eduKate Singapore

VIEW THIS AS

Auto mode follows the Route Engine until you choose a viewpoint.

YOU ARE HERE

ROUTE CHECK

CONNECTED TO

WHAT NEXT

Use the canonical route for this room, or HELP if you are unsure.

Next useful routeCanonical hubHELP

Secondary 2 G3 Mathematics Tutor Bukit Timah | Probability Tree Diagrams and Two-Stage Events

—

by

eduKate Asia
Three learners review open books together at a classroom table, with stacks of textbooks, stationery and a whiteboard in the bright room.

Did you know? The probability of getting a red counter and then a blue one can be different from the probability of blue and then red, depending on what is in the bag and whether the first counter is replaced. A two-stage tree diagram makes those changing chances visible rather than leaving the student to guess.

This Secondary 2 G3 Mathematics guide is written for Bukit Timah families who want the important rule immediately, followed by careful worked examples and progressive practice. Two-stage events can extend beyond a school’s current Secondary 2 test scope, so start with the school’s taught content and treat later probability branches as purposeful extension.

Find the part of this guide you need

Quick answer: multiply along a path and add different routes

For a two-stage probability tree, write the possible outcomes of stage one, then the conditional stage-two outcomes after each branch. Multiply probabilities along a complete path to calculate that joint outcome. Add the probabilities of mutually exclusive complete paths when an event can happen through different routes. Each set of branches leaving a node must sum to one. If you draw without replacement, update the counters left in the bag.

For two fair independent coin tosses, the paths HH, HT, TH and TT each have probability (1/2)(1/2) = 1/4. Exactly one head happens through HT or TH, so the probability is 1/4 + 1/4 = 1/2. The answer is not 1/4 simply because the words ‘one head’ sound singular.

Why a tree is more than a diagram of possibilities

A probability tree keeps two things apart: which outcomes can occur and how likely each outcome is under the model. Each first-stage branch connects to its own second-stage branches. For example, drawing a red counter from a bag without replacement changes the number of red counters available for the next draw. The second-stage fractions must reflect the new contents.

When the trials are independent and conditions do not change, the second-stage probabilities are the same after every first-stage outcome. When an outcome changes the population, the fractions can differ along different branches. A student who copies the original denominator everywhere may not understand the procedure they are modelling.

A reliable six-step route through a school question

  • Define one trial and the event the question asks about. Clarify whether order matters.
  • List stage-one possibilities and assign probabilities according to the stated model.
  • For every stage-one outcome, describe the stage-two situation and record its probabilities.
  • Check that outgoing branches from the same point add to one; this catches missing outcomes.
  • Multiply the fractions or decimals along each relevant complete path.
  • If more than one distinct path satisfies the event, add those path probabilities and explain the result in context.

Independent events compared with dependent events

With replacement, drawing a coloured counter, replacing it and mixing the bag restores the original contents. Under a uniform selection model, the second draw’s colour chances are the same as the first. Without replacement, the composition changes, so the second probability is conditional on what happened on the first draw.

Independence is not a statement that the two results must be different. Two independent fair coin tosses can both be heads. Conversely, without replacement the draws from a finite bag are not normally independent. The relevant question is whether knowing stage one changes the chance assigned at stage two.

Exactly one, at least one and neither

Exactly one means one occurrence and one non-occurrence in the stated number of trials. For two stages, either success then failure or failure then success may qualify. At least one also includes two successes. Sometimes finding the complement, no successes, is easier than adding several qualifying paths.

Translate those English phrases into paths before doing arithmetic. An incorrect set of paths can give a neat fraction that answers the wrong question. In a strong answer, the pupil should name the event in plain language next to the algebraic expression.

Why your probabilities have to total one

All complete and mutually exclusive outcomes of a properly specified experiment have total probability one. On a simple two-stage tree with two outcomes at each stage, there are four full paths; their probabilities should sum to one. If they do not, something has been omitted, counted twice or assigned an incorrect conditional probability.

The total-one rule is a diagnostic, not a substitute for correctly constructing the tree. Two incorrect numbers can sometimes still add to one. After checking the sum, inspect whether each branch’s probability genuinely follows from the story.

Drawing the tree neatly without creating a false model

There is no requirement for the branches to be drawn to scale. Labels and fractions matter more than the angle of a branch. Put probabilities next to the correct branch, not at the end where they can look like a combined result. The probability at the leaf for a complete path should be the product of that path’s stages.

