The first thing to acknowledge about online engagement is that most people are online, with 80% of adults in the UK using the internet daily. The online world is a natural habitat for so many people, and such an established means of communication, that our approach to online engagement should be one of enthusiasm and curiosity. While I and my engagement colleagues love to meet people face to face, we are always mindful of the barriers to F2F engagement. These include the venue’s accessibility, public transport and parking availability, the difficulty some people have with social interactions, and the potential disruption of unforeseen events (which in the last eight months have included a heatwave, a storm, and a pandemic). These all represent potential barriers to F2F engagement, so why not view online engagement as a way to remove such barriers? Well, aside from the question of how to design high-quality online engagement (as Jonathan Bradley says: #NotAnotherSurvey please!1), practitioners and clients are concerned about the possibility of digital exclusion: who is left out of online engagement?
The Office for National Statistics has pulled together various datasets to offer a picture of internet non-usage.
Internet non-usage has been declining consistently: internet non-users (defined as not having used the internet ever or within the last three months) make up 10% of the total adult population of the UK, a total of 5.3m people2. As the chart below shows, this is a reduction from 20.3% in 2011 (10.2m people). If this downward trend has continued, the number of those currently offline will be lower than the 2018 figures.
Internet non-usage correlates strongly with age. It is conventional wisdom that older people are less likely to be online – and this is true. People over the age of 65 make up an increasing proportion of internet non-users, though it should be kept in mind that the overall number of people offline is decreasing. In fact the report makes it clear that more and more older people are internet users. In 2018, 80% of those in the 65-74 year old cohort were internet users; for men aged 75+ the figure was 51% (up from 29% in 2011), while 38% of women aged 75+ were internet users (up from 13% in 2011).
The differences in internet non-usage by ethnicity have been changing relatively rapidly. The gaps in usage between different ethnicities evident in 2011 had narrowed by 2018 as the proportion of internet non-users declined, with white people having the highest percentage of internet non-users. In 2011, 31% of UK Bengali adults were non-users, compared to just 20% of UK adults overall. Yet by 2018, the figure had fallen to 8% - slightly lower than the UK total of 10%.
A total of 56% of internet non-users are disabled, with disabled people forming half or more of all internet non-users in each age cohort. Among those of working age, people who are economically inactive are more likely to be internet non-users; as the ONS also points out, those on lower incomes are less likely to have an internet connection in the home, as are older people who live alone. Finally, women make up 61% of all internet non-users, a proportion that has remained constant even as the number of women (and men) becoming internet users has continually increased.
In describing internet non-users, the ONS emphasises the difference between using the internet, and being capable and safe in navigating the digital world. Lloyds Bank’s UK Consumer Digital Index 2019 uses an Essential Digital Skills framework to separate out Foundation skills from Essential Digital Skills for life, and for work. Lloyds used a sample of one million consumers to generate its findings, as well as phone and F2F interviews. Although this research also shows a general picture of progression through the levels of digital skills, it found that 22% of people do not have the Essential Digital Skills for life, while 53% of employees do not have the Essential Digital Skills required for workplaces. The report-writers also believe that in 2030 “there will still be 8% of the adult population (a predicted 4.5 million people) who show little or no signs of digital behaviour, only a four percentage point decrease” from 2019.
One interesting element of the Lloyds report is its exploration of the barriers to being online. Asked to choose from a range of options, 75% of those who are offline said they have ‘no interest’ in being online, while more than half referred to security concerns; it is notable that the figures for all barriers increased from 2017. Looking at those three-quarters who said that they have no interest in being online, a large majority (89%) were also able to give other reasons, particularly focusing on security (see the details below). This leads to one of the report’s big take-away findings: “[the data] suggests that while a lack of interest may be the perceived barrier, underpinning this is a strong set of concerns revolving primarily around cybersecurity”.
Of course, some people experience practical barriers to online participation: one-fifth of the offline sample referred to a lack of adequate connectivity. While about half of the offline sample said ‘nothing’ could get them online, the other half offered actions that could be taken to get them online, the top three being:
We can picture what good online engagement looks like. At Traverse we definitely agree with Jonathan Bradley’s formulation, which includes:
We’d add to this a sensible strategy for collecting and analysing data from a range of channels, to ensure that you are capturing the same level of information you would F2F – we already use our Magpie analysis system to collate data from a mix of sources.
But how can we respond to the issues around digital exclusion outlined above?
