AI and Investigative Writing: Human Expertise Remains Crucial

Artificial intelligence (AI) is increasingly establishing itself as a potential ally in survey writing, but it still cannot do without human expertise. Although generative AI tools, such as ChatGPT and Claude, are capable of producing survey drafts in record time, their use requires human vigilance to maintain data quality.

User experience (UX) researchers are particularly drawn to these generative AI chatbots due to the growing pressure to quickly deploy surveys. With a well-defined research objective and instructions based on best design practices, these tools can generate a solid first draft in just a few seconds.

However, a solid draft alone is not enough. AI tools, while effective in certain aspects of survey design, have shortcomings in other crucial areas. It is essential to know where AI excels and where it fails to prevent issues from compromising the quality of the survey and, consequently, the research insights.

Strengths and Limitations of Generative AI in Survey Creation

When guided by precise instructions, generative AIs can excel in several fundamental aspects of survey writing. They are capable of:

• Formulating clear and concise questions
• Proposing suitable response formats
• Structuring the survey logically

Nevertheless, to ensure the quality and relevance of surveys, human review remains indispensable. This intervention allows for the detection and correction of design subtleties that AI might overlook, thereby ensuring reliable and actionable data.