A child can read every number in a table and still choose the wrong one. Before adding calculation practice, ask whether they can identify what the rows describe, what the columns describe and which intersection answers the question. The difficulty may begin in locating information, not in doing mathematics with it.

Use a small table with clear headings and one direct lookup question. Ask the learner to name the row and column before giving the value, then explain the answer as a complete fact. Compare that with any later calculation. These observations help choose teaching or access support; they do not diagnose a visual, attention or learning condition.

A number has an address and a meaning

The following table is an invented example, not data from an actual school. It records attendance at three clubs on two different days. Each number is the attendance count for one club on one day.

For “How many attended Coding on Thursday?”, the answer is 14. The 8 is also a valid number in the Coding row, but it belongs to Tuesday. The 9 belongs to Thursday, but it describes Art. The correct cell must satisfy both conditions.

If the learner answers 8, do not immediately conclude that they confused the numerals. Ask what that 8 describes. They may have found the right club while missing the day. That points towards coordinating two labels rather than learning to count.

The useful repair begins by restoring the number’s meaning: “Fourteen attended Coding on Thursday.” A bare numeral can conceal the route by which it was selected.

Reading across and down is not enough by itself

Adults often tell children to go across and then down. That can be a helpful movement instruction, but it does not tell the learner which labels to follow or why their intersection matters. A familiar scanning routine can still arrive at the wrong cell.

Begin with the structure. In this example, the rows distinguish clubs and the columns distinguish days. Ask the child to point to those headings and say what changes as they move to another row or column.

The W3C’s guidance on tables with two headers explains why data cells need clear relationships with both row and column headings. Its focus is accessibility, but the same information relationship is visible in the child’s reading task.

Do not make table reading a race to reach a number. The first goal is to preserve what the number refers to. Speed becomes useful only when the lookup is accurate enough for the task.

Notice the shape of the error

A child may consistently choose from the correct row but the wrong column. They may read the first available number after finding a familiar label. They may choose a total when the question asks for one category, or a category when the question asks for a total.

Other errors appear when headings are abbreviated, when a table spans pages or when the learner must switch between the question and the data. A correct spoken lookup followed by an incorrect copied number points towards another transition to inspect.

Look for what works too. Is the same information understood when written as sentences? Does a smaller table help? Can the child locate the cell but struggle with a comparison calculation afterwards?

Record a few representative examples and the help provided. “Found Coding but used Tuesday instead of Thursday” is more useful than “cannot read tables.” It identifies a decision adults can explain and practise.

Separate lookup from calculation

Start with “How many attended Music on Tuesday?” This requires locating 10, not adding, subtracting or inferring. If that lookup is insecure, a question involving several calculations will make the first difficulty harder to see.

Next ask how much larger Coding attendance was on Thursday than Tuesday. The relevant values are 14 and 8, and the difference is 6. If the learner selects 14 and 8 correctly but subtracts incorrectly, the lookup has succeeded while arithmetic needs attention.

If the learner calculates 14 − 9 = 5, the arithmetic may be correct for the chosen numbers, but the selection does not answer the question. Five is the difference between Thursday Coding and Thursday Art, not between Coding on the two days.

That distinction prevents unnecessary reteaching of a secure calculation. Find whether the first mismatch occurred in interpreting the question, selecting the cells, choosing the operation or executing it.

Ask the child to say the conditions aloud

For the direct Coding question, a short plan is “Coding row, Thursday column.” Then locate the intersection and read its value. The plan is small enough to support the task without becoming a second worksheet.

Ask where each condition came from in the question. Otherwise the adult is still choosing the row and column while the child merely follows. The aim is for the learner to identify the conditions themselves.

The EEF’s metacognition guidance supports modelling and practising planning and monitoring within subject work. The short lookup routine here is an original application to try and review, not a validated test or a universal programme.

Once the relationship is understood, a spoken plan may no longer be needed on every question. Keep suitable supports where they help and reduce unnecessary prompting when the learner can manage the decision.

Use a small contrast to show what each label controls

Keep the day fixed and change the club: Art on Thursday is 9; Coding on Thursday is 14. Ask what changed in the lookup and what stayed the same. The Thursday column remains relevant while the row changes.

Then keep the club fixed and change the day: Coding on Tuesday is 8; Coding on Thursday is 14. Now the row stays the same while the column changes.

This contrast makes the two conditions visible without adding harder arithmetic. A learner who can explain it is doing more than memorising the location of 14 on this particular page.

Do not present a large collection of near-identical questions just because the first comparison worked. Move to a new, appropriate table after the basic relationship is clear, and check whether the child can identify the headings there.

