Digital divide is a frequent and popular topic. The general feeling is that it is something negative that should be fought against, and diminished as much as possible. But is there a consistent and generally accepted definition of digital divide? One that would enable us to measure digital divide, evaluate it, compare, discuss etc.? Or is digital divide more like a Yeti (the Abominable Snowman) in statistics? Something that everybody talks about, but nobody has really seen it and knows much about what it really is?
The usual understanding is that “digital divide” is the imbalance between something that has to do with ICT. And once those numbers, describing such an imbalance do exist, the amount of digital divide (or: level, ratio, coefficient or whatever we call it) can be exactly calculated, using traditional statistical methods.
But the real problem is somewhere else: what exactly should we measure, if we want to evaluate digital divide? Is digital divide the difference in Internet penetration between various countries? Or the difference in fixed phone penetration, mobile phone penetration, PC penetration etc.?
Yes, such statistics do exist, and they are even presented as “digital divide statistics”, see example. And they surely are significant indicators that should be watched and evaluated carefully. But they deal more with economical issues: with the distribution of something (goods and services), and with the take-up: how many people buy or rent this or that.
At least in my opinion, digital divide should deal with more “human” aspects than economical and industrial factors. With knowledge, abilities and skills. But then it is even more difficult to provide a precise definition of what should be measured and evaluated. Fortunately, the task is not impossible.
In my next article, I will discuss the results of one Eurostat survey, monitoring the abilities of EU citizens to perform specific tasks on the Internet. They show some interesting differences between Visegrad countries, and at least in my opinion are more close to how “digital divide” should be interpreted.
But perhaps the most significant result, not only from these particular results but also from discussions with people from statistical offices, is that there is not “one digital divide”, but “many digital divides”. And in the eternal quest for a definition of digital divide, this should be kept in mind. One digital divide is not enough.
Hi Jiri,
You tackle a problem typical for the statistical assessment of “unmeasureable” issues; my post on academic ranking dealt with a similar problem. Fact is, that there are people who don’t work with computers or do not have one at home and/or don’t use the internet etc. and/or do not know how to operate these devices/systems. The percentages vary, the show up in surveys such as “e-readiness Index” etc. A close relative of mine still uses a phone book, I don’t even have one anymore. “Computer Literacy” is a keyword in this context, and I think this is definitely an indicator of the Digital Divide discussion. I ask myself sometimes whether Literacy is the correct expression as many people who know how to make use of computer and internet cannot write correctly. Am I “literate”, if I can use the internet but do not have books at home? If I use an MP3 player but don’t know what happens in the world as I don’t read newspapers? (Those who don’t read them physically don’t read online either…)
There was this huge conference on “e-Inclusion” last week http://ec.europa.eu/information_society/events/e-inclusion/2008/conference/index_en.htm to improve Computer Literacy in Europe. I am not sure anymore if owning and using digital devices is really linked to litercy…
Another side of this issue is the concept of ‘bit literacy’, coined by Mark Hurst. His point is that many people – even the most wired people – just don’t know how to manage the flood of information they are already consuming. How do we extract real meaning and value from all the raw data?
Mark’s solution boils down to knowing how to use the delete key, and being selective about what data you consume and archive.
Here’s the site:
http://bitliteracy.com
And here’s a preview chapter:
http://bitliteracy.com/chap1.html
I am doing a project involving the digital divide in Baton Rouge, Louisiana, US for a quantitative methods class. I am testing for a correlation between high-tech device ownership and socioeconomic and demographic variables. The device ownership was measured by wardriving with Kismet on every street in 8 census block groups. The number of detected Wifi networks per block group was divided by the number of households in the block group. The presence of a Wifi router suggests ownership of broadband and one or more wireless devices. The ratio of networks to households is used as an index that can be compared to other data with regression.
There is more to the project than that, but it has its limitations. However, results from those 8 census block groups (3960 households and 854 networks in total) showed 9 variables with R-squared > .50 with linear regression. The number of networks increases with percentage of whites, vehicle ownership, being married with or without children, and with income. Number of networks detected decreases with percentage of blacks, percentage of children under the age of 5, percentage of single-parent households, and with mean household size. The percentage of whites had the highest R-squared with .897. Interestingly, median household income had the lowest R-squared (above .5) with .503