If a question gives an experimental probability rather than an assumed fair device, use the stated estimate and its limitations. Do not silently replace it with 1/2 or 1/6. The tree is a tool for organising the mathematical assumptions you are given, not for proving those assumptions true.

When to use a simple table instead

Two fair coin tosses can be described with a complete four-entry table or with a probability tree. The table makes ordered outcomes easy to see; the tree is especially useful when second-stage chances depend on the first outcome. Both representations should agree when modelling the same experiment.

Students can compare a table and a tree as a self-check. If the sample space has four equally likely outcomes but the tree has only three leaves, an ordered outcome has been lost. This is one reason the problem should be translated into possible outcomes before multiplying any fractions.

A school-level note on assumptions and real chance

The worked examples below are mathematical illustrations. A fair coin or uniformly selected counter is a model assumption. Real equipment or physical procedures can behave differently. The purpose is to master conditional structure, not to claim that a classroom coin, bag or weather forecast has been scientifically tested.

Under Full Subject-Based Banding, G3 is a subject level, and schools can sequence probability over different terms. If your current assessment covers only single events, use the early branches as a bridge and save the more complex examples for later.

Thirty original worked probability clinics

Two fair coins

Question. Two independent fair coins are tossed once. What is P(HH)?

Solution. Each head has probability 1/2. Following H then H multiplies 1/2 × 1/2 = 1/4.

Reasoning. Independence means the second chance does not change after the first toss. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. The result is not 1/2 simply because heads appears twice.

Independent extension. For HT, multiply 1/2 × 1/2 again: probability 1/4. Have the learner explain whether the second stage changes after stage one before choosing a method.

Exactly one head

Question. What is the probability of exactly one head in two fair independent coin tosses?

Solution. The qualifying ordered paths are HT and TH, each 1/4. Add them to obtain 1/2.

Reasoning. A tree separates the two routes that satisfy the same verbal condition. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Counting only HT leaves out an equally valid ordered outcome.

Independent extension. At least one head also includes HH, so it has probability 3/4. Have the learner explain whether the second stage changes after stage one before choosing a method.

No heads in three tosses

Question. Three independent fair coins are tossed. Find P(no heads).

Solution. The single qualifying path is TTT, probability (1/2)³ = 1/8.

Reasoning. The event is a single complete sequence of three tails. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. At least one tail is a different event and has more qualifying paths.

Independent extension. The complement P(at least one head) is 1 − 1/8 = 7/8. Have the learner explain whether the second stage changes after stage one before choosing a method.

Exactly two heads in three tosses

Question. Three independent fair tosses are made. Find P(exactly two heads).

Solution. Valid sequences HHT, HTH and THH each have probability 1/8. Sum three paths: 3/8.

Reasoning. The arrangement order matters for which leaves satisfy the event. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Do not treat two heads as one leaf when there are three positions.

Independent extension. P(exactly three heads) is only 1/8. Have the learner explain whether the second stage changes after stage one before choosing a method.

A red bag with replacement

Question. A bag has 3 red and 2 blue counters. Draw, replace, mix, draw. Find P(RR).

Solution. Each draw has P(red) = 3/5, so RR is (3/5)(3/5) = 9/25.

Reasoning. Replacement restores the starting contents before stage two. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. A denominator of 4 in the second fraction would falsely model no replacement.

Independent extension. P(RB) with replacement is (3/5)(2/5) = 6/25. Have the learner explain whether the second stage changes after stage one before choosing a method.

Two reds without replacement

Question. Use the same bag with three red and two blue counters, but do not replace. Find P(RR).

Solution. First red chance 3/5. Two red remain among four, so second red chance is 2/4. Multiply to get 3/10.

Reasoning. Stage two is conditional on a red having already been removed. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. The 9/25 replacement answer is wrong for this experiment.

Independent extension. Draw red then blue instead: (3/5)(2/4) = 3/10. Have the learner explain whether the second stage changes after stage one before choosing a method.

Red then blue

Question. Three red and two blue counters, without replacement. Find P(RB).

Solution. First red probability 3/5; after red, two blue remain among four. P(RB) = (3/5)(2/4) = 3/10.

Reasoning. After red, there is one fewer red but unchanged blue count. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Dividing by five again ignores the removed counter.

Independent extension. Blue then red has probability (2/5)(3/4) = 3/10. Have the learner explain whether the second stage changes after stage one before choosing a method.

Exactly one red

Question. Three red and two blue counters, two draws without replacement. Find exactly one red.

