Accuracy and Efficiency of Artificial Intelligence Tools for Transcription in Qualitative Research
American Journal of Qualitative Research · 02.11.2026 · 10 (4) · 266-290
DOI: 10.66815/ajqr/19595 ↗Kurz gefasst
Qualitative Studie in der Zeitschrift American Journal of Qualitative Research, erschienen am 02.11.2026. Verfasst von Monique Hennink und 4 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben.
Abstract (Originalsprache)
Verbatim transcription in qualitative research is hugely time consuming, yet critical for data validity. The rise in artificial intelligence tools for transcription provides new opportunities for researchers. We assessed the accuracy and efficiency of two widely accessible AI tools and human transcription to provide empirical guidance for researchers. We compared 10 verbatim transcripts generated via three transcription modes: human, Otter.ai and MAXQDA. We compared transcription time, cost, errors (volume and type), and correction time. Results show that the AI tools were highly time/cost efficient but had 2-3 times more errors than human transcription. Accounting for average correction time, AI tools generated a transcript in 1.25 - 2.5 hours for USD31-38 per transcript; thus 3–6 times faster and 3 times cheaper than human transcription. However, human review of AI transcription remains critical for data validity; thus, AI tools can assist in the transcription process versus generating the final product. We develop recommendations for each phase of the research process to maximize the benefits of using AI transcription tools.
Zitieren
APA
Hennink, M., Kazzi, B., Yoon, E., Alami, N., & Farhat, G. N. (2026). Accuracy and Efficiency of Artificial Intelligence Tools for Transcription in Qualitative Research. American Journal of Qualitative Research, 10(4), 266-290. https://doi.org/10.66815/ajqr/19595
Vancouver
Hennink M, Kazzi B, Yoon E, Alami N, Farhat GN. Accuracy and Efficiency of Artificial Intelligence Tools for Transcription in Qualitative Research. American Journal of Qualitative Research. 2026;10(4):266-290. doi:10.66815/ajqr/19595
BibTeX
@article{Hennink2026,
author = {Hennink, Monique and Kazzi, Bahaa and Yoon, Erin and Alami, Nael and Farhat, Ghada N.},
title = {Accuracy and Efficiency of Artificial Intelligence Tools for Transcription in Qualitative Research},
journal = {American Journal of Qualitative Research},
year = {2026},
volume = {10},
number = {4},
pages = {266--290},
doi = {10.66815/ajqr/19595}
}
EUVIMED