How Accurate Are Consumer Air Quality Monitors? The Independent Data
How accurate are consumer air quality monitors? AQ-SPEC PM2.5 results by model, what R-squared means, humidity errors, NDIR vs eCO2 accuracy and VOC sensor limits.
Affiliate disclosure: product links on this page go to Amazon, or to the maker's own store for brands that sell only direct. As an Amazon Associate we earn from qualifying purchases, at no extra cost to you; a link to a maker's store earns us nothing. No retailer or brand sees our picks before they are published or has any say in them. How we choose.
Consumer air quality monitors range from close to useless to impressively accurate for PM2.5, and the only way to know which is which is independent testing. South Coast AQMD's AQ-SPEC program runs sensors beside reference instruments in the field and lab. In its summary, the PurpleAir PA-II scored a field R-squared of 0.93 to 0.97 for PM2.5, the Qingping Air Monitor Lite 0.85 to 0.93 and the IQAir AirVisual Pro 0.69 to 0.73, while the Airthings Space Pro scored 0.44 to 0.49.
For CO2, accuracy depends mostly on the sensor type: NDIR sensors are specified to within roughly ±40 to ±75 ppm plus a few percent, while "eCO2" readings are calculated from VOC signals and are not CO2 measurements at all. VOC sensors only show relative change. Here is the data and how to use it.
PM2.5: the AQ-SPEC results
| Monitor | Field R² | Lab R² |
|---|---|---|
| PurpleAir PA-II | 0.93 to 0.97 | 0.99 |
| Temtop LKC-1000S+ | 0.91 to 0.92 | 0.99 |
| Atmotube Pro (original) | 0.88 | 1.00 |
| Qingping Air Monitor | 0.86 to 0.90 | 0.95 to 0.98 |
| Qingping Air Monitor Lite | 0.85 to 0.93 | 0.99 |
| Kaiterra Laser Egg 2+ | 0.81 to 0.85 | 0.99 |
| PurpleAir PA-II-FLEX | 0.78 to 0.88 | 0.99 |
| Temtop M2000 (2nd Gen) | 0.77 to 0.82 | 0.99 |
| PurpleAir PA-I-Indoor | 0.75 | 0.99 |
| IQAir AirVisual Pro | 0.69 to 0.73 (0.63 to 0.81 with firmware 1.1683) | 0.99 |
| Temtop Airing 2000 | 0.57 to 0.59 | - |
| Airthings Space Pro | 0.44 to 0.49 | - |
We did not find the Airthings View Plus, Awair Element, Temtop M10+ or Aranet4 in the summary; the Aranet4 has no particle sensor. A missing model is not a bad result, only an unknown one.
What R-squared means
R-squared measures how closely a sensor's readings rise and fall with the reference instrument, from 0 (no relationship) to 1 (perfect tracking). It does not measure bias: a sensor can track perfectly and still read consistently high or low. AQ-SPEC's individual reports for each model give more detail than the summary. For a home user, a high R-squared means the monitor's spikes and dips are real; whether 12 µg/m³ on the screen is exactly 12 is a separate question.
Lab results are almost always better than field results because the lab uses controlled particles and conditions. The field figure, outdoors beside a regulatory monitor through changing weather, is the one to compare.
Why humidity throws off particle readings
Low-cost laser sensors count light scattered by particles, and most have no way to dry the air first. A peer-reviewed study on humidity correction for low-cost sensors notes that particles take up water at high humidity, so the sensor sizes them too large and reports more mass than a reference monitor. EPA developed a national correction for PurpleAir data used on its AirNow Fire and Smoke Map for this reason. Indoors, the practical effect is that readings in a steamy bathroom or a humid basement can run high.
CO2: sensor type matters more than brand
| Monitor | Sensor | Stated accuracy |
|---|---|---|
| Airthings Wave Plus | CO2 sensor (Airthings) | ±30 ppm ±3% |
| IQAir AirVisual Pro | NDIR | ±40 ppm or ±3% |
| Airthings Wave Enhance | NDIR | ±50 ppm ±5% |
| Airthings View Plus | NDIR | ±50 ppm ±3% (product page) |
| Awair Element | Not stated | ±75 ppm or 10% |
| Atmotube PRO 2 | Thermal conductivity (Sensirion STCC4) | ±(100 ppm + 10%) |
Watch for "eCO2" or "CO2eq". Sensirion's SGP30 datasheet explains that its CO2eq signal is calculated from ethanol and hydrogen measurements using internal algorithms, so it tracks VOCs, not ventilation. Many NDIR sensors also use automatic baseline calibration, which assumes they see outdoor-level air regularly; Atmotube, for example, says its sensor assumes about 400 ppm air at least once a week. A monitor kept in a room that never airs out can drift.
