Model Comparison: Interpretation of Data
═══════════════════════════════════════
Gemini 3 Pro
═══════════════════════════════════════
Generation failed: exception TypeError: fetch failed sending request
═══════════════════════════════════════
GPT 5.2 Pro
═══════════════════════════════════════
What You Need to Know
- Most ACT Science “Interpretation of Data” questions are lookup + logic—find the right graph/table, match the right variable/units, then read the value carefully.
- You must quickly recognize what each axis/column represents, including units, scales (especially non-1 increments), and legends.
- Many questions require comparing, estimating between points (interpolation), or projecting beyond shown data (extrapolation)—but only when the question explicitly asks.
- Light math shows up often—ratios, differences, percent change, and slope—done directly from the displayed data.
Curriculum anchor (ACT official skills, as publicly described by ACT, Inc.): the ACT Science Test assesses scientific reasoning—interpreting, analyzing, evaluating, and reasoning with scientific information presented in graphs, tables, and diagrams. The notes below focus on the “data representation/interpretation” skills repeatedly tested across Science passages.
Reading and Interpreting Tables and Graphs
Goal: Extract correct values from tables, line graphs, scatterplots, and bar graphs by tracking variables, units, and scales.
What to check first (fast routine)
- Title/caption: what relationship is being shown?
- Variables: identify the independent variable (often on the horizontal axis or leftmost column) and the dependent variable.
- Units: don’t separate number from unit (e.g., seconds vs minutes, \text{mg} vs \text{g}).
- Scale/ticks: confirm step size (e.g., tick marks may go by 2s, 5s, or logarithmically).
- Legend/key: match the correct line/symbol/bar to the correct condition.
Common graph types (how ACT tends to use them)
- Line graph: change over an ordered variable (time, temperature).
- Scatterplot: relationship/correlation; may imply a trend line.
- Bar graph: comparisons across categories.
- Table: exact values; may require finding the right row/column.
Worked example (table lookup + units)
A table lists reaction time in milliseconds for different temperatures.
- Question: “At 30^\circ\text{C}, what is the reaction time?”
- Steps: locate the row/entry for 30^\circ\text{C} → read the corresponding value → include milliseconds (not seconds).
Worked example (graph reading with non-1 scale)
A line graph has a y-axis labeled concentration with ticks at 0, 10, 20, 30.
- If a point sits halfway between 10 and 20, read it as 15 (not 11 or 12).
Exam Focus
- Why it matters: A large share of ACT Science questions are direct or near-direct reads from visuals—fast accuracy is a major score driver.
- Typical question patterns:
- “According to Figure/Graph/Table __, what is the value of …?”
- “Which of the following conditions corresponds to the dashed line?”
- “At ____ (given x), what is ____ (read y)?”
- Common mistakes:
- Reading the wrong series (wrong legend symbol/line style)—trace back to the key every time.
- Ignoring units or unit prefixes—say the unit to yourself as you read.
- Misreading tick spacing—check two adjacent labels before reading any point.
Recognizing Trends and Patterns in Data
Trend means a general direction in data (increasing, decreasing, leveling off), while a pattern can include peaks, cycles, plateaus, or thresholds.
Types of trends you should recognize
- Positive association: as x increases, y tends to increase.
- Negative association: as x increases, y tends to decrease.
- No clear relationship: points are scattered with no direction.
- Nonlinear behavior: curve; rate of change is not constant.
- Plateau/saturation: increases then levels off.
- Peak/optimum: rises to a maximum then falls.
Worked example (describing trend)
A graph shows enzyme activity vs temperature: activity rises from 10^\circ\text{C} to 40^\circ\text{C}, then drops at 60^\circ\text{C}.
- Trend description: “Activity increases up to an optimum around 40^\circ\text{C}, then decreases.”
Outliers
An outlier is a point far from the overall trend.
- ACT may ask what could explain it (measurement error, different condition), but you usually answer using only information given.
Exam Focus
- Why it matters: Many questions ask you to interpret “what the data show” rather than compute—trend recognition is the bridge from numbers to conclusions.
