An image of a Learnometer device, switched on and lit.


Learnometer research: helping you to monitor your classroom environment for factors which may hinder learning

We are proud to have partnered with Murray Hudson and his team at Gratnells - also based in Essex UK. Gratnells have excellent products in schools in almost every country in the world and we are using their unique chain and expertise to get Learnometers out to everyone - please go to their site to book yours and be part of the research.

A poor physical environment hurts learning.

Learnometer is a unique combination of hardware, software and analysed data that help learners, and thus schools, perform better by optimising physical environments for learning. The process of making learning better also engages everyone as reflective learners, improving their learning further.

Our research, and others', confirms that poor light levels, the wrong temperatures, inappropriate sound volumes and rhythms, humidity, air pollution, Co2, and air pressure can all impair learning. Our Learnometer research tool automatically samples your classroom environment, and makes suggestions through a unique algorithm as to what might be changed to allow students to learn and perform at their best.

Although each of these variables matter individually, our research embraces the aggregation of marginal gains. What we don't know yet is the extent to which learning in a high CO2 environment damages your thus impaired concentration, so that the impact of too hot, or too dark is more damaging that would otherwise be the case. We have a lot of data - around half a million hours. As our boxes roll out we will have more data to explore this. Don't expect multiple non optimal factors together to make learning anything other than worse however.

Cloud-based logging allows you to see, at a glance, how your environment looks, and lets you compare it with benchmark environments. You can analyse general learning spaces, or specific areas like examination and test rooms. Changes you make can be evaluated and compared, and then finessed over time.

Shortcut me to some possible remedies for less-than-ideal learning spaces

The front and Back of a Learnometer device. dashboard button

Learnometer tracks...

Light levels

Poor lighting is a significant barrier to learning. Our work is confirmed by recent school environment studies, and detailed research for other working environments. For example recent research confirms evidence that good lighting significantly influences reading vocabulary and Science test scores (Barrett et al., 2015). Learnometer reports lumen (lux) levels throughout the day and plots them constantly and automatically for you wherever it is installed.

Our guide would be above 500 lux, with a sensible target of 1,000 - but being careful to get the "right" Kelvin values in your elctrical lighting too.

Temperature & Humidity

Lots of caveats - our literature search was like a jigsaw puzzle, everyone had bits, noone had it all. With much data already logged, we know that many classrooms are either too hot or (less damagingly) too cold for learning. Even though national law (eg UK) does not usually give an upper limit for temperature, research broadly suggests that between 18 and 21 degrees is ideal for learning, although it arguably doesn't start to be significantly damaging for a couple more degrees.

Learnometer contains a sensitive digital thermometer and hygrometer which tell you the temperature of your classrooms throughout the day. You’d be amazed how much it changes.

Joshua Graff Zivin, of UCSD, and two colleagues in an NBER Working Paper noted that "we find that math performance declines linearly above 21C". They thankfully didn't see much accumulation of damage. Fix it and it ends.

More recently, R. Jisung Park, Joshua Goodman,and Mike Hurwitz (who were looking as the equity issue of schools who could cool spaces, and those who could not) found that "Using data from over 12,000 schools and 10 million middle- and high-school students across America, my colleagues and I found that students who experience more hot days during the school year perform worse on subsequent standardized exams. A small 1 degree hotter-than-average academic year reduces learning by about 1%." An opinion piece by R. Jisung Park is here.

Sound volume & rhythms

Studies confirm that the classroom sound signature can affect how well students achieve (Picard and Bradley 2001). Learnometer constantly gives you automated feedback on sound volumes (including from the students!) and rhythms (for example from projector fans or air conditioning units) in your classroom, will highlight changes to be made, and plots the results of your changes by hour, day, and even year.

As a starting point we suggest that sound above around 72 decibels starts to be disruptive - although what the sound is matters too (for example is is hard to write when background music contains a familiar lyric). A simple guide for beats per minute is to stay below 80 bpm


Carbon Monoxide and Carbon Dioxide cannot be sensed by humans, both of course have the potential to kill, but at lower levels, Co2 will affect concentration.

