Ranking-use guide
HK primary-school rankings and secondary pathways
Separate rankings, official secondary relationships and school-published destinations before using school net, fees, travel and child needs to build a verifiable shortlist.
Remove unsuitable schools before looking at positions
The Education Bureau does not publish primary-school banding or a territory-wide league table. Separate government and aided POA from DSS and private direct admission first, then use P1Tracker's Top 100 to find schools worth checking. This non-official research index is not an academic-result table, child-fit score or admission probability.
Remove schools that fail the family's constraints
Remove options that cannot work before reading positions. Part B of Central Allocation is tied to the home school net. DSS and private schools sit outside that net but have separate applications, fees and consequences of accepting a place. Gender, religion, main teaching language, one-way travel, affordable fees and care arrangements matter more than a small rank difference.
Check what the score actually measures
A total is comparable only within the same method. Check the data year, weights, missing-value rule and every component. Operating history is informational only and does not score; teacher and campus data describe structure but do not directly prove academic outcomes. The highest P1Tracker original research score, 83.1, is normalized to 60 points and every other listed school receives score divided by 83.1 and multiplied by 60; research rank and tier do not score separately, and an eligible unlisted school receives a fixed 15/60. A missing official pathway, teaching or campus field receives the applicable 50% category or subcomponent fallback, not a minimum score for every school. No component becomes official banding.
A secondary relationship is not an allocation outcome
The finder uses through-train, direct and linked relationships from the official 2025 Primary School Profiles. A relationship describes how schools are connected; it is not a recent destination result, allocation percentage or progression guarantee.
If a school publishes destinations, verify the publisher, school year, covered cohort, denominator and metric definition. Not published does not mean zero or an inferior school. Education Bureau allocation bands classify pupils during allocation; they are not permanent official labels for a primary or secondary school and cannot predict an individual child's result.
Sort the shortlist into three groups
Worth investigating
Start with eight to twelve schools from the Top 100, district pages or family knowledge without predicting admission.
The route and family constraints work
After checking net or direct admission, fee, travel, language, gender and dates, reduce the list to four to six.
Every school has a next step
Record one official check, one visit or session, one family question and one deadline action for each school.
Compare no more than four schools at once
Differences disappear when too many schools share one screen. Compare up to four using official facts, secondary relationships, fee sources, 2027/28 application status and real family limits. Complete one set before replacing it and retain the reason for excluding a school.
Return to the admission rules before ordering choices
POA choices must follow Education Bureau rules, the applicable net and genuine family preference. A territory-wide list cannot simply become Part B order. DSS and private schools require separate planning for openings, deadlines, interviews, results and place-acceptance terms. A practical fallback is still not a guaranteed offer.
Official verification points
Return to the Education Bureau's annual school-net lists for POA choices and the CHSC Primary School Profiles for school facts. Neither a ranking page nor a comparison tool replaces them.
FAQ
Frequently asked questions
Does Hong Kong have official primary-school banding?
No. The Education Bureau does not publish primary-school banding or a territory-wide ranking. Public lists must remain labelled non-official.
Is the number-one school best for every child?
No. A method measures selected data dimensions and cannot replace child needs, travel, fees, language, culture and family arrangements.
Can a territory-wide ranking be copied into POA Part B?
No. Part B is constrained by the home school net and Education Bureau rules, and order should reflect genuine family preference.
Does a through-train or feeder relationship guarantee progression?
No. A relationship label does not guarantee a place. Verify the current eligibility and arrangement with both schools.
What should I verify in a school-name plus secondary-allocation search?
Separate a through-train, direct or linked relationship in the official profile from school-published destinations for a stated year and from Education Bureau allocation. They are not interchangeable. If the school has not published a verifiable year, cohort and definition, keep the outcome unpublished rather than inferring a percentage or guarantee.