ACT Data Interpretation: Graphs, Tables & Charts

What You Need to Know

On the ACT, data interpretation questions test whether you can quickly extract meaning from graphs, tables, and charts. Most are not “hard math” — they’re about reading carefully, matching units, and avoiding traps like misleading scales or confusing variables.

What this topic includes (what you’re expected to do)

Read values from tables/graphs (including between tick marks).
Compare groups/conditions (multiple lines/bars/columns).
Identify trends (increasing, decreasing, leveling off, peaking).
Compute simple quantities from the data: differences, ratios, percent change, rates, and averages.
Interpolate (estimate between points) and sometimes extrapolate (predict beyond the data — usually with caution).
Handle graph “gotchas”: non-zero axes, uneven scales, log scales, dual axes, stacked bars, error bars.

Core rule

Always answer the question using the correct variables and units from the axes/labels/legend. Most misses come from rushing this step.

Critical reminder: Many ACT questions are “Which is supported by the data?” Your job is to choose what the graph/table actually shows, not what “should be true.”

Step-by-Step Breakdown

Use this process on every graph/table question to stay fast and accurate.

1) Decode the display (10 seconds)

Read the title/caption (what is being measured?).
Identify x-axis (usually the input) and y-axis (output) or the table’s row/column headers.
Check units (seconds vs minutes; $mg$ vs $g$ ; $\%$ vs fraction).
Locate the legend/key (which line/bar corresponds to which condition?).

2) Find what the question is really asking

Underline the two variables you must connect (e.g., “At $t = 4\,s$ , what is the temperature?”).
Decide the operation: read value, difference, ratio, rate, percent change, max/min, trend, or “supported by.”

3) Pull the correct numbers (or estimate correctly)

Direct read: go to the exact x-value and read y.
Interpolation: if between gridlines/points, estimate proportionally.
Table lookup: match the right row and column.

If the axis scale is weird (skips, unequal spacing), slow down and count tick marks.

4) Compute only what you need (keep it simple)

Use quick arithmetic: differences first, then divide if needed.
Keep units attached mentally (or jot them): it prevents wrong-choice traps.

5) Sanity-check (5 seconds)

Does the answer have the right units and magnitude?
Is it consistent with the graph’s shape (e.g., shouldn’t increase if the curve is decreasing)?

Mini worked walkthrough (typical “read + compute”)

A line graph shows distance vs time. At $t = 2\,s$ , $d = 8\,m$ . At $t = 6\,s$ , $d = 20\,m$ . Average speed from $2$ to $6\,s$ :

$\text{speed} = \frac{\Delta d}{\Delta t} = \frac{20 - 8}{6 - 2} = \frac{12}{4} = 3\,m/s$

Decision points:

If the question says “average,” use endpoints.
If it says “instantaneous” on a curve, approximate slope at that point (tangent idea).

Key Formulas, Rules & Facts

High-yield calculations you actually use

Quantity	Formula	When to use	Notes / traps
Difference (change)	$\Delta y = y_2 - y_1$	“How much more/less”	Watch order: “increase” implies positive.
Rate of change (slope)	$m = \frac{y_2 - y_1}{x_2 - x_1}$	“per,” “rate,” “slope,” “average speed”	Units become $\frac{\text{y-units}}{\text{x-units}}$ .
Percent change	$\%\,\text{change} = \frac{\text{new} - \text{old}}{\text{old}} \times 100\%$	“percent increase/decrease”	“Old” is the original baseline.
Ratio / factor	$\text{factor} = \frac{A}{B}$	“how many times,” “ratio”	If asked “A compared to B,” do $\frac{A}{B}$ .
Average (mean)	$\bar{x} = \frac{\sum x}{n}$	Average of listed values	Don’t average endpoints unless asked.
Weighted average	$\bar{x} = \frac{\sum (w_i x_i)}{\sum w_i}$	Different group sizes / frequencies	Often hidden in tables with counts.

Graph-reading rules that matter on ACT

Independent vs dependent: x-axis is usually independent; y-axis dependent.
Trend language:
- “Increasing at a constant rate” looks like a straight line with positive slope.
- “Increasing at an increasing rate” curves upward (slope gets steeper).
- “Increasing at a decreasing rate” rises but levels off.
- “Peak/maximum” is the highest y-value; “minimum” is lowest.
Interpolation vs extrapolation:
- Interpolate between given data points (usually safer).
- Extrapolate beyond data only if asked; be skeptical.

Common chart types and how to read them fast

Bar chart: compare categories; check if bars are grouped/stacked.
Stacked bar: total height is sum; segments show parts.
Pie chart: parts of a whole; convert percent to fraction of total.
Scatterplot: relationship/correlation; may need line of best fit.
Two-line graph: compare conditions; watch legend colors/styles.
Dual-axis graph: two y-axes with different scales/units (major trap).

Unit conversions that show up often (keep them straight)

Time: $1\,min = 60\,s$ , $1\,h = 60\,min$ .
Metric: $1\,g = 1000\,mg$ , $1\,kg = 1000\,g$ .
Length: $1\,m = 100\,cm = 1000\,mm$ .

ACT graphs frequently mix units in labels or answer choices. Always match what the axis uses.

Examples & Applications

Example 1: Table lookup + difference

A table lists reaction yield (in $\%$ ) for catalysts A and B.

