2.1 Table Analysis & Data Interpretation

The Table Analysis question format is a staple of the Executive Assessment's Integrated Reasoning (IR) section. It simulates the real-world task of analyzing a spreadsheet to extract actionable insights. You will be presented with a table of data, typically containing 5 to 25 rows and 4 to 8 columns. A critical feature of this format is the ability to sort the table by any column using a drop-down menu. You will then evaluate three binary statements (e.g., Yes/No, True/False, or Supported/Not Supported) based solely on the provided table.

The Structure of Table Analysis

Each Table Analysis prompt consists of three components:

1. The Scenario: A brief introductory text explaining the context of the data. Read this carefully, as it often defines terms or units used in the table.
2. The Table: The data itself, with a sort function. Columns may contain numerical data (e.g., revenue, population), categorical data (e.g., country names, industry sectors), or percentages.
3. The Question: Three distinct statements. You must select one of two mutually exclusive options (e.g., True or False) for each statement. There is no partial credit; all three must be correct to earn a point for the question.

Strategic Sorting

The ability to sort is your most powerful tool. Never scan raw, unsorted data to find a maximum, minimum, or to evaluate a trend. Always use the sort function.

Common sorting strategies:

• Finding Extremes: To find the highest or lowest value in a category, sort by that column. The answer will immediately appear at the top or bottom.
• Locating the Median: Sort by the relevant numerical column. If there are 21 rows, the median is the 11th value. If there are 20 rows, it is the average of the 10th and 11th values. Counting down the sorted list is vastly faster and more accurate than estimating.
• Evaluating Correlations: If a statement claims that "as Revenue increases, Profit Margin decreases," sort by Revenue. Then, scan the Profit Margin column to see if it generally decreases as you move down the sorted Revenue list.
• Grouping Categories: If you need to evaluate all "Technology" companies, sort by the "Sector" column to group them together.

Calculating Statistical Measures

While sorting is visual, many statements require calculation. You have access to a basic on-screen calculator, but estimation is often sufficient and faster.

1. Percentages and Ratios: Statements frequently ask whether one value is a certain percentage of another, or if a ratio exceeds a specific threshold.

• Example: "Company A's revenue is more than 25% of the total revenue of all listed companies."
• Strategy: Instead of calculating the exact percentage (Revenue A / Total Revenue), calculate 25% of the Total (Total / 4) and compare it to Company A's revenue.

2. Means and Medians:

• Mean (Average): Sum of all values divided by the number of values. Be cautious of statements comparing the mean to the median, especially in skewed data.
• Median: The middle value when sorted. Always sort before finding the median.
• Range: The difference between the maximum and minimum values. Sort by the column, take the top value, and subtract the bottom value.

A Worked Example

Consider the following simplified table representing five fictional cities:

Statement 1: The city with the lowest population density has the lowest percentage of commuters. Evaluation: Density = Population / Area. Let's estimate density: Alpha: 450/50 = 9 Beta: 1200/120 = 10 Gamma: 850/100 = 8.5 Delta: 300/40 = 7.5 Epsilon: 2100/300 = 7 Epsilon has the lowest density (7). Epsilon also has the lowest percentage of commuters (30%). True.

Statement 2: The median population of the cities is greater than 1,000,000. Evaluation: Sort by Population: Delta (300), Alpha (450), Gamma (850), Beta (1200), Epsilon (2100). The median is the 3rd value: Gamma (850). Note the unit trap! The table is in thousands. Gamma's population is 850,000. This is NOT greater than 1,000,000. False.

Statement 3: More than half of the total population across all five cities are commuters. Evaluation: You cannot simply average the commuter percentages. You must calculate the raw number of commuters for each city and sum them up. Total Population = 450 + 1200 + 850 + 300 + 2100 = 4900 (thousands). Half of total = 2450. Commuters: Alpha (450 * 0.65 = 292.5), Beta (1200 * 0.40 = 480), Gamma (850 * 0.55 = 467.5), Delta (300 * 0.70 = 210), Epsilon (2100 * 0.30 = 630). Total Commuters = 292.5 + 480 + 467.5 + 210 + 630 = 2080. 2080 is not more than 2450. False.

Common Traps to Avoid

1. Unit Conversions: Always check the column headers. Are numbers in thousands, millions, or billions? A statement might claim a revenue is "$5,000," but if the column header says "(in millions)," the actual revenue is $5 billion.
2. Subgroup vs. Total Population: A statement might ask about the average of a subset of the data (e.g., "European countries"), but you accidentally calculate the average for the entire table. Always verify the scope of the statement.
3. Averages of Averages: As demonstrated in the example above, you cannot find the overall percentage by simply averaging the percentages of individual rows. You must calculate the weighted totals.
4. Misreading Column Headers: In the rush of the exam, candidates often confuse similarly named columns (e.g., "Total Revenue" vs. "Net Revenue"). Double-check which column the statement refers to before sorting or calculating.