Technology, poverty and education

19th December 2022 by Timo Hannay [link]

Update 19th December 2022: See also this coverage by Tes.

The effects of poverty on education are many and varied, but the switch to online learning during the COVID-19 pandemic threw a harsh light on one in particular: differences in access to technology, whether between local areas, schools or individual pupils. This post looks some related trends before, during and after the pandemic. In the grand scheme of things, the UK appears to be doing quite well with respect to computers and internet use in schools. The focus here is on disparities within the country – specifically, within England – rather than our place in the international pecking order. We conclude that:

• Laptops and tablets distributed for free by the Department for Education during the height of the pandemic went disproportionately to poorer areas, helping to temporarily redress the imbalance in access to such devices, at least during lockdown.
• However, there's no evidence that schools in these same areas spend any more on computers and other digital learning aids, raising a question as to how this technology gap will be mitigated in the longer term and what effects the technology divide will have now that online learning has become so central to education.
• Furthermore, some areas of the country (albeit not the poorest ones) continue to suffer from relatively slow or patchy internet connectivity, whether through fixed or mobile connections.
• Thus technology, in terms of both devices and connectivity, risks becoming yet another axis of educational disparity among the nation's children.

 

Distributed computing

Figure 1 shows a 'cartogram' of local authority (LA) areas in England. Each LA is scaled according to the numbers of pupils who attend school there. Unlike an ordinary map, this gives proper prominence to the urban and suburban areas where most pupils live. The traditional measures of pupil poverty are eligibility for free school meals or (roughly equivalently) the Pupil Premium¹. These tend to be highest in urban areas, especially those in the North East, North West, West Midlands and London. Many educational policies and interventions (not to mention data analyses) are aimed at reducing persistent differences in educational outcomes between these poorer and more affluent areas.

With the first COVID-19 lockdown in March 2020, another consequence of poverty came to the fore: with pupils restricted to online lessons, what if poorer families lacked suitable devices and internet connectivity to take part? Cue an admirable scheme² from the Department for Education (DfE) to distribute free devices (ie, laptops and tablets) and wireless routers to those in need of a computer and an internet connection. Many of these were distributed to schools, trusts and colleges, but about 240,000 devices and 11,000 routers were sent out via LAs with instructions to pass them on to pupils and families in greatest need. Sensibly, these tended to go to some of the same LAs that have high levels of Pupil Premium eligibility. Devices went disproportionately to areas such as South Tyneside, Camden, Tower Hamlets and Islington, while routers (which provide internet connectivity) were donated in greatest numbers to Hartlepool and Sunderland in the North East.

(Use the menu below to switch between Pupil Premium eligibility, devices and routers. Hover over the map to see corresponding LA names and data values.)

Figure 2 shows this relationship between Pupil Premium eligibility and the hardware giveaway more explicitly. Each dot represents an LA, with the line showing a linear regression through all the dots. Though the relationship is not especially strong (bear in mind that this was not the only way that hardware was distributed to poorer families), LAs with more Pupil Premium tended to receive more devices and routers.

It's worth adding that our previous analyses in collaboration with Nesta showed that during 2020 (ie, the first year of the pandemic) users accessing online learning resources from poorer neighbourhoods were more likely to use phones (as opposed to computers) than those from more affluent neighbourhoods, but by spring 2021 (after the DfE had distributed around 1m devices) this gap appeared to have largely closed. While this does not prove a causal link, it suggests a tangible impact on children's learning.

(Use the menu below to switch between devices and routers. Hover over dots to see accompanying values and LA names; hover over the line to see regression statistics.)

So far so good, but what happens now that both the COVID lockdowns and the hardware giveaway are over? Online learning is increasingly embedded in the way children are taught, whether during normal times or during periods of occasional educational disruption. As the pandemic retreats in society's rear-view mirror, is the question of technology access for poorer to be forgotten? For sure, many of the almost two million devices doled out by the DfE over the last couple of years are presumably still out there. But it won't take long for them to be lost to the vicissitudes of cohort turnover, technological obsolescence and plain old wear and tear. A more permanent solution appears to be in order.

One obvious approach would be for some of the aforementioned Pupil Premium funding to be used to improve technology access for those without. But school financial returns show few signs that this has been happening. Figure 3 shows per-pupil spending on "ICT learning resources" (read: computers, software and online learning systems) during the 2020-2021 academic year. Among primary schools, this varied from under £20 per pupil per year (in North Somerset) to well over £100 (eg, in Dudley). The most striking thing about this pattern is that there isn't one: the distribution of high and low spending seems essentially random. The same is true across secondary schools, where there is an even bigger range (less than £5 per pupil per year in North Somerset to over £150 in South Tyneside). But once again, the areas of high spend are not obviously clustered in the same places that we saw above to be home to large proportions of poorer families.

(Use the menu below to switch between primary and secondary schools. Hover over the map to see corresponding LA names and data values.)

This is shown even more clearly in Figure 4: at the LA level, the correlation between Pupil Premium eligibility and school spending on ICT learning resources is negligible among both primary and secondary schools. In other words, school spending on computers and the like is hardly any higher in poorer areas than it is in more affluent ones. As things stand, it therefore seems unlikely that the technology divide will be closed in this way.

(Use the menu below to switch between primary and secondary. Hover over dots to see accompanying values and LA names; hover over the line to see regression statistics.)

 

Only connect

Finally, it's worth considering the fact that we live in an age when a computer has little educational (or any other) value unless it's connected to the internet. And connectivity, too, shows considerable variation around the country, as shown in Figure 5. Premises with fast internet connections (defined as 30Mbps, which is usually enough for several simultaneous video streams) are least common in the South West, as well as other rural and coastal areas – and also, rather bizzarely, in Westminster and the City of London. The distinction for mobile data connectivity is even more stark, with disadvantaged users (ie, those with lower connectivity) tending to reside overwhelmingly in rural areas. It is important not to overstate the scale of the problem: even in under-served areas coverage approaches 90%. But that may be small comfort if you're in the other 10%.

(Use the menu below to switch between fixed-line connectivity and mobile connectivity. Hover over the map to see corresponding LA names and data values.)

Figure 6 shows these connectivity indicators plotted against local Pupil Premium eligibility rates. As you can see, areas that are disadvantaged with respect to connectivity, whether fixed-line broadband or wireless mobile data, are often relatively affluent in traditional socioeconomic terms.

(Use the menu below to switch between fixed-line and mobile connectivity. Hover over dots to see accompanying values and LA names; hover over the line to see regression statistics.)

 

The whole internet or the internet hole?

England is a relatively wired country, but differentially so depending on your family circumstances and location. As online learning becomes ever more important, these gaps need to be measured, understood and addressed. Such disparities are not new, but they are more important than ever, with effects not only on education but also on young people's mental health.

The pandemic was a calamity, but also forced us to face some under-appreciated truths: collective action can be as important as individual liberty, delivery drivers are every bit as essential as bankers, and not everyone has an internet-enabled computer at home. Eager as we are to bid COVID-19 goodbye, it would be a huge mistake to forget the lessons it taught us. One of these is that online learning is here to stay (see this recent DfE-commissioned study). Another is that a concerted effort is required to avoid this becoming yet another axis of inequality in an education system that is already stacked against the poorest.

As ever, we welcome your feedback: [email protected].

 

Footnotes:

1. Schools receive Pupil Premium funds for children who have been eligible for free school meals at some time in the preceding six years, or who have been in care.
2. This was not the only programme to support poorer households. Various charities and trusts also distributed hardware, and in early 2021 mobile networks temporarily waived data fees for selected educational resources.