Table 1 Steps of Critical Discourse Analysis.
Step | Process | Application and Example |
|---|---|---|
1 Description | Description involves the textual analysis of the selected document, with a focus on the choice of words, grammar and structure of the text and its arguments. The focus is on identifying how language is used to construct meaning, as well as how word choices and structure may communicate certain underpinning ideologies. There is no definite list of devices that could be used, but typically these are: word order, word choice, voice (passive/active), tense, coherence (Mullet, 2018). | • Familiarisation with the document through multiple readings • High level notation to identify passages of relevance and interest • Micro-level analysis, highlighting word choices and structures (e.g., will+infinitive to express confidence and certainty of outcomes), choices between active/passive voice, references to citizens versus to industry. • Example: We highlight and coded excerpts such as “deep pools of capital” and “excellent funding ecosystem”, “we understand”, “critical importance”. |
2 Interpretation | Interpretation focuses on how the document may be or is interpreted by the intended audience, focusing specifically on said audience’s prior experiences and knowledge. Macro-coding. | Using the examples provided above, we focused on identifying the intended audience of each of these examples, with the aim to explore possible hidden meanings. For instance, “deep pools of capital” are aimed at industry and market actors, with the intention to communicate that funding exists and can be made available to digital businesses. Further, other elements, such as “we understand”, immediately followed by “critical importance” communicates interest alignment and expertise. |
3 Explanation | Explanation moves Interpretation beyond the analysed document, where the latter is contextualised within the wider political and social context. It specifically entails examining power relations, ideological underpinnings, social structures, and tensions between different audiences and conflicting agendas and priorities. Macro-coding. | • Extensive review of relevant literature (e.g., Avgerou & Bonina, 2020; Cooper et al., 2021; Eubanks, 2018; Hustad et al., 2019; Mandelbaum, 2020; Vassilakopoulou & Hustad, 2023). • Contextualisation based on the review of auxiliary policy-related documents (e.g., Black et al., 2023; House of Lords, 2023; Tobin, 2023) and other relevant UK policies and strategies (DCMS, 2017; Department of Education, 2019; DHSC, 2021; DHSC and NHS, 2022; GOV.UK, 2013; HM Government, 2021). • Example: In the “Interpellation of citizens as national subjects” we discuss that the document portrays the UK as a digital powerhouse, where this status needs to be maintained, and that this requires further digitisation and investments in technology. This interpretation was confirmed and validated through the UK AI Strategy document (HM Government, 2021). |
4 Review and production of chains of evidence | Review entails moving back and forth between the three previous steps, continuously refining the analysis via consultation, extended literature review and reading of other documents. | • Consultation between the authors to address discrepancies, differences and possible misinterpretations. • Relabelling of themes where needed. • Excerpts from the document were chosen as exemplars. • Developed chains of evidence and supplementary material (Supplementary Tables 1, 2, and Supplementary Fig. 1) • Findings triangulation (with 2 external experts in CDA) |
