Connecting to the Internet: A survey of Diplo’s online students

We designed our online learning environment knowing that many of our participants are located in developing countries, including some of the poorest countries in the world. We wanted to ensure that course materials, tools and conversations would always be accessible for all course participants, even those with very slow Internet connections. But the technology for accessing the Internet is changing rapidly, and high-speed Internet access is reaching more parts of the world. In order to better guide work on a new version of our online classroom, we decided to check how course participants were currently accessing the Internet. Were our assumptions correct?

In June 2011 we invited participants in our current diplomacy and Internet Governance online courses to take part in a short survey. We asked them what kind of Internet connections they were using to take part in our courses, and the speed of those connections. The 181 participants who responded were located as follows:

• North America: 7% (13 respondents)
• Central America and Caribbean: 16% (29 respondents)
• South America: 7% (12 respondents)
• Europe: 10% (18 respondents)
• North Africa and the Middle East: 4% (8 respondents)
• Sub-Saharan Africa: 38% (69 respondents)
• Asia: 12% (21 respondents)
• Australia and Pacific Islands: 4% (7 respondents)
• No response: 2% (4 respondents)

The survey showed that just over half of our course participants access the Internet in more than one way (for example, using the office network at work, and using a modem of one kind or another at home). The table below shows the types and distribution of use of different kinds of Internet connection. 

The majority – some 66% – of respondents access our online courses using an office Internet network. Perhaps more significant is the fact that 25% of respondents access their online courses ONLY using the office network. Without more data we cannot know whether this is because their employers support study time during working hours, or because they do not have other options such as an Internet connection at home.

Some 8% (15 respondents) of course participants use dial-up Internet connections (typically not faster than 50 kb/s) to access their online courses. This may seem like a small number, however it means that one in every twelve participants may be using dial-up. However, only three of the respondents reported that dial-up was their only Internet connection; for the rest, it was one of two or more ways they connect. Participants using dial-up were mostly located in Sub-Saharan Africa.

Respondents were also asked to indicate their Internet speed, if known. Reported Internet connection speeds ranged from 32 kb/s through to 100 Mb/s. Here is the distribution:

To put this into perspective:

• A connection speed of 350 kb/s is needed to comfortably view the small video clips used in some Diplo courses.
• Online video-conferencing works best with connection speeds of 500 kb/s or more.
• Some materials on YouTube require connection speeds of 1 Mb/s or more.
• Loading an online course text with hypertext entries will take approximately 10 times as long using a 100 kb/s connection as with a 1 Mb/s connection. So if it takes 30 seconds for you to load the text, using a 1 Mb/s connection, your colleague using a 100 kb/s connection will have to wait up to 5 minutes for the same thing.

Not surprisingly, Internet speed tends to relate to location: with just two exceptions, course participants in Europe and North America all had connection speeds of 1 Mb/s or higher, while the slowest reported connection speeds were concentrated in predominantly developing regions. In addition, many participants from developing regions reported connection speeds of less than 500 kb/s even using ADSL, office networks, Mobile dongles, and mobile phone connections. There are many reasons this might be the case, including poor telephone wire infrastructure, too many Internet service providers using the same Internet cables; and Internet service providers over-selling their capacity.

Of course this is just a snapshot taken at a specific point in time. But it does suggest that a significant number of course participants continue to access our classrooms using dial-up or other slow methods. Diplo’s learning tools need to continue to allow flexibility and collaborative learning, without excluding participants located in the world’s poorest countries.