Interpreting Information

Interpreting Information (II) gives you a passage of text or a data exhibit (chart, table, graph) followed by 5 statements to mark Yes or No. Like Syllogisms, II uses the drag-and-drop format: 5/5 = 2 marks, 4/5 = 1 mark, ≤3/5 = 0.

These make up ~5-6 questions (~14% of DM). Target time: 60-90 seconds per set.

Target time and triage

II is one of the best mark-per-minute trades in DM alongside RA and Probability. Two marks available, technique is mechanical, the trap library is small. Strong opening-pass priority.

The catch: passage length varies enormously. A 3-sentence passage on a familiar topic resolves in 60 seconds. A 10+ sentence dense paragraph on an unfamiliar topic can swallow 3 minutes. Triage on length and topic familiarity before committing.

1. Passage-based II. A block of scientific, medical, or statistical text + 5 yes/no statements. Same technique as VR with one critical difference (below).
2. Data-based II. A chart, table, bar graph, or pie chart + 5 yes/no statements. Generally easier and faster than passage-based. You usually get 1-2 data questions per paper and 3-4 passage questions.

This distinction changes the standard of evidence for every statement.

1. Scan the first sentence and triage. Read sentence 1: what topic? How long is the passage?

• ≤6 sentences on a familiar topic → attempt.
• 7-10 sentences → attempt but scan, don't read word-by-word.
• >10 sentences on an unfamiliar topic → flag and skip. Best-guess all 5 to keep the 1-mark partial credit possible.

2. Read the first statement. Don't read all 5 up front. Take them one at a time.

3. Pick a keyword - the irreplaceability test. A good keyword is hard to swap for a synonym: a proper noun, a technical term, a number, a date. "Upwelling" is irreplaceable. "Important" is not. Bad keywords scan to the wrong sentences. The Verbal Reasoning module covers keyword selection and scanning technique in depth - the same approach applies here.

4. Scan the passage for the keyword. Read one or two sentences around it. Do not re-read the whole passage.

5. Verdict. Apply the yes/no rule (below). Move to the next statement.

   Find the matching content. Then ask:

   ├── Passage directly supports the statement (or one-step inference)
   │     → Yes
   ├── Passage contradicts the statement
   │     → No
   ├── Passage doesn't address this topic / claim
   │     → No
   ├── Statement covers only part of a multi-part claim
   │     → No
   └── Statement exaggerates the passage
       ("can reduce" → "eliminates"; "up to 50%" → "by half")
         → No

Unlike VR, silence is "No" not "Can't Tell." If the passage doesn't support it, it's No.

Watch the statement's modal words. They flip verdicts.

Upwelling is a local response of the ocean to surface wind. The warmer, nutrient-depleted surface water is replaced by dense, cooler water that rises to take its place. Speculation about the association of these cooler waters with the enhanced growth of phytoplankton on western tropical coasts has gained ground since chemical analysis of the water revealed a high concentration of dissolved gases along with nutrients like nitrates, phosphates, and silicates. Phytoplankton uses light-harvesting chlorophyll and forms the base of several aquatic food webs.

Mental map after one scan: S1 = upwelling defined. S2 = mechanism (warm replaced by cool). S3 = speculation about phytoplankton + chemical evidence. S4 = phytoplankton's ecological role.

Data questions are often the easiest marks in the entire DM section. Different sequencing:

1. Read the statement first. Don't study the chart up front.

2. Identify what data point the statement claims. "Individual gift donations rose between 2020 and 2021" → I need individual gifts in 2020 vs 2021. Nothing else.

3. Go straight to the data and check. Eyeball the bars/cells. Verdict in seconds.

4. Watch for inferred causation. If the statement adds "because of an economic downturn" to a falling bar, the data shows the fall but not the cause. Verdict: No.

A bar chart shows charity income across three sources (individual gifts, corporate sponsorship, government grants) in 2020 and 2021. Individual gifts rose; corporate sponsorship fell; government grants held steady.

Trap 1: Exaggeration

The passage qualifies a claim. The statement removes the qualifier.

Trap 2: Correlation vs causation

Any time the statement uses "because," "leads to," "causes," "results in," check whether the passage actually establishes that causal link or just describes co-occurrence. The data shows two things moving together; the statement says one causes the other. Almost always No.

Trap 3: Misreading scales and axes

For data questions:

• What does the y-axis measure? (absolute numbers vs percentages vs rates per 1,000)
• Where does the y-axis start? (a chart starting at 30 makes 40→60 look enormous)
• Are the intervals even? (a time series with missing years can mislead)

Trap 4: Going beyond the passage

The passage doesn't mention X. The statement makes a claim about X. Default to No - silence is not support.

II passages often embed numbers that you need to calculate with.

"It takes approximately 3.5 hours to add layers of egg white and shell. The complete egg weighs about 58 g, of which 18 g is the yolk."

Statement: "The layers other than the yolk weigh about 2.23 times more than the yolk."

58 − 18 = 40 g (non-yolk). 40 / 18 = 2.22. Statement says "about 2.23" → Yes.

Rule: "About" and "approximately" give you rounding room. Within a few percent is fine.

Scan sentence 1 of the passage.
    |
    ├── ≤6 sentences, familiar topic
    │   → Attempt. 60-75s target.
    │
    ├── 7-10 sentences OR moderately technical
    │   → Attempt. Scan-don't-read. 90s target.
    │
    └── >10 sentences AND opaque topic
        → Flag. Best-guess all 5 statements (you might still get 1 mark
          partial credit). Return only if banked time after Venn/Prob.

Data-based II should almost never be skipped.

1. Applying VR's "Can't Tell" threshold. In II, reasonable one-step inference is allowed. "Auto-immune" → "the body's own immune system attacks itself" is valid in II but Can't Tell in VR. Adjust your threshold.
2. Accepting "could be true." Could ≠ must. If there's a plausible reading where the statement is false, it's No.
3. Reading the whole passage before looking at statements. Wastes time. Map the passage (10s scan) then test each statement against its keyword.
4. Missing the certainty trap. "Directly," "must," "always," "never" almost always flip the answer to No.
5. Importing causation from correlation. A "because" clause attached to a real trend is usually a No.

II questions test three skills:

• D1: Scientific / Technical Passage Inference - close reading of dense text, distinguishing what's stated from what's inferred. The keyword-scan technique avoids re-reading.
• D2: Numerical / Quantitative Passage Inference - extracting numbers from text and verifying claims arithmetically. Arithmetic must be exact unless "about" appears.
• D3: Graphical / Chart Data Interpretation - reading bar charts, tables, line graphs. Watch axes and units; the data shows trends, not causes.