Give the learner a way to keep the place

A finger, a plain ruler or a temporary marker can help some learners follow a row on paper. Ask whether it is comfortable and useful. The tool should support finding the intended cell, not force a particular posture or become a test of coordination.

On a screen, zooming or changing the view may help, provided the headings remain visible. Enlarging a number while moving its header offscreen can solve one problem and create another.

If the child uses assistive technology, the source needs to preserve the table’s structure. An image of a table may not provide the same access as a properly structured table. Ask the school for an appropriate version rather than assuming the learner must adapt to every format.

Improvement with a marker or clearer layout identifies a useful support. It does not prove a visual-processing or attention diagnosis. Keep the educational observation separate from professional assessment.

Check the table before testing the child

A cropped photograph may lose the column headings. A worksheet copied faintly may make lines or numbers difficult to distinguish. A table pasted into a message may wrap so that values no longer sit beside the correct labels.

Ask what the learner actually received. Compare it with the original where possible. Do not judge their response against an intact teacher copy if their own version is incomplete or unreadable.

W3C’s tables guidance emphasises structural relationships rather than relying on visual cues alone. For adults preparing materials, clear headers and a meaningful caption are part of making information accessible, not decorative extras.

A better source does not remove the learning task. It allows the child to encounter the intended one. Providing the missing header is different from identifying the answer cell for them.

Abbreviations and units can be the hidden language barrier

A child may understand rows and columns but not know what a heading means. “Mass/g” and “Time/min” require knowledge of both the quantity and the unit. A numeral without that relationship is incomplete information.

Ask the learner to read a cell as a sentence: “The time was eight minutes,” rather than just “eight.” If the meaning is uncertain, explain the heading and unit before asking for interpretation.

Do not confuse unfamiliar language with inability to reason. In multilingual settings, check whether the task vocabulary has been taught and whether the child understands the equivalent idea in familiar language.

The existing guide to reading tables and visual texts provides a broader language route. Here the immediate question is which label or relationship the learner needs to make sense of the selected value.

A total is a different request from an intersection

The question “How many attendances were recorded on Thursday?” requires adding the three Thursday values: 9 + 14 + 6 = 29. No single cell in the original table contains that total.

A learner may correctly identify Thursday and then choose 14 because it is the largest visible entry. Ask whether the question refers to one club or all the clubs. The missing step may be understanding the scope of “total,” not finding Thursday.

Similarly, the recorded Tuesday attendances total 12 + 8 + 10 = 30. Comparing the two daily totals gives one fewer attendance on Thursday. That requires several valid lookups and calculations, so teach those components deliberately.

Do not assume that every table question asks for aggregation. A child who adds all the numbers after hearing “how many” needs help distinguishing a specific lookup from a total.

Do not turn repeated attendance into unique people

Adding Art’s two counts gives 21 attendances across Tuesday and Thursday. It does not establish that 21 different children attended Art. Some children may have attended on both days.

This is an important limit in the meaning of the data. A learner can calculate 12 + 9 correctly and still label the result incorrectly. The problem is not the sum but the claim made from it.

Ask what was counted and whether the same person could appear more than once. The table’s caption and surrounding explanation should make that clear where the task requires it.

Keep the lesson age-appropriate. A younger child may only need the distinction between “visits” and “different visitors.” The broader principle is that correct arithmetic does not grant permission to change what the values measure.

Blank, zero and not recorded are different states

Suppose an additional club has a zero on Tuesday and “not recorded” on Thursday. The zero reports no attendance for the recorded Tuesday entry. “Not recorded” does not provide a Thursday attendance count.

Do not ask the child to turn every empty cell into zero. A blank can have different meanings, and a dash may require a note or legend. When the meaning is unclear, the correct action is to check the source rather than invent a value.

The existing eduKate explanation of what a blank means develops this distinction beyond school tables. It is a useful next route when missing information is being mistaken for a measured result.

At home, a simple question is enough: “Does the table tell us nobody attended, or does it not tell us how many attended?” Those statements are not interchangeable.

When the rows and columns exchange places

The same attendance information could be presented with days down the side and clubs across the top. A child who memorised “find the club on the left” may now struggle even though no data has changed.

Ask the learner to inspect the headings again. Which direction now contains the clubs? Which contains the days? The relationship survives the change in layout, but the movement required to reach the answer changes.

Oak National Academy’s lesson on two-way outcome tables notes that rows and columns can be transposed. Its probability context is more advanced; the useful introductory point here is to follow the labels rather than a memorised orientation.

Use a small suitable example before moving to complex tables. A layout change should reveal whether the relationship is understood, not become an attempt to overwhelm the learner.

Headers can have more than one level

An older learner may meet a table where Tuesday and Thursday each contain separate morning and afternoon columns. Selecting the correct day is no longer sufficient; the time period also matters.

