DDx has introduced a new Performance dashboard for longitudinal learner analytics: a single place to see a learner's progress over time. The Performance dashboard brings to life all the underlying case data and insights in DDx by Sketchy, the AI-enabled clinical readiness platform. DDx delivers realistic, multi-role case simulation across every clinical competency and phase of education. The AI-enabled platform helps develop clinical reasoning, communications, and technical skills through 500+ faculty-built cases scored using its competency-based rubrics. Each case is also built around key findings: specific moments in conversations where asking the right question surfaces the information that guides a learner through the encounter. These key finding results are factored into a score for the case as a whole.
The Performance dashboard takes the rubric-level scoring generated by every case a learner completes and turns it into the longitudinal view faculty, administrators, and learners themselves can act on. Now, the same assessment that scores a single encounter can be used to show whether a learner, cohort, program, or institution is trending in the right or wrong direction (with enough detail for a nuanced understanding).
Before building it, we researched and established what would actually make a longitudinal analytics dashboard most useful to the people who'd rely on it. That work produced three principles, criteria we held to throughout the design. This post introduces the Performance dashboard and the Learner and Administrator views within it, and walks through what each of the three principles looks like in practice.
Key takeaways
- DDx's new Performance dashboard is live now, giving faculty, administrators, and learners a single place to track progress over time instead of reconstructing trends from individual case grades or memory.
- The dashboard is built to answer questions that used to take manual digging, were siloed across individuals, or were inconsistently addressed: exactly where a learner is struggling, broken down by instructional focus; and whether a low score is an isolated incident or a persistent pattern. It also helps answer what to do next: assigning targeted cases for one learner, enriching lecture material for a cohort, or making a curriculum change when a gap shows up program-wide.
- The dashboard is built on three principles: analytics that trace a learner's full arc across the learner, cohort, and program levels; visibility scoped to each audience ; and a design that surfaces data and next steps without making decisions on anyone's behalf.
Why longitudinal data is preferable
A single assessment score has never been a reliable read on where a learner actually stands. The reliability research is specific about why single assessment encounters fall short. One assessor scoring 10 encounters produced a reliability coefficient of just 0.39, while 10 different assessors each scoring one encounter reached 0.83, a reminder that any one grading moment carries a lot of noise, no matter the source (Norcini & Burch, 2010). Faculty have long known this intuitively; it's why a single bad case rarely changes how they think about a strong learner.
That's the problem longitudinal performance analytics is intended to solve: not a better single score, but a clearer view of the trend across every case a learner completes, over time. Faculty don't need to be told to look at trends instead of single scores.They need a system that assembles the trend for them in a digestible and actionable way. Building a new tool to scale this data into actionable longitudinal analytics presented an opportunity for DDx to define a new set of criteria.
What a good longitudinal analytics dashboard has to get right
Any dashboard can help you save time by tracking assignment scores. However, a 2024 analysis cautioned that poorly designed dashboards risk becoming opaque, comparison-driven tools, flagging that when "the source and veracity of the learning recommendations cannot be questioned," dashboards can start to feel like surveillance rather than support (Adv Health Sci Educ, 2024). That's a useful bar to hold any analytics tool to: the goal isn't an algorithm quietly making judgments about a learner.
DDx’s perspective is that three principles comprise a strong longitudinal analytics dashboard design:
- Deep flow-through analytics: DDx takes faculty from a single snapshot to the full arc of learner progress. They can see performance at any single moment or track trends over time — by learner, cohort, case, or instructional focus — with longitudinal analytics built for both real-time and long-range insight. Progress can also be tracked over time at three levels: the individual learner, the cohort, and the program.
- Shared visibility: Instructors, administrators, and learners all see the same underlying data, so every conversation about progress starts from common ground instead of three different accounts of it. Faculty see the courses they teach and only the learners enrolled in those courses. Administrators see every course and every learner across the institution for accreditation purposes, saving them countless hours of paperwork. And learners get their own version of the Performance tab: a full assignment history showing their score, instructional focus, due date, completion status, and time spent for every assignment, filterable by course, instructional focus, or date range, with one-click access into the full results for any assignment and an exportable performance report.
