Overview

We propose the creation of a new authorable content type in Open edX called Diagnostic, designed to assess a learner’s prior knowledge before starting a course. The goal for this new content type is to allow for diagnostic results to personalize the learning path, starting with simple entrance gating and evolving toward advanced adaptive progression, where learners can skip content they already master and focus on what they still need to learn.

Problem

Learners often enter courses with varying levels of prior knowledge, but Open edX currently lacks native tools to evaluate their readiness or tailor their path accordingly. This results in repetitive learning, disengagement, and wasted time, especially for returning or advanced learners.

Use cases

• As a learner, I want to demonstrate what I already know so I can skip unnecessary content and focus on what I still need to learn.

• As a learner, I want to confirm if I’m ready to take a course before enrolling, so I don’t waste time on something that’s too advanced or too basic for me.

• As a course author, I need to build diagnostics that allow me to:

  • Restrict access to learners who meet certain knowledge criteria,

  • Mark parts of the course as completed based on mastery,

  • Or route students to more appropriate alternatives if needed.

• As a platform admin, I want to offer adaptive, personalized learning experiences that improve engagement, retention, and outcomes.

Supporting market data

Extensive research supports the use of diagnostic assessments as a foundation for improving learning outcomes and enabling personalized instruction. A meta-analysis found that diagnostic and formative assessments have a strong positive effect on student achievement. In Competency-Based Education (CBE), diagnostics are essential to determine prior mastery and accelerate learning: for instance, Texas-based CBE programs use diagnostic pre-tests to allow learners to skip competencies they already demonstrate. Additionally, studies show that initial diagnostics help reduce academic dropout rates by identifying students in need of early support. A popular consumer example is the Duolingo Placement Test, which adaptively assesses a learner’s language level at the start of a new course and places them at the appropriate skill checkpoint, allowing users to skip redundant lessons and engage only with content they haven’t yet mastered. These cases confirm that diagnostics are not only valuable as gatekeepers but also as drivers of efficient, personalized, and engaging learning experiences.

Proposed solution

We propose a new Studio authorable content type: Diagnostis separate from courses, libraries, and taxonomies. This content type allows authors to create pre-course or in-course assessments that measure a learner’s prior knowledge and trigger specific rules for progression or access:

• In the simplest case, gating access to courses or sections based on minimum scores.

• In more advanced use, linking diagnostic outcomes to completion of specific modules, so learners can skip content they already master.

• Ultimately, recommending or unlocking entirely different courses, based on the learner’s diagnostic profile.

Importantly, this design ensures that diagnostics are not limited to a single course—they can be linked to multiple content objects across the platform. For example, a single physics diagnostic could place learners or determine partial completion for an entire department’s set of physics courses. This cross-course flexibility supports more strategic, scalable, and personalized learning pathways, meeting institutions wherever they are on the path from basic entry exams to more advanced, tailored learning experiences.

Other approaches considered

We considered extending or improving entrance exams. While this allows gating access to a course, it falls short of delivering true personalization. It doesn’t support skipping content, linking to specific modules, or recommending alternative courses. The proposed Diagnostic content type goes further, adapting the learning path based on what the learner actually knows, not just if they’re allowed to start.

Furthermore, entrance exams are embedded within individual courses and can’t easily support cross-course or cross-content use cases. In contrast, diagnostics as a separate content type can operate at multiple levels: they can be linked to a single course as an entry gate, or flexibly aligned with multiple learning objects across a program or department. This flexibility supports more scalable and comprehensive placement strategies. For example, a single diagnostic could place or advance learners across a suite of related courses.

Competitive research

Several major platforms have introduced features that partially address the need for pre-course diagnostics and adaptive learning paths, though none currently offer a unified, built-in diagnostic content type that enables flexible, multi-stage learner routing as proposed here.

Moodle – Conditional Activities

Moodle allows instructors to set up conditional activities, where access to certain resources or modules depends on predefined conditions, such as scores from quizzes or completion of previous tasks. While this allows some adaptation, it is rule-based and linear. There’s no native support for diagnostics that evaluate multiple competencies and trigger deeper personalization like marking modules complete or recommending alternate learning paths based on results.

Official Moodle Docs on Conditional Activities

Open

Canvas – Mastery Paths

Canvas offers a feature called Mastery Paths, which enables instructors to create conditional content sequences based on students’ performance on an initial quiz or assignment. Learners are branched into different sets of content depending on how well they score. While powerful, Mastery Paths are limited to content within a single course and require manual configuration. They do not function as diagnostic tools that can conditionally mark parts of the course complete or skip whole sequences.

Canvas Mastery Paths Overview

Open

Coursera – Coach AI and Adaptive Diagnostics

Coursera has introduced Coach AI, a learning assistant that uses AI to provide personalized guidance, recommendations. Coursera Coach offers reactive, AI-driven guidance based on learner behavior, while the proposed Diagnostic content type enables proactive, instructor-defined assessments that control learning paths based on demonstrated prior knowledge.

Advance your learning with Coursera Coach

Implementation Plan

• Lead organization: Schema Education (design, development, and coordination).

• Funding: Seeking collaboration and co-funding from interested Open edX community members and partners.

We propose an iterative development approach, with each stage adding deeper functionality to the Diagnostic content type:

Stage 1: Diagnostics as basic entrance gates for courses

• Develop a diagnostic authoring tool that mirrors the course unit page authoring experience, allowing course authors to create simple assessments that determine whether a learner can access a course based on a minimum score or set of criteria.

• Enable linking of diagnostics to courses as prerequisites, extending beyond the current entrance exam functionality to ensure minimum preparedness before enrollment.

• Allow for diagnostic questions to be aligned to content taxonomies similarly to how course content blocks can be aligned to tags.

Stage 2: Diagnostics as in-course progression accelerators

• Expand diagnostic authoring to support multi-page assessments that evaluate a range of skills in more depth.

• Allow course authors to link specific diagnostic questions to course sections or subsections, and also to overall diagnostic results that determine if certain sections can be skipped entirely.

• Let learners skip sections or modules that teach skills they already master, ensuring they focus only on the content that adds the most value to them. This aims to improve engagement and reduce dropout rates, especially for advanced learners.

• Provide learners with visibility into what learning content modules have progression / completion linking possibilities to help communicate the value of diagnostics up front.

Stage 3: Advanced diagnostic exam creation tools

• Enhance authoring capabilities to allow course authors to create more sophisticated types of diagnostic exams, such as tests that modify the difficulty of subsequent questions based on a learner’s answers.

• While we have considered many other improvement areas for diagnostics to support adaptivity or personalization, we have left those ideas for future proposals / product idea submissions.

Plan for long-term ownership/maintainership

Would be maintained by the Open edX Studio maintainers.

Open questions for rollout/releases

• How could / should Diagnostic outcomes be stored in the learner’s profile for reuse across courses?

• How should authors define rules via UI?

• What options exist for providing learner visibility to these diagnostics: visible from learner dashboard, enrollable via catalog listing, discoverable within courses linked to that diagnostic, other options?