Ask the learner to name every applicable heading before reading the cell. If the relevant question is Thursday afternoon Coding, all three conditions need to be preserved.

Begin with a clear small version and connect each header to its group of values. Do not infer that the child cannot use tables because a crowded multilevel example caused difficulty.

For material design, W3C’s multilevel-table guidance explains the need for explicit header associations and notes that splitting complex tables can sometimes improve usability. The educational demand should be chosen deliberately rather than created accidentally by poor layout.

A table and a graph do not use the same lookup routine

Both organise information, but a graph may require reading a scale or estimating a position between marked values. A table usually provides labelled entries directly. Do not assume that success with one format proves success with the other.

If a child selects cells accurately but misreads graph axes, move to the graph-reading need rather than repeating table practice. The existing article on why visual graph comparisons can fail addresses axes and scale separately.

Likewise, a correct table lookup does not establish that the learner can explain a scientific trend. Finding values, comparing them and explaining an underlying mechanism are different demands.

The practical-data learning manual is the broader science route. Use this guide to identify which earlier information-reading step needs support.

Find the point where copying changes the answer

Ask the learner to point to the intended cell and say its meaning before writing. If the oral answer is correct but the written response changes the numeral, the source selection may not be the main difficulty.

Check the transfer from table to answer space. Did the child return to a neighbouring cell? Did they copy a digit incorrectly? Was the answer box so far from the source that they lost their place?

A small reference note or a deliberate read-back may help. The objective is to preserve the selected information through the handoff, not to add a lengthy copying exercise.

Keep this distinct from the place-value column problem. The word “column” appears in both, but locating data and arranging digits for arithmetic are different reader jobs.

Use feedback that names the mismatch

“Check again” leaves the child searching the entire task. “You found the Coding row, but this value belongs to Tuesday” identifies what was correct and which condition was lost.

Ask the learner to locate the Thursday heading and repair the selection. Then use another question where the wrong-row and wrong-column alternatives are both plausible. The child should learn to coordinate the labels, not simply remember that the previous answer was 14.

If the operation was wrong after correct selection, say so. If the data meaning was unclear, teach it. Feedback should not collapse every incorrect final answer into a vague warning about carelessness.

Do not correct every task at once. Choose the recurring mismatch that most limits the learner’s current work and provide a manageable new opportunity to apply the repair.

Some comparisons need two complete addresses

A question may compare Art on Tuesday with Music on Thursday. Now there is no single row or column containing the entire answer. The learner needs two complete selections: Art–Tuesday is 12, and Music–Thursday is 6. The difference between those attendance counts is 6.

Ask the child to name both addresses before calculating. A learner may correctly find the first value and then remain in that row while searching for the second. Their usual scanning routine is working mechanically, but the second target requires changing both conditions.

A short note can preserve the comparison: “Art Tuesday 12; Music Thursday 6.” This is not unnecessary working if it prevents the values from losing their meaning. Once the learner can coordinate the selections reliably, they may need less visible support.

Check the direction of the requested comparison too. “How many more?” asks for a difference in the relevant direction; “How many altogether?” asks for a sum of the specified counts. A correct pair of values does not by itself choose the operation.

For a fresh task, change the clubs or days without making the arithmetic substantially harder. Notice whether the learner reads each full target rather than following the movement remembered from the previous question. The purpose is flexible information selection, not memorising a route around one table.

A practical school conversation

A helpful message might say: “My child locates the correct row but often uses the first number instead of checking the column heading. They can explain the answer after naming both labels. Could we practise a consistent lookup routine?”

Alternatively: “The cell selection is accurate, but totals and individual entries are confused.” That asks for a different explanation. Include a representative task and the child’s original attempt when useful.

Ask whether the teacher sees the same pattern and whether the source format creates an access barrier. A classroom version, a printed copy and a phone photograph may not be equivalent learning conditions.

Share information privately through the school’s normal channel. There is no need to post a child’s mistakes in a public group to obtain a more dramatic example.

What improvement should look like

Look for a new action: the learner reads the headings before choosing a number, names both conditions, or notices that the result belongs to another category. A fresh suitable table provides stronger evidence than repeating the same lookup until it is memorised.

Record the support that remains and keep necessary accommodations available. Independence can include using a line guide, an accessible table or a brief self-prompt.

Persistent substantial difficulty across tasks despite appropriate instruction deserves discussion with the teacher or learning-support staff, who can advise whether further assessment is needed. Table-reading errors alone do not identify a developmental or health condition.

Start with the number’s address: what does the row say, what does the column say, and what fact does their intersection express? The Parent Learning Support Directory helps you choose another route when the difficulty lies elsewhere.