WHAT is Being Taught in the IntroDS Course?
(Data collected from 2022 - 2024)
Prerequisites
Based on the data collected, 66% of IntroDS courses have no required prerequisite, and of those with a required prerequisite statistics is the most common.
Technology
Instructors were asked about technology use (exclusively, significant, we touch on it, or not at all) in their IntroDS course. The graph below summarizes exclusive + significant technology use by prerequisite of the course.
Data Science Definitions
Definitions were coded into seven conceptual themes: workflow, knowledge extraction, interdisciplinary thinking, problem solving, sensemaking, data characteristics, and communication. The heatmap summarizes the percentage of courses within each prerequisite group whose definitions included each theme, illustrating how instructors conceptualize data science for students with different prerequisite backgrounds.
The prevalence of each conceptual theme across prerequisite groups is shown above. While all groups described data science as an interdisciplinary field to some extent, the relative emphasis on workflow, knowledge extraction, sensemaking, and data characteristics differed by prerequisite structure.
More details on the DS definition thematic analysis can be found here.
Time Spent
There is a lot of variability in the amount of time allocated to different topics among IntroDs courses. However, there is no discernible difference in the time spent on topics across prerequisites. The following figure displays the time spent by prerequisite of the course, and the topics with diagonal lines represent core data science topics.
Specific Topics / Skills
The following graphics summarize information collected from the IntroDS Course Survey. Respondents were asked to indicate whether the following skills/concepts were taught in their course. The summaries are categorized by having a CS prerequisite, a Stat prerequisite, or no CS or Stat prerequisite.
Learning Objectives in IntroDS
Learning objectives from n = 29 introductory data science syllabi were analyzed using qualitative content analysis and coded into 11 instructional themes. Multiple themes could be assigned to each syllabus using binary coding (0 = absent, 1 = present). Courses were grouped by prerequisite structure (None, Computer Science, and Statistics) to examine how prior preparation influences the emphasis of introductory data science instruction. Note syllabi from the STAT+CS group were not submitted. The figure below summarizes the percentage of syllabi within each prerequisite group that included each instructional theme.
Key Findings
Across all prerequisite groups, introductory data science courses emphasized a shared instructional core centered on data preparation, visualization, statistical thinking, communication, and critical thinking. Differences across prerequisite groups reflected the depth and emphasis of instruction rather than entirely different curricular content.
Courses with No Prerequisites
Courses without prerequisites emphasized broad data literacy and foundational analytic skills. Learning objectives most frequently focused on visualization (58%), communication (58%), critical thinking (58%), and data preparation (47%), with comparatively less emphasis on modeling and professional software practices. These courses introduced students to the complete data science process while developing confidence working with unfamiliar data.
Computer Science Prerequisite
Courses requiring a computer science prerequisite placed greater emphasis on modeling and prediction (60%), problem solving (60%), and communication (60%), while maintaining strong coverage of data preparation (60%) and statistical thinking (60%). Computational skills were introduced when programming itself was an instructional objective rather than simply a tool for completing other tasks. These courses emphasized implementing scalable computational workflows and applying predictive models.
Statistics Prerequisite
Courses requiring statistics prerequisites emphasized statistical thinking and inference (60%), visualization (60%), data preparation (60%), and critical thinking (60%). Learning objectives focused on inference, uncertainty, methodological justification, and interpreting evidence. Modeling was included but generally framed as an extension of statistical reasoning rather than purely predictive computation.
More deatails on the learning object thematic analysis can be found here.
Data Science Program Topics
As part of the MASDER project, we also worked to synthesize the Data Science curriculum (beyond just IntroDS). The final synthesis of the information can be found in the interactive Data Science Program Topics Shiny App.