<aside> 🔒
Recreated for this portfolio. I built Tableau dashboards on student performance task results at Green Dot Public Schools, at the school's request. Student data is private, so all figures below are illustrative, built to show the design decision.
</aside>
The school wanted dashboards showing how students were doing on performance tasks, so teachers and leaders could see where to focus.
The easy build was one number: the class average against one benchmark. I didn't build it that way.
I tracked English learners, students in special education and general education students separately, each against their own benchmark. A blended average answers "is the class okay?" Segmented views answer the question leaders actually need: "which students need what?"
| View | Result | Benchmark | What it says |
|---|---|---|---|
| Blended class average | 70 | 70 | On track. Nothing to see. |
| General education | 78 | 75 | Slightly ahead |
| English learners | 61 | 70 | 9 points behind: needs support now |
| Special education | 55 | 50 | 5 points ahead: a win the average erased |
xychart-beta
title "Each group against its own benchmark (illustrative)"
x-axis ["General ed", "English learners", "Special ed", "Blended"]
y-axis "Score" 0 --> 100
bar [78, 61, 55, 70]
line [75, 70, 50, 70]
Bars are results; the line is each group's benchmark. The blended bar sits exactly on its line, while English learners fall well below theirs and special education students clear theirs. Each group is measured against its own growth benchmark, set from where that group started, not against a single class-wide target.
<aside> 💡
The blended average hid two stories at once: a group falling behind and a group succeeding. A dashboard is a decision about what people get to notice.
</aside>
A single snapshot says where students are. Leaders also needed to know whether what the school was implementing, the Next Generation Science Standards (NGSS), was moving each group. So I tracked every group across performance tasks through the year, each against its own benchmark.
| Group | Task 1 | Task 2 | Task 3 | Task 4 | Benchmark |
|---|---|---|---|---|---|
| General education | 72 | 75 | 78 | 80 | 75 |
| English learners | 55 | 58 | 61 | 66 | 70 |
| Special education | 46 | 50 | 55 | 57 | 50 |
The snapshot above is Task 3. Over time, the story changes again: English learners are still below benchmark, but closing the gap every task. That's a different conversation from "9 points behind."
xychart-beta
title "English learners across performance tasks (illustrative)"
x-axis ["Task 1", "Task 2", "Task 3", "Task 4"]
y-axis "Score" 40 --> 80
bar [55, 58, 61, 66]
line [70, 70, 70, 70]
Bars are English learners' results; the flat line is their benchmark.