Student Data Analysis (SDAS): From Mark Sheets to Multi-Year Tracking — 8 Questions Principals and Panel Heads Should Ask
A mark sheet tells you who scored what; a Student Data Analysis System answers why, what the trend is and what to do next. Eight questions show how standard scores, percentiles, progress rates, multi-year tracking and teacher analysis are used — and how AI natural-language queries let every teacher ask.
Short answer: the value of SDAS is not seeing scores but using standard scores, percentiles and multi-year tracking to answer eight decision questions — who needs support, which strand is weakest, which practices work. AI natural-language queries let every teacher ask, instead of relying on one "data teacher".
From mark sheet to decision: what is missing?
Most schools have an Excel mark sheet after every exam, but it usually stops at averages and pass rates. Missing are cross-subject comparison (difficulty differs), multi-year tracking (the same student over time), strand analysis (where the weakness is) and the teacher level (which practices work). SDAS fills those four gaps.
Basics in two minutes
- Standard score (z-score) = (raw − grade mean) ÷ standard deviation. 0 is the grade average; +1 is one SD above.
- Percentile: the share of the grade a student scored above.
- Progress rate: change in standard score between exams (not raw marks).
- Strand: each paper split by curriculum strand (e.g. "reading comprehension", "number and algebra").
Eight questions for principals and panel heads
| # | Question | How SDAS answers |
|---|---|---|
| 1 | Which students lag across subjects persistently? | List with z < −1 in multiple subjects, two or more times in a row |
| 2 | What is the grade's weakest strand? | Strand mean standard-score ranking |
| 3 | Which class improved most between terms? | Progress rate by class (change in z) |
| 4 | Trend for the same cohort from P4 to P6? | Multi-year line chart by student ID |
| 5 | Is the pass-rate change due to difficulty or real progress? | Compare standard vs raw scores |
| 6 | Whose classes improved markedly in which strand? | Teacher × strand progress (for professional sharing) |
| 7 | Are middle students (40th–60th percentile) overlooked? | Progress by percentile band |
| 8 | Did the remediation programme work? | Tagged list, before/after comparison |
Three steps from Excel to SDAS
- Import: CSV / Excel results mapped to student ID, class, subject and strand.
- Compute: standard scores, percentiles, progress rates; absences and exemptions handled by rule.
- Query: default dashboards plus natural-language questions (e.g. "Which P5 students dropped more than 0.5 SD in Maths between terms?").
AI natural-language queries: let every teacher ask
Data analysis used to sit with one or two Excel-savvy teachers. Natural-language queries drop the barrier to zero — teachers ask in Chinese or English, the system builds the query, shows a table and chart, and explains. Panel heads can ask live in a panel meeting and see results instantly.
Three principles for using data
- Trend before single points: one exam's fluctuation means little.
- Data starts the conversation, it is not the conclusion: find the class that improved and ask "what did you do?"
- Scoped access: principals see the school, panel heads their subject, class teachers their class — a privacy requirement that also keeps conversations focused.
Link to the grant report
SDAS multi-year tracking is exactly the before/after evidence the grant's "student learning outcomes" requires: standard-score change before and after AI use, for participating versus comparison classes, exportable as report charts.
Takeaway
A mark sheet tells you what happened; SDAS tells you the trend, the reason and the next step. When every teacher can question the data in their own words, school data finally becomes school decisions.
FAQ
A Student Data Analysis System imports exam results, automatically computes standard scores and percentiles, analyses pass and progress rates by grade, class, subject and teacher, and tracks students within a year and across years.
Different subjects and papers differ in difficulty, so raw marks cannot be compared directly. Standard scores (z-scores) convert each result into a position relative to the whole grade, making comparisons across subjects and years meaningful.
It depends on how the school uses it. The healthy approach treats it as the start of professional dialogue — find which classes improved markedly in which strands and invite those teachers to share — rather than ranking or penalising.
Related articles
How AI Reduces Teacher Admin Workload: Notices, Minutes, Parent Letters and eAdmin in Practice
Teachers can spend 8–10 hours a week on paperwork. Six examples show how AI admin agents (eAdmin) draft eNotice, eCircular, parent letters, meeting minutes, event proposals and iPortfolio comments — and why grounding in the school knowledge base and matching the school's voice are the keys.
Read moreThe PIE Cycle Meets AI Dashboards: Evidence-Based School Decisions
Every school runs Plan–Implement–Evaluate, but "evaluate" often relies on impressions. See how PIE KPIs link directly to platform data — AI usage, objective mastery, teacher training records, SDAS results — so planning has a baseline, implementation has tracking, evaluation has evidence, and grant interim and final reports come out automatically.
Read moreThe School Knowledge Base (RAG): Letting AI Cite Your Own Materials and Policies
Generic AI does not know your school rules, pacing or notice format. A school knowledge base uses RAG (retrieval-augmented generation) so the AI searches school documents before answering and cites its sources. This article covers how it works, what to upload, permissions and export, and six use cases for teaching and administration.
Read more