Solution. Two qualifying paths are RB and BR, each 3/10. Their probabilities add to 3/5.

Reasoning. Exactly one red includes both possible orders, but not RR. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. The event is not just RB because the first draw is not specified.

Independent extension. Check complementary possibilities: RR probability 3/10 and BB 1/10; 1 − 4/10 = 3/5. Have the learner explain whether the second stage changes after stage one before choosing a method.

At least one red

Question. Three red and two blue counters; draw twice without replacement. Find at least one red.

Solution. The only excluded event is BB. P(BB) = (2/5)(1/4) = 1/10. Complement gives 9/10.

Reasoning. A complement avoids summing RR, RB and BR separately. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Exactly one red has probability 3/5 and misses RR.

Independent extension. Verify direct sum: 3/10 + 3/10 + 3/10 = 9/10. Have the learner explain whether the second stage changes after stage one before choosing a method.

Two blues

Question. Three red and two blue counters; two draws without replacement. Find P(BB).

Solution. P(BB) = (2/5)(1/4) = 2/20 = 1/10.

Reasoning. Only one blue counter remains after the first blue draw. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. A tree branch should show the changed numerator and denominator.

Independent extension. With replacement P(BB) would instead be (2/5)² = 4/25. Have the learner explain whether the second stage changes after stage one before choosing a method.

Even on both dice

Question. Two fair six-sided dice are rolled independently. Find probability both show even numbers.

Solution. Each die has three even faces of six, probability 1/2. Multiply 1/2 × 1/2 = 1/4.

Reasoning. The two independent die results have their own even/odd branches. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Do not divide by 6 once for an event involving two dice.

Independent extension. P(odd then even) is also 1/4 under the same model. Have the learner explain whether the second stage changes after stage one before choosing a method.

Sum of seven on two dice

Question. Two fair dice are rolled. Find P(sum = 7).

Solution. There are 36 ordered pairs; favourable pairs are (1,6), (2,5), (3,4), (4,3), (5,2), (6,1). Answer 6/36 = 1/6.

Reasoning. The first die and second die form distinguishable stages. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Do not count (2,5) and (5,2) as the same ordered outcome.

Independent extension. The ordered-pair table and a six-branch-first-stage tree agree. Have the learner explain whether the second stage changes after stage one before choosing a method.

Two fair spinners

Question. One spinner has four equal sectors A-D; another has three equal sectors 1-3. What is P(A then 2)?

Solution. P(A) = 1/4 and P(2) = 1/3, giving P(A2) = 1/12 for independent spins.

Reasoning. Different stage spaces may have different denominator sizes. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Writing 1/4² would incorrectly pretend the second spinner has four sectors.

Independent extension. There are twelve equally likely ordered sector pairs altogether. Have the learner explain whether the second stage changes after stage one before choosing a method.

An event described by two branches

Question. One spinner gives red with 1/3 probability; a second independent spinner gives red with 1/2. Find exactly one red.

Solution. Red then not-red gives (1/3)(1/2) = 1/6. Not-red then red gives (2/3)(1/2) = 1/3. Sum = 1/2.

Reasoning. The two colour probabilities can differ between stages, so the two path chances differ. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Assuming both paths are 1/6 would ignore the 2/3 chance of not-red on the first spinner.

Independent extension. Both red has probability 1/6, and neither red 1/3; the four leaves total one. Have the learner explain whether the second stage changes after stage one before choosing a method.

An intentionally biased coin model

Question. A coin is modelled with P(H) = 0.6 and P(T) = 0.4 on independent tosses. Find P(HT).

Solution. Multiply 0.6 × 0.4 = 0.24.

Reasoning. The probability tree uses the stated biased model rather than silently substituting 1/2. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. A theoretical probability should not be changed because the coin is called a coin.

Independent extension. Exactly one head over two tosses has probability 0.24 + 0.24 = 0.48. Have the learner explain whether the second stage changes after stage one before choosing a method.

At least one head under a biased model

Question. For the independent coin model P(H)=0.6, find probability of at least one head in two tosses.

Solution. No heads means TT, probability 0.4² = 0.16. Therefore at least one head is 1 − 0.16 = 0.84.

Reasoning. The complement is only one path, so it is quicker than adding three leaves. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Using 0.6² = 0.36 finds two heads, not at least one.

Independent extension. Check direct: HH 0.36 plus HT 0.24 plus TH 0.24 = 0.84. Have the learner explain whether the second stage changes after stage one before choosing a method.

