Social Research Methods in Collaboration with Generative AI

Abstract

 As is well known, ChatGPT and other generative artificial intelligence have a significant social impact on the state of knowledge production. This paper discusses the transformation of skills predicted from the perspective of the knowledge required by social research professionals working with generative AI, which has become one of the pillars of universities' institutional research. To what extent can generative AI replace personnel with expertise in social research? If new specialists are required to collaborate with generative AI, what type of expertise would support this?

 Using two frameworks: agenda-setting and framing effects that were developed through mass communication research, the following findings were obtained:

1) Decisions that have a significant impact on the results are embedded in the questionnaire design phase. Social research designers, analysts, and decision-makers must be aware of the decisions that are interwoven into the questionnaire design phase.

2) If a problem is found in the decision embedded in the questionnaire, it will be necessary to retroactively modify the upstream process of selecting the subject and point of view in accordance with the purpose of the research to redesign the questionnaire.

3) By responding to these requirements, social research designers can achieve a high level of collaboration with generative AI, which can automatically create questionnaires.