How to Use P6 Chinese Frequency Lists (and How Not To)

We have published seven articles of word lists, covering the multiple-choice question types in the language-use section of P6 Chinese Paper 2 that can be counted word by word. Each list was counted from the same 266 prelim papers set by Singapore's top primary schools between 2021 and 2025. (The 'which sentence is correct' questions cannot be reduced to a word list, so there is none for them.) This article is the one to read first if you have not read any of the others, and the one to read last if you have read them all, because it is about what to do with the numbers.

About the data, once more: 266 different P6 prelim papers, 2021–2025, set in-house by top primary schools. They are not SEAB's PSLE papers. A word that appears often in prelims is a word that experienced setters consider worth testing — which is useful — but nobody can promise it will appear in any particular exam.

The seven lists at a glance

The number of different words is the most important column in that table, and the one people skip. A pool of 233 words for pinyin means the whole question type can be covered: a child who knows the pinyin of 233 words has, in a real sense, finished. A pool of 2,450 for vocabulary-in-context means the opposite. No list of 30 or 80 words covers it, and the frequency list is a starting point for a habit, not a syllabus.

The mistake most families make

The mistake is to print the list and memorise it top to bottom. It feels like the obvious thing to do with a ranked list, and it fails for two reasons.

The first is that the lists are ranked by how often setters used the word, not by how likely your child is to get it wrong. A P6 child who has done their homework already knows 努力 and 功课. Time spent memorising words they know is time not spent on the ones they do not.

The second is that most of the question types do not test the word itself. Connectives test whether the pair matches. Sentence completion tests logic and tone. Character choice tests the difference between two characters, not the word they belong to. Learning the word in isolation trains the wrong thing.

What to do instead

Step 1 — find the weak question type, not the weak word

Take one full prelim paper or a mock, and sort the mistakes in Paper 2 by question type. Most children lose their marks in one or two types, and it is rarely the type the parent expected. That type is where the corresponding list earns its keep.

Step 2 — test before you teach

Go down the list for the weak type and have your child answer, not read. For pinyin, say the word and have them write the pinyin. For character choice, show the two candidate characters and ask which one. Mark the misses. You will usually find that a list of 30 collapses to 8 or 10 words that actually need work.

Step 3 — work the difference, not the word

For each miss, practise the thing the question tests. A connective goes with its partner. A character goes next to the one it is confused with. A sentence half goes with the logic that connects it. Each of the seven articles ends with a section on how to do this for that type.

Step 4 — come back on day 3, day 7 and day 21

A word corrected once is a word forgotten within the week. The spacing does not need to be exact, and it does need to happen: the first return, a few days after the miss, is the one that matters most, and it takes five minutes.

A 15-minute routine

1. Five minutes: the day's returns. Re-test the words missed 3, 7 and 21 days ago. Anything missed again goes back to the start.
2. Eight minutes: new material from the weak type's list, tested first, then worked as pairs or differences.
3. Two minutes: your child tells you, in their own words, why one of today's wrong options was wrong. If they cannot, that item is not done.

That routine covers a list of 30 words in about two weeks, with returns built in, and leaves the child with something a memorised list never gives them: a way of thinking about the question type that works on words that were never on any list.