One green after an orange draw

Question. A bag has 4 green counters and 1 orange. Without replacement, first draw orange. What is P(green second | orange first)?

Solution. After removing the only orange, four green remain among four counters. Conditional probability = 4/4 = 1.

Reasoning. A branch probability can equal one if that outcome becomes certain. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Do not reuse the first-stage green chance 4/5.

Independent extension. The path P(orange then green) is (1/5)(1) = 1/5. Have the learner explain whether the second stage changes after stage one before choosing a method.

Two green from a five-counter bag

Question. With 4 green and 1 orange, draw twice without replacement. Find P(GG).

Solution. P(GG) = (4/5)(3/4) = 3/5.

Reasoning. Second green numerator becomes three after a green leaves the bag. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Multiplying 4/5 by 4/5 assumes replacement.

Independent extension. At least one green is certain because only one orange exists and two counters are taken. Have the learner explain whether the second stage changes after stage one before choosing a method.

Exactly one orange

Question. A bag contains 4 green and 1 orange. Draw two without replacement. Find exactly one orange.

Solution. Orange then green is (1/5)(4/4)=1/5. Green then orange is (4/5)(1/4)=1/5. Add to get 2/5.

Reasoning. The two distinct paths are mutually exclusive and both count. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Two oranges cannot occur without replacement when only one orange existed.

Independent extension. Complement: no orange is GG probability 3/5, so exactly one orange is 2/5. Have the learner explain whether the second stage changes after stage one before choosing a method.

Three draws without replacement

Question. A bag has 2 red and 3 blue. Find P(red, then blue, then red) across three draws without replacement.

Solution. The path is (2/5)(3/4)(1/3) = 1/10.

Reasoning. The conditional denominator shrinks from 5 to 4 to 3 while colour counts change. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Using 2/5 in the final branch ignores what was removed.

Independent extension. Red-red-blue has (2/5)(1/4)(3/3) = 1/10 under the same bag. Have the learner explain whether the second stage changes after stage one before choosing a method.

A branch complement

Question. A particular two-stage event has probability 0.37. What is probability it does not occur?

Solution. P(not event) = 1 − 0.37 = 0.63, provided ‘not’ means the complement within the same complete experiment.

Reasoning. Every elementary outcome either satisfies the event or its complement. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Subtracting from 100 without converting between percent and decimal is a unit mistake.

Independent extension. If the event probability is 37%, the complement is 63%. Have the learner explain whether the second stage changes after stage one before choosing a method.

An expected count from a two-stage chance

Question. A two-stage event has probability 1/4 under the model. Across 80 independent repeats, how many occurrences are expected?

Solution. Expected count = 80 × 1/4 = 20.

Reasoning. Expectation describes the average model count over repeated trials, not a guaranteed observed number. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Do not assert exactly 20 outcomes must appear in every run.

Independent extension. If 24 occurrences are observed, report 24/80 = 0.30 as sample relative frequency. Have the learner explain whether the second stage changes after stage one before choosing a method.

Weather and transport, hypothetical model

Question. Suppose P(rain)=0.3; P(delay|rain)=0.6 and P(delay|dry)=0.1. Find P(delay).

Solution. Delay through rain is 0.3 × 0.6 = 0.18. Delay through dry is 0.7 × 0.1 = 0.07. Add = 0.25.

Reasoning. Both distinct weather branches can lead to delay, so both path probabilities belong in the total. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. This is a fictional model, not a current weather or transport forecast.

Independent extension. No-delay probability is 0.75, obtainable by the complement. Have the learner explain whether the second stage changes after stage one before choosing a method.

Conditional inference from a joint path

Question. In the same hypothetical model, what fraction of the delayed journeys occurred in rain under the model?

Solution. P(rain and delay)=0.18; P(delay)=0.25. So P(rain|delay)=0.18/0.25=0.72.

Reasoning. Conditional probability changes the denominator to the cases in which delay actually happened. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. The result is not simply the original rain chance 0.30.

Independent extension. Say in words: among journeys with delay, the model allocates 72% to rain conditions. Have the learner explain whether the second stage changes after stage one before choosing a method.

Check outgoing branches add to one

Question. After a red counter is removed from a 3-red/2-blue bag, what are the second-stage fractions?

Solution. Remaining 2 red and 2 blue among four. P(red next|red first)=2/4; P(blue next|red first)=2/4; sum one.

