How to Tell When a Missing Value in a PSLE Science Table Can Be Inferred—and When It Must Stay Unknown

Wait, what? A blank box in a Science table is not an invitation to fill it.

Sometimes a missing value can be worked out exactly. Sometimes it can only be estimated. Sometimes the scientifically correct answer is that the value is not determined by the information given. Strong PSLE Science reasoning means knowing which of those three situations you are in before you touch the numbers.

Quick Answer

A missing value can be inferred only when the information in the question fixes that value. Look for an explicit total, rule, relationship, paired value, conservation statement, calculation rule or other supplied constraint that makes only one answer possible. If you merely see a pattern between neighbouring results, do not pretend an unmeasured value is known unless the question gives a rule that justifies that step. If several values could fit, the missing value must stay unknown.

Owned PSLE Science Learning Job

This guide owns one learner job: how a Primary 5/6 learner decides whether a blank, missing or unrecorded table value can be scientifically deduced from the given information, or whether it must remain unknown.

This is not a Mathematics owner, not a generic graphing page and not a science-concept page. Arithmetic may appear, but the dominant job is scientific evidence discipline: preserve what was measured, what was calculated, what was inferred and what was never established.

Why This Matters in PSLE Science

The 2026 PSLE Science examination is based on the 2023 Primary Science syllabus. Its official assessment objectives include applying scientific facts, concepts and principles; interpreting and analysing information; evaluating observations, information and methods; and communicating explanations and reasoning. Tables are therefore not just places to read numbers. They carry scientific objects, quantities, conditions, measurements and evidence limits.

The official documents do not create a universal rule saying every blank table cell must be filled. The learner has to read what the question actually establishes.

The Central Distinction: Deduced, Estimated or Unknown

Status	What it means	What the learner may do
Deduced	The given information determines one value.	Calculate or infer it and show the scientific basis.
Estimated	The question explicitly permits or requires an approximation based on a stated relationship or representation.	Give an estimate and keep it labelled as approximate.
Unknown	The evidence leaves more than one value possible.	Leave it unknown or state that the information is insufficient.

The most common mistake is to turn the third case into the first because the learner feels that a complete-looking table must have a complete set of numbers.

The Mechanism: Ask What Constrains the Blank

A missing value becomes inferable only when another piece of information constrains it. Work through these questions in order.

What quantity belongs in the blank? Identify the row, column, unit, object, time and condition.
Was this quantity directly measured? If not, do not call it an observation.
What rule or relationship connects the blank to known information?
Is that rule actually given or scientifically established for this exact case?
Does the rule produce one unique value?
Would a different value also fit the supplied evidence? If yes, the blank is not determined.
If the value is calculated, can you keep it separate from measured evidence?

The PSLE Science Reasoning Chain

READ THE TABLE AND GIVEN INFORMATION → IDENTIFY THE SCIENTIFIC QUANTITY → DISTINGUISH MEASURED FROM CALCULATED OR INFERRED → SELECT THE RELEVANT RULE OR CONCEPT → CHECK THE CONDITION UNDER WHICH IT APPLIES → DETERMINE WHETHER ONE VALUE FOLLOWS → STATE THE VALUE OR KEEP IT UNKNOWN → CHECK AGAINST THE EVIDENCE.

Worked Example 1: A Missing Value That Is Determined by a Given Total

Consider an original practice table describing how 24 identical counters representing particles are distributed between two labelled regions of a model. The question explicitly states that no counters are added or removed during the stage shown. The table records 9 counters in Region A but leaves Region B blank.

Here, the blank can be deduced because the question itself fixes the whole-system total at 24 and states that the model contains only the two counted regions. Region B must therefore contain 15 counters.

But notice the evidence status. “15” was not directly observed in the table. It was calculated from the stated total and the recorded value. That distinction matters if a later question asks which values were measured and which were derived.

Worked Example 2: A Blank Between Two Time Points

A table records a quantity as 12 units at 0 minutes, leaves the 5-minute value blank, and records 20 units at 10 minutes.

