Appendix B — The Peer-Program Survey
Several chapters of this book justify their inclusion by reference to a survey of the graduate curricula of peer United States biostatistics programs: Chapter 6 on the command line, Chapter 3 on de-identification, Chapter 21 on SQL, Chapter 23 on missing data, Chapter 14 on pipeline tools, Chapter 27 on continuous integration, and Chapter 29 on SAS. This appendix records the sources on which those claims rest, so that a reader can check them.
B.1 Scope
The survey covers master’s-level curricula in biostatistics and allied statistical-computing courses at United States institutions. It was conducted by consulting the public course catalogues, syllabi, and course pages listed below. Where a program’s catalog was the only public record, the catalog is cited; where an individual course published a syllabus, the syllabus is cited in preference, because it states the software and the topics rather than merely the title.
Two properties of the evidence base deserve statement, because they bound what the survey can support. First, a public catalog records what a program says it teaches, not what is taught; a topic covered in a single lecture and a topic given a whole course both appear as a line in a syllabus, and the survey cannot distinguish them except where the syllabus is detailed. Second, curricula change annually, and the sources below were consulted in the years their entries record. A claim of the form ‘most peer programs teach X’ should be read as a claim about the published curricula at that time, not as a durable fact about the field.
B.2 Sources consulted
Course-level sources, where a syllabus or course page was available:
| Institution | Course | Source |
|---|---|---|
| Johns Hopkins | Statistical Computing (140.776) | (Johns Hopkins Bloomberg School of Public Health, 2024) |
| UNC Chapel Hill | Statistical Computing (BIOS 735) | (University of North Carolina at Chapel Hill, 2024) |
| U. Washington | Statistical Computing (BIOST 561) | (University of Washington Department of Biostatistics, 2024) |
| U. Washington | Statistical Methods for Missing Data (BIOST 531) | (Sadinle, 2019) |
| U. Michigan | Statistical Computing (BIOSTAT 615) | (University of Michigan School of Public Health, 2024b) |
| U. Michigan | Missing Data (BIOSTAT 880) | (University of Michigan School of Public Health, 2024a) |
| Columbia | Advanced Statistical Computing (P8160) | (Columbia University Mailman School of Public Health, 2025) |
| UC Berkeley | Introduction to Statistical Computing (STAT 243) | (University of California, Berkeley Department of Statistics, 2024) |
| U. Florida | Statistical Programming (PHC 6089) | (University of Florida College of Public Health and Health Professions, 2025) |
Program-level sources, where the public record is a catalog or curriculum page rather than a syllabus:
| Institution | Source |
|---|---|
| Yale | (Yale School of Public Health, 2024) |
| Emory | (Emory University Rollins School of Public Health, 2024) |
| U. Minnesota | (University of Minnesota School of Public Health, 2024) |
| UCLA | (UCLA Fielding School of Public Health, 2024) |
| Duke | (Duke University Department of Biostatistics and Bioinformatics, 2024) |
| Vanderbilt | (Vanderbilt University Department of Biostatistics, 2024a) |
| U. Iowa | (University of Iowa College of Public Health, 2024) |
| U. Pittsburgh | (University of Pittsburgh School of Public Health, 2024) |
| U. Pennsylvania | (University of Pennsylvania Perelman School of Medicine, 2024) |
| Brown | (Brown University School of Public Health, 2024) |
| Boston University | (Boston University School of Public Health, 2024) |
| New York University | (New York University School of Global Public Health, 2024) |
| Northwestern | (Northwestern University Feinberg School of Medicine, 2024) |
| UT Health Houston | (UT Health Houston School of Public Health, 2024) |
B.4 The Vanderbilt materials
One entry in the program-level table warrants a note, because counting it as a curriculum understates it. The Vanderbilt University Department of Biostatistics publishes, without charge or registration, the largest body of biomedical statistical-computing teaching material that this survey encountered anywhere: R Workflow (Harrell, 2025b), the Biostatistics for Biomedical Research course notes (Harrell, 2025a), the Regression Modeling Strategies course and its accompanying book (Harrell, 2015), the author checklist (Harrell, 2024), and the department’s page on reproducible reporting (Vanderbilt University Department of Biostatistics, 2024b). These were assessed against this book’s chapters and are cited where they bear on one: Chapter 10 and Chapter 17 on workflow and data checking, Chapter 18 and Chapter 22 on information loss and on displaying distributions, Chapter 23 on missing-data reporting, and Chapter 24 and Chapter 32 on the author checklist.
Two differences bound how far that material can be adopted here rather than merely cited. The first is dialect: R Workflow is built on data.table and Hmisc, and this book is tidyverse-first, so its arguments are borrowed and its code is not. Appendix A exists partly so that a reader who follows the citation is not stranded. The second is altitude: BBR and RMS teach statistical methods, which this book deliberately does not, and they are therefore cited as the place to go for the reasoning behind a computing practice rather than as sources for the practice itself.