AI-Driven Journalism in Iran: Challenges and Opportunities
Description
51 statements for analyzing the opportunities and challenges of using artificial intelligence in journalism have been extracted from the analysis of the responses of the panel members based on the Delphi method.
Files
Steps to reproduce
Research Design and Data Collection Summary This research investigated the role of artificial intelligence (AI) in journalism in Iran using the "Delphi method", a structured approach for building consensus among experts. The method's iterative nature facilitated informed discussions, ensuring validity and reliability in reaching consensus on 51 AI-related statements. Panel Composition The panel consisted of 25 experts selected through purposive sampling to ensure diverse perspectives: - "10 journalists" with extensive media experience. - "4 media managers" familiar with organizational strategies. - "8 university professors" specializing in media and technology. - "3 AI experts" experienced in AI applications within media contexts. These participants were chosen based on their expertise in journalism, media management, and AI technologies. Survey Instrument and Protocols The study's survey included 51 statements developed from: - Literature reviews on AI in journalism. - Preliminary expert interviews to identify challenges and opportunities. - Theoretical frameworks such as Actor-Network Theory (ANT) and Diffusion of Innovations (DOI). The Delphi process included three rounds: 1. Round 1: Experts rated their agreement with each statement on a Likert scale (1-5) and provided qualitative feedback. 2. Round 2: Aggregated responses and anonymized feedback from Round 1 were shared with the panel, allowing participants to adjust their responses. 3. Final Round: A summary of scores and comments was circulated to confirm consensus. Statements achieving 75% or higher agreement (19 or more members) were considered as reaching consensus. Tools and Software - Survey Distribution: Online forms ensured accessibility and participant anonymity. - Data Analysis: - Quantitative data (Likert scale responses) were analyzed using "SPSS". - Qualitative feedback was coded and thematically analyzed with "NVivo" to identify key patterns and divergences.