- AI-guided, human-led: A good dashboard supports human decision-making. In the faculty’s Learner view, every instructional focus card includes an "assign more cases" link that opens the case library pre-filtered to that focus area, so faculty can spot a gap and assign more practice without leaving the screen. A criterion-level rubric breakdown provides an even more focused review. Administrators have access to longitudinal data for every course and learner across the institution enrolled in a DDx course, supporting their need for oversight and meeting accreditation requirements. As for learners, they see their own scores transparently, ensuring they feel safe and equipped to take ownership of their learning progress.
The questions you can now answer with this dashboard
From key findings, to final score, faculty, in one place, the Performance dashboard now lets you answer questions like these:
- Where specifically are learners struggling? Beyond a single average score, the dashboard breaks performance down by instructional focus: clinical reasoning, communication, routine care, clinical skills, clinical foundations each aligned to accredited competency frameworks. That's how you know exactly where a gap is, not just that one exists.
- Is the issue I’m seeing an isolated struggle or persistent pattern? A chart plots performance across every instructional focus over time, so you can tell immediately whether a low score is a one-off or the start of a trend, instead of piecing it together manually or from memory.
- If I’m seeing a persistent issue, what should I do about it? And what is the improvement result of that guidance? A gap in one learner calls for assigning additional cases specifically aligned to weaknesses, then watching the chart to see whether the next few cases move the needle. The same gap across a cohort calls for enriching lecture materials before it compounds. When a gap persists program-wide, a deliberate curriculum change can be made at the end of a term rather than a piecemeal fix.
See it in DDx
If you're already using DDx, the Performance dashboard with the Learner view is live now! The Administrator view will be released in October. Log in and the Performance tab is the first thing you'll see. If you're evaluating platforms for your program, request a short demo to see how other programs are using DDx to close performance gaps.
FAQ
What is the DDx Performance dashboard? The Performance dashboard is a new tab in DDx that shows a course carousel and a sortable, filterable learner table, with average scores and an instructional-focus breakdown for every learner in a course. At the learner level, performance results include a summary card and per-focus score card showing how many assignments each score reflects, so that trends are visible directly, rather than a project for faculty to reconstruct from individual case grades. The performance dashboard replaces manually piecing together data from individual case results.
Who can see learner performance data in DDx? Visibility depends on the role. Faculty see performance data for the courses they teach and the learners enrolled in them. Learners see their own performance data across every course they're enrolled in, regardless of which faculty member teaches it. Administrators can see analytics for every course and learner across the institution enrolled in a DDx course.
Can faculty assign more cases based on a performance gap? Yes. Every instructional focus card in the Learner view includes an "assign more cases" link that opens DDx's case library pre-filtered to that specific focus area, so faculty can move from spotting a weak area to assigning targeted practice without leaving the page.
What does this look like in practice: When an educator looks at the Performance dashboard, the tab opens with a course carousel that displays scrollable cards showing each course's overall average score and learner count at a glance. This sits above a full learner table for the faculty’s cohort. That table shows every learner's average score alongside a breakdown by instructional focus (such as: clinical reasoning, communication, routine care, clinical skills, clinical foundations), and it can be sorted by name, assignment count, average score, or any single instructional focus score, then filtered by course and date range. Faculty can jump from a course card straight into that course's assignment overview and come right back without losing their place.
From there, the Learner view goes deeper on a single learner: an average performance summary (average score, multiple-choice performance, assignments completed, average time to complete) so faculty get a birds eye view of a learner at a glance, an average score card for each instructional focus so faculty can quickly discern a learner’s strengths and areas for improvement, a bar chart per rubric that breaks down performance to the criterion level so that faculty can see specifically where learners need more support in an instructional focus, and, most notably, a chart plotting performance across all cases over time, so a trend is visible at a glance rather than reconstructed from memory.