- Typical question patterns:
- “Which statement best describes the relationship between …?”
- “At what value of ____ is ____ maximized/minimized?”
- “Which interval shows the greatest increase/decrease?”
- Common mistakes:
- Overstating conclusions (e.g., claiming causation from correlation)—stick to “associated with” unless the experiment design supports causation.
- Missing a nonlinear change—check multiple intervals, not just endpoints.
- Ignoring the question’s specified interval—re-read the range asked.
Translating Between Data Representations
Translating representations means moving between words, tables, and graphs (and sometimes between two different graphs of the same situation).
Common translation tasks
- Graph → sentence: describe trend, compare conditions, identify maxima/minima.
- Table → graph idea: recognize which variable belongs on which axis.
- One graph → another: use one figure to interpret another (e.g., identify which curve matches a described condition).
Quick method: map “what changes” to axes
- The variable you control or that progresses naturally (often time, concentration steps) usually belongs on the horizontal axis.
- The measured outcome belongs on the vertical axis.
Worked example (table → comparison statement)
If a table shows that Group A has higher values than Group B at every listed x:
- Correct translation: “For all measured values of x, Group A’s y exceeds Group B’s y.”
Exam Focus
- Why it matters: ACT Science often provides multiple visuals—questions test whether you can align them consistently.
- Typical question patterns:
- “Which graph best represents the data in Table __?”
- “Which statement is supported by Figure __?”
- “Which curve corresponds to condition ____?”
- Common mistakes:
- Swapping axes (treating y as x)—identify variables before interpreting.
- Confusing absolute value with rate of change—a higher value is not the same as a steeper slope.
Interpolation and Extrapolation
Interpolation is estimating between shown data points. Extrapolation is estimating beyond the shown range.
When ACT expects you to do this
- Interpolation is common when a value is asked at an x that lies between labeled points.
- Extrapolation is less common and typically explicitly requested—be cautious and conservative.
Linear interpolation (typical ACT approach)
If the graph segment is approximately straight, estimate proportionally.
Worked example (interpolation):
- Suppose y=20 at x=4 and y=30 at x=6.
- At x=5 (halfway), estimate y\approx 25.
Extrapolation caution
If the relationship is curved or leveling off, extending a straight-line assumption can be wrong.
- Only extrapolate if the question asks—and use the local trend near the edge of the data.
Exam Focus
- Why it matters: Interpolation is a quick way to get values not explicitly printed; extrapolation tests whether you respect what the data can (and can’t) support.
- Typical question patterns:
- “Approximately what is ____ at x = (between ticks)?”
- “Based on the trend, what would you expect at x = (just beyond range)?”
- Common mistakes:
- Treating a curve as linear—check whether equal x-steps give equal y-steps.
- Extrapolating far beyond the data—ACT rarely expects big leaps; if answers seem extreme, re-check.
Mathematical Reasoning with Data
Mathematical reasoning on ACT Science is usually arithmetic applied to values you read: differences, ratios, percent change, averages, and sometimes slope.
High-frequency calculations
- Difference (absolute change): \Delta y = y2 - y1
- Rate of change / slope (from a graph): \text{slope} = \frac{\Delta y}{\Delta x}
- Percent change: \%\text{ change} = \frac{\text{new} - \text{old}}{\text{old}}\times 100\%
- Ratio: \frac{A}{B} or “A is how many times B?”
- Mean (average): \bar{x} = \frac{\text{sum of values}}{\text{number of values}}
Worked example (percent change)
A value increases from 50 to 65.
- \%\text{ change} = \frac{65-50}{50}\times 100\% = 30\%
Worked example (slope comparison)
Two lines on the same axes:
- Line 1 rises 20 units over 4 units of x: slope =\frac{20}{4}=5.
- Line 2 rises 12 units over 4 units of x: slope =\frac{12}{4}=3.
- Line 1 has the greater rate of increase.
Exam Focus
- Why it matters: Many questions are “one-step math” embedded in science contexts—doing the right operation quickly prevents time loss.
- Typical question patterns:
- “What is the change in ____ between x1 and x2?”
- “Which condition has the greatest rate of increase?”
