Table Data Interpretation: Columns, Rates and Missing Values

A table gives you exact numbers, so there is no reading error from a scale, but it hides the answers in columns you have to build yourself. A typical bank or SSC table has four to six rows and two to four columns, and the questions ask for rates, totals and comparisons that are not printed.

The table below holds invented practice data: the number of students who appeared for, and passed, an exam in five schools.

The last column and the total row are not given in a real question; you add them. Spending a few seconds on derived columns before reading the questions pays off, because most questions in the set use them.

This is the single most common trap in table DI. The overall pass rate is total passed ÷ total appeared, not the average of the five pass rates.

If the options include both numbers, the examiner is testing this exact point. Use totals unless the question says "average of the pass percentages".

Some tables leave cells blank and give a rule or a total that lets you fill them. Solve for the missing cells first, then answer the questions.

In caselets, the whole table is "missing" and has to be built from a paragraph; the method is in caselet DI.

A table set usually has five questions on the same rows, and they rarely need every column. A short routine keeps the set quick:

1. Skim all five questions first. Note which derived figures they need, such as a rate, a failure count or a total.
2. Build only those columns, in pencil beside the table, and add a total row if any question mentions "overall" or "all schools".
3. Answer the look-up questions first. A single-row rate or ratio takes seconds once the column exists.
4. Leave the "how many rows satisfy" questions for last. They need a comparison against every row, but by then all the rates are already written down.

Two habits prevent most errors. Read the column headings for units, because one column may be in hundreds and another in actual numbers. And when a question names two rows, circle both before reading across, since adjacent rows in a dense table are easy to swap.

• ✘ Averaging percentages when the question needs the overall rate.
• ✘ Using the wrong row as the base in "more than" and "percentage of" questions.
• ✘ Forgetting that a column can be derived (failures = appeared − passed).
• ✘ Mis-copying a number from a crowded table; underline the row before you read across.

Related methods: bar graph DI, pie chart DI, line graph DI and weighted averages. Shortcuts are in the Quant cheat code.

Why is the overall percentage different from the average of the percentages?

The average of percentages gives every row equal weight, but rows have different sizes. The overall percentage weights each row by its size, so large rows count for more. Always divide the column totals unless the question asks for a simple average.

How do I find a missing value in a table?

Look for a relationship between columns, such as passed equals appeared times pass rate, or a total row. Write that relationship as an equation with the missing value as the unknown, then solve it before attempting the questions.

All problems and solutions on this page are original. No outside facts are relied on; every numeric answer is recomputed by software before publishing.

• Quant Cheat Code: 20 Must-Know Shortcuts & Formulas