The study looks at how many and which faculty are using artificial intelligence in data management, and how they are using it, currently, and planned for the future. The study presents unique data sets for use of AI in many specific facets of data management, including, but not limited to data visualization, data cleaning, finding data, data summary, as a plug in with stat programs, hypotheses generation, data analysis, data upload, data tagging or cataloging, data training and other uses. In addition, survey respondents evaluate their interest in specific generative AI applications, and their level of trust in AI-generated results. Respondents also rate the usefulness of AI in data management, evaluate their impact on the productivity of their data-driven efforts, and comment on their plans.
Just a few of this 129-page report’s many findings are that:
-
About 11% of respondents reported that their academic libraries offered training in the use of AI applications in data management.
-
2.36% of respondents felt that their productivity in data management had increased by more than 50% though the use of AI tools.
-
Non-tenured faculty who were on a tenure track were much likelier than tenured faculty or those not on a tenure track to feel that AI was extraordinarily useful in data management.
-
Faculty in physics showed the highest use of AI in data visualization.
Data in the report is based on a survey of 339 higher education faculty, predominantly in STEM subjects at research universities. Data in the report is broken out by many institutional and personal variables, enabling the study’s end users to pinpoint use of AI in specific facets of data management by academic field, college type, age, gender and other variables. Data is based on a survey of 339 faculty, drawn predominantly from research universities and medical schools.