Many learners immediately write 16 because it is halfway between 12 and 20. That answer assumes a constant rate of change. But the two measured endpoints alone do not establish what happened at 5 minutes. The quantity could have changed quickly at first and then slowly, slowly at first and then quickly, remained unchanged for a period, or followed another path entirely.

If the question separately states that the quantity increases at a constant rate throughout the ten minutes, the situation changes: the missing value is now constrained by an explicit rule. Without that rule, 16 is an invented interpolation rather than established evidence.

Worked Example 3: A Pattern That Looks Tempting but Does Not Prove the Next Value

An original table shows results of 4, 8 and 12 units for three tested conditions. The fourth condition is listed but its result is blank.

It is tempting to write 16. But three equally spaced results do not, by themselves, prove that the relationship must continue in the same way. If the scientific rule supplied in the question says the result increases by exactly 4 units for every one-step increase in the tested condition across this range, then 16 follows. If the table is simply experimental data, the fourth result remains unknown until it is measured or a justified model is supplied.

This is why a visible pattern and a scientific law are not the same thing.

Worked Example 4: A Blank That Represents “Not Recorded”, Not Zero

A results table has a blank cell because one measurement was not taken. Another cell contains 0.

Zero is a recorded numerical result. Blank may mean missing, not measured, not applicable or not provided, depending on the table. Treating the blank as zero silently changes the evidence. Read the legend, heading, footnote and question text before deciding what the blank means.

Worked Example 5: A Calculated Quantity in a Science Table

Suppose a table gives a starting mass and a final mass for a model system and has a blank column titled “change in mass”. If the question defines change as final value minus starting value, the blank can be calculated from those two supplied values.

The calculation is legitimate because the table heading and rule identify the quantity. But the resulting number remains a calculated value. It should not be described as a direct measurement unless the method actually measured that change directly.

Five Tests Before You Fill a Blank

Test	Question	If the answer is no
Quantity test	Do I know exactly what this cell represents?	Stop and read headings, units and labels.
Rule test	Is there a stated or justified relationship connecting it to known values?	Do not calculate yet.
Condition test	Does that relationship apply under this exact condition, time and object?	Do not transfer the rule.
Uniqueness test	Does only one value satisfy the information?	The value is not determined.
Evidence-status test	Can I say whether the value is measured, calculated, estimated or inferred?	Repair the provenance before using it later.

Observable Failure Signatures

The learner fills every blank automatically.
The learner assumes equally spaced rows must give equally spaced results.
The learner treats a missing value as zero.
The learner uses a trend outside the tested range without permission or evidence.
The learner calculates with values that measure different quantities.
The learner ignores a unit change in the missing column.
The learner writes an exact number even though only a range is supported.
The learner calls a calculated value an observation.
The learner uses a rule from one set-up after the condition has changed.
The learner refuses to leave anything unknown because an incomplete table feels uncomfortable.

Earliest Weak-Link Diagnosis

What you observe	Earliest likely weak link	Repair
Wrong number but correct arithmetic	Quantity or condition misread	Relabel the cell before calculating.
Assumes smooth change between measurements	Evidence-limit failure	Separate measured points from unmeasured intervals.
Uses a rule that was never given	Inference stacking	Name the source of every relationship.
Cannot decide whether to calculate	Role confusion	Ask what the column means and how it relates to known values.
Gives a plausible but non-unique answer	Uniqueness failure	Try to construct a second value that also fits the evidence.
Correctly leaves blank but cannot explain why	Evidence reasoning incomplete	State which required relationship or measurement is missing.

Misconception Repair

“Tables should be complete.” Scientific records are allowed to contain missing information. Completeness is not more important than truth.

“A neat pattern lets me fill the next value.” A pattern may suggest a prediction, but a prediction is not the same as a measured result or a logically determined value.

“Between two points, the halfway time has the halfway value.” Only if a suitable linear relationship or constant rate is given or otherwise justified for that interval.

“If I can calculate something, it must be what the table wants.” First identify the scientific quantity and unit. Arithmetic that answers the wrong quantity is still wrong.

