Resources
The instruments, the material repositories, and the questions adopters ask before they commit.
Assessments
Common instruments were used across all eight institutions, so results can be compared between sites — and so you can compare yours against theirs.
The shared course-wide instrument: student background and demographics, self-reported experience with general computing and with PDC topics, interest, perceived importance and understanding.
Used at all eight institutions · what it asks · Chapter 1 § 1.5
Short knowledge checks bracketing an individual activity, so its effect can be isolated from the rest of the semester.
An engagement instrument adapted for unplugged activities, with results grouped by aspect category. Used with Penny Sorting and Flag Maker.
Per institution, a table mapping each PDC concept to the Bloom level reached by each activity — the fastest way to check an activity against your own learning outcomes.
Tables 2.1, 4.1, 5.1, 6.1, 7.1, 8.1, 9.1 · find the chapter →
Chapter 1’s activity finder gives the time each activity takes and which institutions ran it, in one place — so you can fit an activity into a real class period rather than discovering the overrun live. Table 1.1 covers the activities described in that chapter; Table 1.2 the additional institutional ones.
Every chapter repository carries an Evaluation_Data_and_Analysis directory with survey
instruments, de-identified responses, evaluation metrics and statistical analyses.
Pre-selection, post-selection, annual and post-adoption instruments for the instructors themselves — prior PDC background, perceived effectiveness, students taught, and whether interventions will be retained, reduced or expanded.
Informed-consent forms alongside the surveys, quizzes, lab reports and programming-assignment reports collected between Spring 2024 and Spring 2026.
Survey design
Students rate their experience with basic programming skills, then their familiarity with PDC concepts, then a short set of attitude statements. All items use a seven-point Likert scale: programming and PDC concept questions run from 1 = no experience at all to 7 = extremely experienced; attitude questions from 1 = strongly disagree to 7 = strongly agree.
Because every site ran the same items, an adopter can administer them locally and compare directly against the published per-institution tables.
FAQs
Flag Maker or Penny Search, unplugged. Both need only paper, markers or coins, run in a single 40–75 minute session, require no programming background, and surface speedup, load balance and race conditions concretely. Start with Chapter 1, then look at whichever institutional chapter most resembles your class size.
That was the explicit design constraint — the basic CS1 learning outcomes are preserved. Two checks are available. Casper’s data shows perception of learning gains in non-PDC topics stayed broadly flat between baseline and intervention semesters while PDC gains rose. MSU compared a no-intervention semester against the Flag Maker intervention semester and found gains across all basic programming categories in both, with comparable post-survey means.
Three broad strategies, often combined: selectively reducing or deferring existing content, modifying current assignments and activities, and adding short PDC-focused modules. Topics commonly reduced included binary and random-access files, C-style strings and arrays, detailed output formatting, two-dimensional arrays and advanced class relationships — usually judged coverable more briefly or revisitable in CS2.
Some institutions removed nothing at all. One placed three asynchronous PDC modules outside class time; another used existing lab and recitation periods. Scheduled PDC instructional time ranged from none to about seven hours.
Yes. All unplugged activities — Flag Maker, Penny Search and Sorting, Coin Addition, the More Processors writing activity — are language-neutral, as are the animations and the codeless modules. Casper College’s physical computing projects use Python’s multiprocessing package. A Python CS2 package covering threading, multiprocessing, OpenCV image processing and performance engineering is planned for the CS2 release.
The development teams reported more than 100 hours across the initial development year — creating examples, finding curriculum insertion points, revising assignments, testing software and datasets, and refining materials. Adopting already-developed modular materials substantially reduces that, but still requires local adaptation, instructor preparation and refinement based on student responses.
Budget separately for coordinating with lab instructors and teaching assistants: they need enough familiarity with the concepts and tooling to guide students, and this was repeatedly named as one of the harder parts.
Not to teach them. You need IRB approval if you intend to collect and publish student data. Each testing team’s IRB experience and instrument design is documented in the chapters if you want to run your own study.
The set was chosen to span community college, liberal arts, private undergraduate, public master’s, mid-sized public and R1, on both semester and quarter calendars, with class sizes from about 24 to about 100 and student populations including dual-credit high-school students and part-time adult learners. Filter by the constraint that binds hardest for you — usually preparation time or class size, not institution type.
Anticipate differences among student computers, compiler configurations, OpenMP support and execution timing — all of which the teams hit. Before using any resource, verify that software versions, API links, datasets and institutional assessment requirements are still current. The unplugged and codeless pathways exist precisely to avoid this class of problem.
Yes — some consultation from project personnel is available to help adopters select and adapt materials. Request adoption help.
Finding material quickly
The fastest way to use the volume is to combine chapter-level navigation with search. The table of contents gives the major chapters and sections; each institutional chapter opens with an abstract and descriptor elements for a quick read of what it covers; and PDF search finds activity names, PDC concepts, tools and evidence terms wherever they appear.
Flag Maker · Penny Search · Penny Sorting · OpenMP · Earthquake Tracker · Greenfoot · animation · visualization · data parallelism · speedup · load balance · event handling · remote data · API
pre/post survey · student feedback · engagement · assessment · reflection · data analysis
Beyond this project
The NSF/IEEE-TCPP topic list with Bloom levels that defines the scope this project draws from. Version I (2012) and Version IIβ (2020).
The wider CDER collection of community-contributed teaching material — syllabi, slides, assignments and labs across the undergraduate curriculum, beyond this project's exemplars.
Peer-reviewed parallel programming assignments, selected and presented each year at EduPar and EduHPC — ready to drop into a course once students are past the introductory sequence.
Volumes 1 and 2 (Morgan Kaufmann, Springer) available as free preprint chapters; Volume 3, focused on adopter experience and resources, expected from Springer in late 2026.
Syllabi, handouts, slides, assignments, starter code, surveys and anonymized evaluation data for every chapter — curated periodically as adopters revise and extend them.