Reasoning. Every node must account for all mutually exclusive possible outcomes from that state. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. A pair of fractions 2/5 and 2/5 sums to 4/5 and reveals a missing count.

Independent extension. After a blue first, the second fractions instead are 3/4 red and 1/4 blue. Have the learner explain whether the second stage changes after stage one before choosing a method.

An order-sensitive prize rule

Question. Two independent fair coins are flipped. A prize requires heads on the first and tails on the second.

Solution. Only HT qualifies, giving 1/4.

Reasoning. The condition specifies order, so TH is not another qualifying path. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Confusing ‘one of each’ with ‘first heads then tails’ doubles the chance incorrectly.

Independent extension. If the prize instead requires one head of either order, the probability is 1/2. Have the learner explain whether the second stage changes after stage one before choosing a method.

The chance of two identical coin results

Question. Two fair independent coins are flipped. Find probability both outcomes match.

Solution. Matching results are HH and TT. Each is 1/4, so total P(match)=1/2.

Reasoning. Different complete paths can represent one verbal event. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. Writing only HH overlooks the tails-tails case.

Independent extension. The complementary probability of different results is also 1/2. Have the learner explain whether the second stage changes after stage one before choosing a method.

Independent results are not forced to alternate

Question. A fair coin has shown four tails in succession. What is P(heads on the next toss)?

Solution. Assuming independent fair tosses, the next heads chance is still 1/2.

Reasoning. Previous outcomes do not alter the model’s probability for the next independent trial. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. The claim ‘heads is now due’ is the gambler’s fallacy.

Independent extension. If the fifth toss is heads, that does not retroactively change any earlier chance. Have the learner explain whether the second stage changes after stage one before choosing a method.

A probability tree with unequal branch counts

Question. Stage one has two outcomes with probabilities 1/4 and 3/4. After the first, event X has chance 1/2; after the second, X has chance 2/3.

Solution. Total P(X) = (1/4)(1/2) + (3/4)(2/3) = 1/8 + 1/2 = 5/8.

Reasoning. One event can be reached by more than one stage-one branch with different conditional probabilities. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. A simple average of 1/2 and 2/3 ignores the differing weights of the stage-one branches.

Independent extension. P(not X) = 3/8; compute the two complementary leaves to confirm. Have the learner explain whether the second stage changes after stage one before choosing a method.

Three draws: exactly two reds

Question. With 2 red and 1 blue counter, draw all three without replacement. What is probability of exactly two reds?

Solution. It is certain: all three counters are drawn and exactly two of them are red. Probability is 1.

Reasoning. Sometimes the event is logically guaranteed without multiplying the paths. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. A zero probability would confuse the order of reds with their total count.

Independent extension. Different orders RRB, RBR and BRR are possible, but all satisfy the event. Have the learner explain whether the second stage changes after stage one before choosing a method.

A total probability audit

Question. For a red-blue two-draw bag model without replacement, the complete path probabilities are RR 3/10, RB 3/10, BR 3/10 and BB 1/10.

Solution. Their sum is 3/10 + 3/10 + 3/10 + 1/10 = 1, so the tree accounts for every ordered two-draw outcome.

Reasoning. Total probability one is a useful last-line check on a complete tree. Always name the ordered path before performing its probability calculation, so the multiplication corresponds to a real branch on the tree.

Check the likely misconception. It would be wrong to add every event count to 1 if those events overlap rather than partition the outcomes.

Independent extension. Circle the paths satisfying exactly one red; only RB and BR should be included. Have the learner explain whether the second stage changes after stage one before choosing a method.

How to draw one clean probability tree by hand

Put the first-stage choices at the left and the second-stage choices to their right. For a bag with three red and two blue counters, label the first branches R with probability 3/5 and B with probability 2/5. Under the red branch, write the state ‘2 red, 2 blue remain’, then draw red and blue branches each with probability 2/4. Under the blue branch, write ‘3 red, 1 blue remain’, then branches with 3/4 and 1/4. The state labels prevent the common mistake of copying the first fractions into every second node.

At the end of each complete branch, multiply the path probabilities and write the ordered outcome: RR is 3/10, RB is 3/10, BR is 3/10 and BB is 1/10. Summing all four gives one. If the question asks for at least one red, shade RR, RB and BR and add. If it asks for exactly one red, shade only RB and BR. One accurate tree can answer several related questions, but the event definition determines which leaves count.