VOCs: relative, not absolute
Consumer VOC sensors are metal-oxide sensors that respond to many gases at once. They cannot tell you which VOC is present or give a concentration you can compare with a health limit, and EPA notes there are no federally enforceable standards for VOCs in homes. Use the reading as a trend: it should jump when you paint or clean and fall when you ventilate.
Radon: slower but specified
Digital radon monitors take days to settle. Airthings states ±10% after 7 days and ±5% after 2 months at 200 Bq/m³ for the View Plus. EPA lists continuous monitors among the devices used for short-term testing, but its follow-up rules still apply: a result of 4 pCi/L or more calls for a second test before you decide to fix. See the radon detector guide.
How to sanity-check your own monitor
- Fresh-air CO2 check. Put the monitor by an open window or outdoors in the shade for 15 minutes. An NDIR sensor should read near the outdoor level, which NOAA's Mauna Loa record put at 426 ppm in September 2026; city air can run higher.
- Particle response check. Make toast a few feet away. A working laser sensor jumps within seconds and falls back over minutes as the air clears.
- Outdoor comparison. On a calm day with windows open for an hour, indoor PM2.5 should move toward the outdoor figure on AirNow.
- Side by side. Two monitors next to each other should agree on trends even if their absolute numbers differ.
Which monitors to trust
Prefer a model with a published AQ-SPEC field result above about 0.8 for PM2.5 and an NDIR CO2 sensor. The Qingping Air Monitor Lite meets both for particles at a low price; for CO2 alone, the Aranet4 uses NDIR. See the best air quality monitors, the Qingping Lite review, the PurpleAir Flex review and our explainer on how monitors work.
Frequently asked questions
Are cheap air quality monitors accurate?
Some are. In South Coast AQMD's AQ-SPEC tests, the Qingping Air Monitor Lite scored a field PM2.5 R-squared of 0.85 to 0.93, while the Airthings Space Pro scored 0.44 to 0.49. Price is not a reliable guide; check the independent result.
Which air quality monitor is the most accurate?
Among the models in AQ-SPEC's PM summary, the PurpleAir PA-II scored highest in the field at 0.93 to 0.97 for PM2.5. It is an outdoor sensor; among indoor monitors, the Qingping Air Monitor Lite scored 0.85 to 0.93.
Is AQ-SPEC a certification?
No. It is South Coast AQMD's evaluation program that tests sensors beside reference instruments and publishes the results. A model not in the summary has simply not been evaluated there.
Why does my monitor read high in the bathroom?
Laser particle sensors generally do not dry the air, so at high humidity particles absorb water and are counted as larger. Readings in steamy or damp rooms can run high.
Is eCO2 accurate?
It is not a CO2 measurement. Sensirion's SGP30 datasheet says CO2eq is calculated from ethanol and hydrogen signals, so it follows VOCs rather than ventilation. Choose an NDIR sensor for CO2.
How do I calibrate an air quality monitor?
Follow the maker's instructions. Many NDIR CO2 sensors self-calibrate by assuming they see outdoor-level air regularly, so airing the room or placing the monitor by an open window now and then helps.
Sources
- AQ-SPEC Sensor Evaluations: Summary of PM Sensors, South Coast Air Quality Management District
- Developing a Relative Humidity Correction for Low-Cost Sensors Measuring Ambient Particulate Matter, Sensors (MDPI), via PubMed Central
- EPA Research Improves Air Quality Information for the Public on the AirNow Fire and Smoke Map, US Environmental Protection Agency
- Datasheet SGP30: Indoor Air Quality Sensor for TVOC and CO2eq Measurements, Sensirion
- Recommendations for CO2 Sensor Usage in Atmotube PRO 2, Atmotech
- AirVisual Pro Indoor Air Quality Monitor, IQAir
- Awair Element: Air Quality Monitor For Your House, Awair
- Volatile Organic Compounds' Impact on Indoor Air Quality, US Environmental Protection Agency
- A Citizen's Guide to Radon, US Environmental Protection Agency
- Trends in CO2, NOAA Global Monitoring Laboratory
Related guides
Air Quality Monitor for a Nursery: What to Track and Which to Choose
A nursery monitor is most useful for temperature, humidity and ventilation, with PM2.5 as a bonus. It does not replace a smoke or CO alarm. Here is what to track and what to buy.
Air Quality Monitors Explained: Sensors, Readings and What They Can Miss
A home monitor is a bundle of small sensors, each good at one thing. Knowing which sensor sits behind each number tells you which readings to trust and which to treat as a trend.
Airthings View Plus Review: Radon, PM2.5 and NDIR CO2 in One Monitor
The View Plus is the most complete home air monitor Airthings makes: radon plus particles, CO2 and VOCs, on batteries or USB-C. It is expensive, and worth it mainly if you want radon too.