DPHU1007 Chap.10 Digital Divides and the Algorithmic Black Box
Digital Divides and the Algorithmic Black Box
Week 10 asks who benefits from digital media. The course quotes Kvasny's definition of digital inequalities as disparities in access to and use of information and communication technology, together with the ways existing social inequalities shape expectations about it.
Van Deursen and Helsper distinguish three levels: unequal access, unequal skills and uses, and unequal capacity to turn internet use into favourable offline outcomes. Lutz summarises research showing that digital inequalities tend to mirror existing ones in socio-economic status, education, gender, age, location, employment and race.
The second half of the week opens the algorithmic black box.
Bucher defines a black box as a device whose inputs and outputs are known but whose inner workings cannot be observed, and Willson notes that many algorithms are proprietary. Gillespie shows how algorithms are presented as impartial, how categorisation is a political act and how some publics exist only because an algorithm called them into being.
Mansell's reading links these themes to declining transparency and accountability.
The chapter models an analysis of an invented tenant-ranking platform.
Where this shows up in assessment. Digital divides suit essays drawing on personal experience, since many students can compare their own access, skills and outcomes with those of family members.
The black box suits stimuli about platforms, recommendations or automated decisions, especially when paired with a discourse about fairness or neutrality.
What this chapter covers
- 01
Digital inequalities: gaps in access and use that reflect long-standing social inequalities.
- 02
First-level divide: unequal access to devices and connection.
- 03
Second-level divide: unequal skills and ways of using technology.
- 04
Third-level divide: unequal ability to convert use into offline benefits.
- 05
Algorithmic black box: known inputs and outputs, hidden inner workings.
- 06
Discourses about algorithms: claims of impartiality that protect automated decisions from critique.
- 07
Algorithmic publics: groups that exist because a system sorted people into a shared category.
Worked example · free
Placing a case on the divide ladder
- +1First level: no gap, because both have the same device and connection.
- +1Second level: a gap in skills and uses, since one student searches and networks strategically.
- +1Third level: a gap in outcomes, since only one converts use into career benefits; the course link to Lutz shows how such gaps often follow socio-economic background.
Key terms
- Digital divide
- Unequal access to, use of and benefit from digital technologies.
- First-level divide
- Unequal access to technological resources such as devices and internet connection.
- Second-level divide
- Inequalities in digital skills and in the ways people use technology.
- Third-level divide
- Gaps in people's capacity to turn internet access and use into favourable offline outcomes.
- Black box
- An opaque technical device of which only the inputs and outputs are known.
- Categorisation
- Deciding what categories exist and what belongs in them, a powerful political intervention.
Digital Divides and the Algorithmic Black Box FAQ
What are the three levels of the digital divide?
Van Deursen and Helsper describe a first level of unequal access, a second level of unequal skills and uses, and a third level of unequal capacity to turn internet use into favourable offline outcomes.
Why is an algorithm called a black box?
Bucher defines a black box as a device whose inputs and outputs are known while its inner workings cannot be observed. Many algorithms are also proprietary, so their parameters and assumptions are hidden from outside critique.
How can talk about algorithms protect platforms?
Gillespie notes that algorithms are positioned as assurances of impartiality. Describing a ranking as neutral mathematics hides the value choices in its categories and makes unfair outcomes harder to contest.
How do digital divides connect to structures of power?
Lutz shows that digital inequalities tend to mirror existing ones in socio-economic status, education, gender, age, location, employment and race, so they can reproduce structures such as capitalism, ageism and racism.
Do I need technical knowledge to discuss algorithms?
No. The course concepts let you argue from inputs, outputs, official descriptions and user experience. Saying clearly that the inner workings are hidden is part of the black box concept rather than a gap in your analysis.
Assessment move
Map your own household or friendship group on the three levels of the divide, noting where people differ in access, skills and outcomes. Read Mansell's article and record one quotation about transparency. Then compare what two people see for the same search or feed, and write a short paragraph explaining the difference without claiming to know the code.
Copy Gillespie's lines on impartiality and categorisation into your concept sheet. Before the essay, practise linking one divide level and one algorithm concept to a structure of power, since this pairing suits questions about challenging power.