Sampling
It is helpful firstly to think about sampling. When designing a research sample for deliberative engagement, we know that not all people will be included. The sampling framework will include relevant attributes, and though the sample will often be based on national data, a programme of research will often ensure that particular voices are heard by using top-up quotas or including specific additional strands of work to capture the voices of certain groups. This is an approach that works well because it surfaces the breadth of issues on a topic. Online engagement may not be able to reach those who are completely offline, but a carefully-designed sample can reach people with similar attributes to those who are offline, in the same way as those who will attend a F2F event in effect stand in for those who will not attend.
Given the correlation between disability and digital exclusion, for example, we can use top-up sampling for disabled people, to be confident that we are building in appropriate input from them. Reaching out to existing groups (either online forums or existing offline social groups who could disseminate the invitation) is one way to access people who could otherwise be difficult to reach. Working through existing groups always requires us to consider how inclusive its membership is – we can’t take this for granted – but the benefit is that a group dynamic will often bring a level of trust and familiarity that makes engagement easier.
To understand the reach of a project, we could ask participants questions to assess whether we are reaching those who are, in terms used by the Lloyds report, ‘digitally disengaged’ rather than completely offline. This can give clients confidence that the research has not been conducted solely with digital natives whose views may not reflect those of society at large. In our F2F engagement we commonly make use of persona cards and other techniques to prompt participants to reflect on the requirements of different kinds of people. Explicitly asking online participants to think of the needs of those they know who are offline or otherwise digitally disengaged can help us to surface a wider range of responses as well as understand the complexity of what it means to be digitally excluded (do friends and family act as an online proxy? do those who are offline have devices available to them?).
Inclusive design
We should make use of general good practice for accessibility to make the experience as simple as it can be for a broad range of users, with appropriate design considerations, clear language and speech functionality. We know from the ONS data that older and disabled people are less likely to use internet ‘on the go’, so any online engagement should be appropriate for use across a range of devices, including PCs, laptops, and smart devices, in order to include the widest range of internet users.
People not being online – or lacking confidence online – is clearly a barrier to online engagement. But we can also see online engagement as an opportunity to build people’s digital skills, provided we can put the time and resources into doing so. Practitioners can for example collaborate with people who are recently online to assess whether their proposed methodologies meet the needs of people in their circumstances. These people could become advocates who invite others to join the engagement, or who support them to join. Engagement can also include ‘getting online’ support for individuals as part of the engagement process – for instance, including the less digitally-confident in a recruited sample, then providing them with help so that they can acquire the skills needed to take part.
Clarity about the purpose of the research and the security of the tools used, including anonymity for users who need it, can build participants’ confidence in navigating the online environment. Practitioners can consider offering an additional general session on staying safe online, to further the social impact of the research. This approach could allow participants to understand the value of the internet, potentially overcoming scepticism about the online environment. The Lloyds research refers to the ‘digital dividend’ - savings that can be made by those who use the internet, particularly in the cost of their utilities, so bringing participants online could materially improve their lives beyond the lifetime of the project. There may also be wellbeing benefits if, during the current Covid-19 pandemic, we can include people who are self-isolating – as the NHS advises older people who are experiencing loneliness: ‘learn to love computers’4.
Having the right kit
Some people may be confident online, or keen to take part in online engagement or other activities, but lack the equipment needed; they may be able to access the internet only in particular circumstances (e.g. in a library or café) or have a poor quality connection. Engagement practitioners should find out what devices people have available to them before the activities begin, as well as what their connectivity is like. Really inclusive engagement would involve working with participants to ensure their engagement is as good as possible, which might include providing devices or connections to those who lack the kit to take part.
The research shows that there are people who, as things stand, would be excluded from online engagement, but we should bear in mind that no approach will include everyone – and for some, online will be more accessible than face to face, rather than less. There are also actions we can take to reduce the number who are not currently confident online so that certain groups likely to be under-represented are included. Online engagement will often only be part of an engagement process. It is possible, for instance, to carry out telephone interviews or conference calls to supplement any online engagement, and at some point in the future findings from online groups can be tested against those from F2F research. But before F2F work becomes feasible again, we have an opportunity to explore and embed online engagement and see what we can all learn from it.
1. All references to Jonathan Bradley taken from the podcast at:
2. All ONS data taken from:
3. All references to the UK Consumer Digital Index taken from:
4. https://www.nhs.uk/conditions/stress-anxiety-depression/loneliness-in-older-people/