Catalyst A at $40\,^\circ C$ : $68\%$
Catalyst B at $40\,^\circ C$ : $74\%$

Question: “How much higher is B than A at $40\,^\circ C$ ?”

Compute:

$74\% - 68\% = 6\%$

Key insight: It’s a difference in percentage points, not a percent increase.

Example 2: Percent change from a graph

A line graph shows population increasing from $1200$ to $1500$ .

Question: “What is the percent increase?”

$\%\,\text{increase} = \frac{1500 - 1200}{1200} \times 100\% = \frac{300}{1200} \times 100\% = 25\%$

Key insight: The denominator is the original value.

Example 3: Slope/rate with units

A distance–time graph shows:

$t = 1\,h$ , $d = 55\,km$
$t = 3\,h$ , $d = 175\,km$

Average speed between $1$ and $3\,h$ :

$\text{speed} = \frac{175 - 55}{3 - 1} = \frac{120}{2} = 60\,km/h$

Key insight: Rate questions are slope questions, and units must match.

Example 4: “Supported by the data” with two lines

A graph shows temperature vs time for Material X and Y.

From $0$ to $10\,min$ : X rises from $20\,^\circ C$ to $50\,^\circ C$ .
Over the same time: Y rises from $20\,^\circ C$ to $35\,^\circ C$ .

Claim choices include:
A) “X heats faster than Y from $0$ to $10\,min$ .”
B) “X will reach $100\,^\circ C$ by $20\,min$ .”

Correct logic:

A is supported (bigger increase over same time).
B is not necessarily supported unless the graph shows linear behavior or includes that range.

Key insight: ACT loves tempting extrapolation claims.

Common Mistakes & Traps

Reading the wrong axis/variable
- What goes wrong: You treat the y-axis as time or mix up rows/columns.
- Why it’s wrong: You’re answering a different question.
- Fix: Before reading a value, say (in your head): “At $x = \_\_$ , $y = \_\_$ .”
Ignoring units (seconds vs minutes, $mg$ vs $g$ )
- What goes wrong: Your number is correct but in the wrong unit.
- Why it’s wrong: Answer choices are unit-sensitive.
- Fix: Write the unit next to your extracted value (even just a letter).
Missing non-zero baselines or broken scales
- What goes wrong: You compare bar heights as if the axis starts at $0$ .
- Why it’s wrong: Visual differences look larger/smaller than reality.
- Fix: Check the lowest labeled tick. If it’s not $0$ , comparisons must use actual values.
Confusing percent change with percentage-point change
- What goes wrong: From $40\%$ to $50\%$ , you say “up $10\%$ ” when the question wants percent increase.
- Why it’s wrong: Percent increase is relative to the starting value.
- Fix: If it says “percent increase/decrease,” use
  $\frac{\text{new}-\text{old}}{\text{old}} \times 100\%$ .
Picking a correlation/causation trap
- What goes wrong: You claim “A causes B” from a scatterplot.
- Why it’s wrong: A trend does not prove causation.
- Fix: Unless the experiment design is stated (controlled variables), stick to “is associated with.”
Using the wrong two points for slope
- What goes wrong: You grab points that aren’t on the line/curve or mix series.
- Why it’s wrong: Rate depends on consistent points for the same dataset.
- Fix: For a straight segment, pick two clear grid-intersection points on that segment.
Forgetting the legend when multiple lines/bars exist
- What goes wrong: You read Material Y’s line thinking it’s Material X.
- Why it’s wrong: Many questions hinge on small differences between series.
- Fix: Trace from the legend to the actual line style/color before reading.
Over-extrapolating beyond the data
- What goes wrong: You assume linear continuation when the curve is bending.
- Why it’s wrong: The ACT often tests “not enough info.”
- Fix: Only extrapolate if the prompt explicitly asks; otherwise, prefer “cannot be determined.”

Memory Aids & Quick Tricks

Trick / mnemonic	What it helps you remember	When to use it
“TLC: Title, Labels, Caption”	The fastest way to orient yourself	Start of every graph/table set
“X then Y”	Always read $y$ _at_ a given $x$	Any “At $x = \_\_$ ” question
“Old on bottom”	Percent change baseline is original value	Percent increase/decrease
“Slope = \u0394 over \u0394”	Rate is change in output over change in input	Speed, density-like rates, “per”
“Legend before numbers”	Prevents mixing datasets	Multi-line or multi-bar displays
“Check the zero”	Bar graphs can mislead if y-axis doesn’t start at $0$	Bar/column comparisons
“Inside = interpolate, Outside = extrapolate (skeptical)”	How safe your estimate is	Estimation questions

Quick time-saver: If answer choices are far apart, you often only need a rough read (nearest tick) rather than a perfect estimate.

Quick Review Checklist

[ ] I read the title/caption and know what’s being measured.
[ ] I identified x vs y, plus units.
[ ] I checked the legend and selected the correct series.
[ ] I noticed any weird scale (non-zero start, uneven ticks, dual axes).
[ ] I know whether the question needs a value, difference, ratio, percent change, or slope.
[ ] My computed answer has the right magnitude and units.
[ ] I did not assume causation or extrapolate unless asked.

You don’t need advanced math here — just disciplined reading and clean, minimal calculations.