A fifteen-minute teaching activity for a group of three

Give each learner the same two-draw bag description but a different question: one calculates both reds, another exactly one red, and another at least one red. Ask them to draw their own trees before discussing solutions. The underlying tree should match, but the selected leaves should differ. This makes the class discussion about interpretation rather than copying the fastest learner’s answer.

After comparing, change the experiment from no replacement to replacement. Each student must redraw the second-stage fractions and calculate again. For exactly one red, the no-replacement probability is 3/5; with replacement it becomes 2 × (3/5)(2/5) = 12/25. The comparison reveals how a small wording change alters the model, even though the original counter counts are unchanged.

A two-week learning plan for accurate conditional reasoning

In the first week, begin with coins and an outcome table, followed by a two-colour bag with replacement. Check whether the student can state why the second-stage probabilities stay the same. Then remove replacement, keeping the starting bag unchanged, and ask the pupil to describe the new contents after each first-stage result. A sketch of the remaining counters often helps more than another equation.

In the second week, mix exact-one and at-least-one wording, conditional weather-like fictional models and a three-draw extension. Have the student predict whether addition, multiplication or a complement will be needed before starting the numerical work. Finish with a question that asks the learner to reject a flawed tree in which outgoing probabilities fail to sum to one. Correct diagnosis is part of mastery.

Five parent questions about tree diagrams

Does the tree have to look like a professional diagram? No. It should clearly label stages, branches and probabilities and show all possible outcomes. Straight, tidy branches and a consistent reading direction are helpful, but visual symmetry is not a mathematical requirement. The fractions must describe the stated model.

Why do we multiply along a path? Because a complete outcome requires the stages on that path to happen together. The second probability is conditional on the first stage; their product gives the chance of that joint path. For independent events, the second stage probability happens to be unchanged, but the path product principle still applies.

Why do we add different paths? If the desired event can occur through separate mutually exclusive complete outcomes, any of those routes qualifies. For ‘exactly one red’, RB and BR are distinct and cannot both occur on the same two-draw experiment. Their probabilities can therefore be summed.

Do two reds always mean red then red? In a two-draw experiment, yes, RR is the only path for two reds. In a three-draw experiment, exactly two reds can appear in several positions: RRB, RBR or BRR. The number of stages and the event wording determine the sample space.

Can a pupil simply memorise ‘multiply and add’? That is not enough. They must identify the sample space, distinguish replacement from non-replacement, and choose exactly which paths satisfy the verbal condition. A tutor should remove the topic heading and require an explanation of why the branches have those probabilities.

A quick diagnostic for three different learning needs

Repair: If a pupil cannot simplify (3/5)(2/4), pause for fraction multiplication, numerator changes and denominator meaning. Use counters or coloured squares to make each fraction a count out of the remaining sample space. Algebraic speed should not hide a misunderstanding of the experiment.

Stabilise: If the pupil can draw a correct tree but chooses the wrong leaves, practise converting ‘exactly’, ‘at least’ and ‘not’ into a list of ordered outcomes. Ask the learner to circle leaves before doing any arithmetic, then explain why the unselected leaves are excluded.

Extend: If the pupil is confident, present a three-stage bag or a fictional model in which an outcome changes the second-stage probability. Have the learner calculate one conditional chance and check the complete tree totals one. This extension is useful only after the core two-stage reasoning is secure.

Programme reference, curriculum and related Mathematics

G3 refers to a subject level under Full Subject-Based Banding. A Secondary 2 student should follow the chapters and weighted-assessment scope set by their school. The MOE G2/G3 Mathematics syllabus and 2027 SEC G3 syllabus index provide progression context, not a guarantee that two-stage trees are examined in every Secondary 2 class.

The unchanged eduKateSG small-group Mathematics reference describes three-pupil weekly tutorials lasting about 1.5 hours, with carefully guided corrections. The published teaching address is 8 Fourth Avenue, Singapore 268674, near Sixth Avenue MRT. The Bukit Timah title identifies families who may travel there; it does not claim a separate branch or any currently available place.

For helpful prerequisites read single-event probability and sample spaces, sets and Venn diagrams, and experimental probability. These are complementary guides, not replacement descriptions of a separate tutorial room.

Parents can contact eduKate Singapore or enquire about Secondary 2 G3 small-group tutorials by WhatsApp. Bring a marked question and the next assessed chapters to make the discussion useful. Confirm lesson times and placement directly.

Editorial review: 26 October 2026. These are original illustrative probability models, not real weather forecasts, gambling advice or copied examination questions. Conditions such as fairness, independence and replacement must be stated before conclusions are drawn.