Our pilot work suggested a correlation between absence / illness levels and high classroom Co2 levels (for staff and students) and we have been dismayed by the damaging levels we have observed in examination and test rooms. Learnometer logs and tracks the levels of Co, Co2 and particulate matter ("pollution") to help you identify airflow issues and optimum levels for learning and performance.

A useful guide would be to be aware that above as little as 1,000 parts per million (and arguably lower still) Co2 will be inducing sleepiness, poor concentration with abnormal heart rate and nausea. Problems ncrease towards 5,000 ppm which is a workplace limit in most countries, but which, disappointingly, many learning spaces often exceed daily.

This 2015 study from the Harvard School of Public Health found that carbon dioxide (CO2) has a worse, direct and negative impact on human cognition and decision-making than was previously understood.

Air pollution

When we began the learnometer project the link between wellbeing and pollution was clear, but pollution's impact on learning was largely hypothesised by us, without much good evidence. Since then however some substantial research projects, like this one using data from China, published by Xi Chen et al, at Yale School of Public Health in the US, have suggested that high levels of urban pollution have a major impact on attainment - some children dropping a whole year of progress in their school lives.

Although it is unlikely that relocating a school will be possible (but see our remedies section), battles against planning approval for siting polluting industries near schools will be a lot easier to win, given clear pollution data from within classrooms. However, a little micro-research study at ARU's Cambridge campus has shown us that opening windows to let the heat and CO2 out, can let high levels of urban pollution in - a nice indicator of the complexity of all this which our learnometer and its algorithms might hopefully inform.

Pollution levels can be damaging in the case of a one-off exam too. Avraham Ebenstei,n Victor Layv and Sefi Roth found published in 2016 that pollution hurt the exam marks, but that the impact f that lasted into later attainment and employment. Another nail in the coffin of examination equity?

"Exploiting variation across the same student taking multiple exams, we find that transitory PM2.5 exposure is associated with a significant decline in student performance. We then examine these students in 2010 and find that PM2.5 exposure during exams is negatively associated with postsecondary educational attainment and earning."

Air pressure

There is much we have yet to learn about air pressure, or indeed other variables captured by our little Learnometer research units. As you optimise your learning spaces, we will analyse the data aggregates too, and add them to the knowledge we are building for you.

At a common sense level many people complain of pressure related sinus headaches that can certainly cause distraction and, we might assume, a loss of concentration. Of course weather and altitude play a role in the pressure we experience. We might hypothesise that cultural norms from living at sea level, or at high altitude, will impact of our tolerance levels too.

Whilst we evolve our research and understanding, Learnometer will allow you to look at your own data too and to engage in your own practice based research.

Professor Stephen Heppell.

To learn well is complex. Alongside the hard work, the pedagogy, collegiality, passion, delight, knowledge, joy, and engagement there is the detail of physical learning spaces themselves. Learnometer builds on work I have been involved in for three decades, designing better classroom environments and understanding how learner-led design increases achievement in measurable ways.

We’re now at an exciting point where the advent of affordable Internet connected smart devices has combined with further research on how factors such as light, heat and air-quality affect classrooms. The team at Learnometer have built some amazing devices to measure this data automatically, and we’d love to see them help your school.

There is plenty yet to learn, but by joining the Learnometer research family, you will be changing and improving learning forever, and informing significant change in your own learning spaces too.


Professor Stephen Heppell

Get updates

We'd love to tell you more about Learnometer. We're looking for real teachers, and real schools to record and add to our real data. As we progress we are continually developing our hardware and algorithms - and you will gain continuously from those updates.

If you'd like to learn more, please sign up below via Mail Chimp. We'll never sell or give away your email address, ever. That is our absolute promise.

all rights reserved - patents pending

We are constantly updating this site - keep coming back for more! or follow @learnometer on Twitter

Some examples of the original data from 2014 and annotated are here