A Missing-Value Protocol for PSLE Science

Read the row label, column heading and unit.
Name the object, condition and time attached to the blank.
Mark known values as measured, given or calculated.
Find the exact relationship that might determine the blank.
Check whether that relationship applies to this cell.
Ask whether more than one value could still fit.
If one value follows, calculate it and label its evidence status.
If several values remain possible, leave the value unknown.
Use the inferred value later only with the same scope and conditions.

Original Practice: Deducible, Estimable or Unknown?

Create short practice tables where the learner’s only job is to classify each blank before calculating anything.

Case A: total explicitly stated; one component missing.
Case B: two time points measured; middle time blank; no rate rule.
Case C: a formula or relationship is supplied in the question.
Case D: a categorical condition is missing rather than a number.
Case E: the blank means measurement not taken.
Case F: an approximate graph-reading task explicitly asks for an estimate.

The learner should first say deducible, estimable or unknown, then explain why. This trains evidence boundaries before arithmetic fluency takes over.

Unfamiliar Transfer Challenge

An unfamiliar table uses letters instead of familiar Science objects. Column 1 is a tested condition. Column 2 is a directly measured outcome. Column 3 is a second quantity calculated from Columns 1 and 2 using a rule printed below the table. One result in Column 2 is blank, while Column 3 for that row is supplied.

The learner should not guess from the visible pattern. Instead, read the supplied calculation rule and ask whether it can be reversed uniquely to recover the missing measured outcome. If it can, the missing number can be mathematically derived from the given calculated quantity—but it is still not a direct observation. If the rule allows more than one possible original value, the blank remains unresolved.

Delayed Independent Return Test

After a meaningful gap, give a mixed table containing one true zero, one unrecorded blank, one deducible calculated value and one tempting but non-deducible pattern gap. Do not label the cases. Ask the learner to identify the status of each one before doing any calculation.

Success means the learner can resist completing the non-deducible blank even when the visual pattern strongly suggests a number.

Answer-Checking Receipt

I know what scientific quantity the blank represents.
I checked its unit, object, condition and time.
I know whether the missing value was supposed to be measured or calculated.
I can point to the rule or evidence that determines it.
I checked that the rule applies here.
I checked whether the answer is unique.
I did not turn a visual pattern into a law.
I did not turn “blank” into zero.
I can leave the value unknown when evidence runs out.
I preserve whether the value is measured, calculated, estimated or inferred.

Common Traps

Interpolating between experimental points without a stated relationship.
Extrapolating past the measured range as though the trend must continue.
Using a total that belongs to a different system or time.
Adding categories that overlap and then using the false total to infer a blank.
Mixing final value with amount of change.
Comparing numbers that have different units or denominators.
Ignoring a note that says a result was not recorded.
Giving false precision to an approximate graph reading.

Parent and Tutor Teaching Guide

If a child fills a blank too quickly, do not begin by correcting the arithmetic. Ask, “What makes that number the only possible number?” This single question often reveals whether the learner is using a given rule or merely following visual pattern familiarity.

If the child says, “It obviously goes 4, 8, 12, 16,” ask whether the values were measured or generated by a rule. Then construct a counterexample: 4, 8 and 12 can be followed by many possible experimental results if no fixed rule is supplied. This teaches the difference between pattern recognition and evidence.

When the learner correctly refuses to invent a value, treat that as scientific success rather than lack of effort. Being able to say “the information does not determine this” is part of disciplined inquiry.

Useful Internal Routes

Authoritative and Research References

The official sources establish the current Singapore curriculum and assessment frame. The deduction checks and practice sequence here are teaching tools, not official marking formulas.

Quiet Return

Science is not made stronger by filling every space. It is made stronger by knowing exactly where the evidence stops.

When one value follows, derive it carefully. When only an estimate is justified, say so. When the information cannot decide, leave the blank honest.

Series route: Previous: How to Answer “Justify” Questions in PSLE Science · PSLE Science Learning Guide · Next: How to Diagnose a PSLE Science Failure After a Delay

Small Group Tutorials