census figures for some districts and way lower for others. Is the dataset wrong, or am I doing something …

---

## [Choosing a no-code tool to map population data for a local health team?](https://community.worldpop.org/t/choosing-a-no-code-tool-to-map-population-data-for-a-local-health-team/170)

<div class="topic-metadata">

**Author:** [@samuelmiller](https://community.worldpop.org/u/samuelmiller)\
**Replies:** 0\
**Last updated:** [November 11, 2026, 7:11pm UTC](https://community.worldpop.org/t/choosing-a-no-code-tool-to-map-population-data-for-a-local-health-team/170 "2026-11-11T19:11:19Z")

</div>

I volunteer with a small health outreach group in North Carolina, and we use WorldPop estimates to sketch out where to focus our door-to-door visits. Right now the process is clumsy: someone downloads the population rast…

---

## [Constrained vs unconstrained layers give different national totals](https://community.worldpop.org/t/constrained-vs-unconstrained-layers-give-different-national-totals/65)

<div class="topic-metadata">

**Author:** [@Graeme\_test](https://community.worldpop.org/u/Graeme_test)\
**Replies:** 2\
**Last updated:** [October 7, 2026, 10:30am UTC](https://community.worldpop.org/t/constrained-vs-unconstrained-layers-give-different-national-totals/65 "2026-10-07T10:30:26Z")

</div>

I’m getting different national totals from the “constrained” and “unconstrained” population layers for the same country and year. Which one should I be using, and why do they disagree in the first place?

---

## [CRS mismatch when overlaying WorldPop raster with admin boundaries](https://community.worldpop.org/t/crs-mismatch-when-overlaying-worldpop-raster-with-admin-boundaries/62)

<div class="topic-metadata">

**Author:** [@Graeme\_test](https://community.worldpop.org/u/Graeme_test)\
**Replies:** 2\
**Last updated:** [October 7, 2026, 10:22am UTC](https://community.worldpop.org/t/crs-mismatch-when-overlaying-worldpop-raster-with-admin-boundaries/62 "2026-10-07T10:22:56Z")

</div>

Hey WorldPop! I downloaded the 2020 population count raster for Kenya and my admin boundary shapefile from GADM, but when I load them in QGIS the raster doesn’t line up with the boundaries at all — it’s offset by what l…

---

## [Random forest model R² looks "too good to be true"](https://community.worldpop.org/t/random-forest-model-r-looks-too-good-to-be-true/61)

<div class="topic-metadata">

**Author:** [@Graeme\_test](https://community.worldpop.org/u/Graeme_test)\
**Replies:** 0\
**Last updated:** [September 2, 2026, 1:13pm UTC](https://community.worldpop.org/t/random-forest-model-r-looks-too-good-to-be-true/61 "2026-09-02T13:13:45Z")

</div>

Hi WorldPop, I’m building a random forest model for dasymetric redistribution similar to WorldPop’s approach, and I’m getting an R² of 0.97 on my validation. That seems suspiciously high. Am I doing something wrong, or …
