Synergizing Qualitative Research: The Role of Artificial Intelligence (AI) in Online Focus Group Transcription and English Translation

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Continuous advancement in artificial intelligence (AI) and large language models (LLMs) has considerably transformed the process of collecting and analysing qualitative data. The aim of this literature review is to assess and provide evidence on the merits of synergizing AI in conducting research involving focus groups in qualitative studies, through the production of transcripts, and translating transcripts written in different languages into English. The key concepts examined in the literature review include the use of AI/LLM models in data recording and transcription, and the use of AI in translation and analysis. A review of current literature shows that researchers utilize AI and LLM tools such as Google AI and ChatGPT to record and translate into English focus group transcripts. The main benefits attributed to the use of these technological tools in qualitative research are that less time and finances are used to collect and analyse transcribed data. On the use of AI/LLM models in recording and transcription, the literature review demonstrates that the adoption of the technologies in qualitative research improves participant engagement in online data collection processes for focus groups. Applications such as AI-supported natural language processing (NLP) and LLMs enable data translation in large-scale, multilingual qualitative research. NLP also helps to save costs and time associated with manual transcription, especially in large multilingual research projects. Reliable audio considered to be credible research evidence for analysis can also be generated using modern NLP models. Furthermore, the use of AI or data transcription in online focus group research enhances the accuracy and applicability of the generated data. Literature evidence highlights that modern AI/LLM tools can be used to attain more than 90% accuracy in data transcription to generate reliable data for initial drafts. On the use of AI for data translation, the literature review indicates that advanced LLMs improve efficiency and fluency in data translation, but that human moderation is essential to detect errors since some AI models are unreliable in translating complex sentences with culture- specific phrases. In summary, the literature review emphasizes that the use of AI, LLMs, and NLP models in online focus group research makes it easy to record, transcribe, and translate data.