Artificial Intelligence in South Africa: Survey Data of Master’s Students
Description
The dataset, titled “Artificial Intelligence in South Africa: Survey Data of Master’s Students,” comprises two sheets, namely “Uncoded data” and “Coded data.” Both sheets comprise the raw data provided by the participants. The categorical data gathered via the ranking and statement selection questions were coded to enable variance analysis, regression analysis, and correlation analysis of the data. Therefore, in addition to the raw data, the “Coded data” sheet contains the data code associated with each survey data point. The “Uncoded Data” sheet contains, amongst others, columns for Participant ID, Gender (text), and Academic Year (text). The “Coded Data” sheet mirrors these columns and includes columns with numeric codes. The subsequent section details the coding process. Each row corresponds to a single participant’s responses. As the survey was completed anonymously, unique numeric participant IDs were automatically assigned to the participants’ answers by Microsoft Forms, the survey instrument. The dataset was organized into thematic sections to improve clarity, each corresponding to related questions. The combination of coded categorical variables and demographic information supports a range of statistical techniques, including reflective and formative measurement models, variance analyses, and regression analyses. This versatility allows other scholars to replicate or extend studies on AI, GenAI, and technology acceptance and usage within higher education. By serving as a baseline, this dataset can be integrated into longitudinal research. Future studies can compare these findings with cohorts from different regions or fields of study.
Files
Steps to reproduce
The survey was adapted from the survey instrument used by Johnson et al. (https://edintegrity.biomedcentral.com/articles/10.1007/s40979-024-00149-4), with slight modifications to suit our specific context. As their study included 2555 student responses, the survey questions were assumed to be well-tested and verified. The survey instrument was developed on Microsoft Forms, and participants were invited to participate via email. All the students enrolled in the program were invited to participate. Participation was entirely voluntary and anonymous. The survey was open for one month.