E-Thesis 70 views
Dimensions of Algorithmic Bias: Exploring its Influence on Organisational Decision-making Processes in Saudi Arabia / RAZAN ALOWAYFI
Swansea University Author: RAZAN ALOWAYFI
DOI (Published version): 10.23889/SUThesis.71643
Abstract
Algorithms are now central to organisational decision-making, influencing how platforms are organised and resources allocated based on complex data. Their opaque nature, however, often hides decision-making processes, raising concerns about fairness and potential discrimination. Against this backgro...
| Published: |
Swansea
2026
|
|---|---|
| Institution: | Swansea University |
| Degree level: | Doctoral |
| Degree name: | Ph.D |
| Supervisor: | Dennehy, D., Dwivedi, Y. K., and Cotterell, D. |
| URI: | https://cronfa.swan.ac.uk/Record/cronfa71643 |
| Abstract: |
Algorithms are now central to organisational decision-making, influencing how platforms are organised and resources allocated based on complex data. Their opaque nature, however, often hides decision-making processes, raising concerns about fairness and potential discrimination. Against this background, this doctoral study examines how algorithmic bias emerges, is experienced and is managed within Saudi organisations during a period of rapid digital transformation shaped by Vision 2030. The study investigates the socio-technical, organisational, ethical and cultural conditions that influence fairness, transparency and accountability in algorithmic decision-making. An interpretivist, inductive qualitative design was adopted, using a multiple case study strategy and drawing on semi-structured interviews with thirty-five practitioners across public, private and semi-government sectors. Data from interviews, observations and documentation were analysed using thematic analysis and the Gioia methodology to produce a structured interpretation of practitioner experiences. The findings identify four interrelated dimensions of algorithmic bias:algorithmic pollution arising from poor contextual alignment and data quality; fairness concerns linked to opacity and uneven oversight; racial and gender bias embedded in training data and organisational processes; and broader organisational and societal implications affecting legitimacy and trust. These issues are shaped by local linguistic and cultural factors, uneven governance maturity, manual bias detection practices and reliance on imported technological systems. Alongside these challenges, the study shows that algorithmic systems can provide operational benefits when supported by strong governance and human oversight. The thesis contributes a contextually grounded framework for responsible algorithmic governance in Saudi Arabia, demonstrating that algorithmic bias is not solely a technical failure, but a systemic issue shaped by organisational routines, institutional expectations and societal context. The study offers practical recommendations for leaders, regulators and technology developers seeking to strengthen fairness, transparency and accountability in AI systems and provides a foundation for future research on responsible AI in rapidly developing digital economies. |
|---|---|
| Keywords: |
Algorithmic Bias; Organisational Decision-Making; Algorithmic Governance; Socio-Technical Systems; Saudi Vision 2030 |
| College: |
Faculty of Humanities and Social Sciences |
| Funders: |
Imam Mohammad Ibn Saud Islamic University. |