Brandon A. BakiPortfolio

Demonstration site. Every student, teacher, class and result in these samples is fictional, created to show how the tools work.

Back

Case study · Lion Tracks

Three assessment platforms, one picture of every student

MAP Growth, CAT4 and DIBELS each arrive as their own export, in their own format, after their own testing window. Lion Tracks puts them together for school leaders, division heads and teachers – with the method in plain sight. It is in development, shaped with the assessment coordinator and school leaders.

  • 3assessments: MAP Growth, CAT4 reasoning, DIBELS early reading
  • 3 viewswhole school, a division, a teacher’s class
  • 11pages, from the briefing to the method
  • K–12three divisions; DIBELS for K–6

In development. The draft runs on a made-up school; nothing here is a real student’s result.

The problem

After every testing window the same spreadsheets were rebuilt by hand. The three platforms don’t share a format or even the same student identifiers; leaders received percentiles without context, and teachers received exports they had no time to read. Nobody could see a student’s reasoning ability, attainment and early reading side by side.

What it does

  • A briefing first. A plain-language headline – what changed, why and where to act – then four figures, each with its comparison and its number of students.
  • Growth by grade and class, UAE benchmarks, and Grades vs MAP to see where teachers’ grades and test results disagree.
  • CAT4 reasoning profiles next to attainment: the able students not yet attaining, and those whose language may be hiding their ability.
  • DIBELS benchmark levels for K–6, with who needs support, at what tier, and the actions agreed for them.
  • Viewing as decides the scope: leadership sees the school, a division head their division, a teacher their own class.
Lion Tracks briefing: a headline about attainment and growth, ratings, four figures with comparisons
The briefing, viewed as leadership, in the draft’s made-up school.

The data work

Bring in the files shows the roster and the three exports as they really arrive – leading zeros, “Last, First” names, duplicates, retests, UK year groups – and reconciles them in five steps, counting every rule it uses; the few rows it can’t be sure of are left for a person, with suggested matches, rather than guessed. The same join is written out in SQL. Data & method defines every measure: what it is, how it is calculated, which file it comes from, when it refreshes and who sees it.

Every figure on the briefing can show the evidence behind it: the number, the group, the comparison and how it was worked out.

Working with people

It started as Assessment Insights, for MAP Growth alone; CAT4 and DIBELS joined in October, when it became Lion Tracks, and DIBELS covers K–6 on the assessment coordinator’s advice. Retest-flagged results are left out by default, and the page says so.

What I’d do next

Try it with division leaders on the Winter MAP results and the middle-of-year DIBELS window (January 2027), and take the files straight from the platforms’ exports instead of uploads.

More case studies