- “By what factor/percent does ____ change?”
- Common mistakes:
- Using the wrong baseline for percent change—“old” goes in the denominator.
- Mixing up “greatest value” with “greatest slope”—identify what’s being asked.
- Dropping units—keep them to avoid mismatched answers.
Comparing and Contrasting Data Sets
Comparing data sets means identifying similarities/differences across groups, trials, or conditions using the same variable(s).
What comparisons look like on ACT
- Same axes, multiple lines: compare at a fixed x, or compare slopes/peaks.
- Separate graphs: align variables/units before comparing.
- Multiple tables: compare corresponding entries (same row/column meaning).
Comparison language to master
- “Greater than / less than” at a specific x.
- “Increases faster” (steeper slope).
- “Reaches maximum earlier/later.”
- “Higher overall” vs “higher at the end” (endpoints can mislead).
Worked example (fixed-x comparison)
At x=10, Dataset A has y=40 and Dataset B has y=35.
- Correct conclusion: “At x=10, A exceeds B by 5 units.”
Worked example (overall vs local)
Dataset A starts lower but rises faster and overtakes B.
- You may be asked for the crossover point—approximately where the lines intersect.
Exam Focus
- Why it matters: Many ACT Science questions involve choosing between experimental conditions—comparison is central to interpreting results.
- Typical question patterns:
- “Which condition produces the highest ____ at x= …?”
- “During which interval is A greater than B?”
- “At approximately what x do A and B have equal values?”
- Common mistakes:
- Comparing values from graphs with different scales—verify axis ranges and units first.
- Assuming “higher at one point” means “higher everywhere”—check multiple points or the specified interval.
Quick Review Checklist
- Can you identify the independent and dependent variables from a table or graph?
- Can you read a value accurately when the axis ticks are spaced in non-1 increments?
- Can you use a legend to match the correct line/symbol to the correct condition?
- Can you describe whether a relationship is positive, negative, nonlinear, or shows a plateau/peak?
- Can you translate a table into a correct verbal statement comparing two conditions?
- Can you estimate a value by interpolation between two points on a line?
- Do you know when extrapolation is risky and should be used only if requested?
- Can you compute \Delta y, slope \frac{\Delta y}{\Delta x}, and percent change correctly from displayed data?
- Can you compare two datasets at a fixed x and also compare their rates of change?
Final Exam Pitfalls
- Reading the right graph but the wrong series — Always match the question’s condition to the legend before reading values.
- Ignoring units or prefixes — Treat the unit as part of the value; check axis labels and table headings every time.
- Misreading the scale — Confirm tick spacing and whether the axis starts at zero before estimating differences.
- Confusing “highest value” with “fastest increase” — Highest value is about y; fastest increase is about slope \frac{\Delta y}{\Delta x}.
- Over-extrapolating — Only project beyond the data when explicitly asked, and keep the estimate close to the trend near the edge.
═══════════════════════════════════════
Claude Opus 4.6
═══════════════════════════════════════
What You Need to Know
- Interpretation of Data makes up roughly 45–55% of the ACT Science section (approximately 18–22 of the 40 questions). It is the single largest reporting category — mastering it is the fastest way to raise your Science score.
- You do not need advanced science knowledge. These questions test whether you can accurately read, analyze, and draw conclusions from tables, graphs, and figures presented in the passages.
- The core skills are: locating specific data points, identifying trends, translating between different representations, estimating values through interpolation/extrapolation, performing simple calculations, and comparing multiple data sets.
- Speed matters — you have about 52 seconds per question. Practice reading visuals quickly and systematically so you don't waste time re-reading passages.
Reading and Interpreting Tables and Graphs
The most fundamental skill on the ACT Science section is accurately pulling information from data displays — tables, line graphs, bar graphs, scatter plots, and occasionally pie charts or diagrams.
Tables
- Always read the column and row headers first, including units.
- Identify the independent variable (what was changed) and the dependent variable(s) (what was measured).
- Scan for any footnotes, special symbols, or conditions listed above or below the table.
Graphs
- Read the title, axis labels, and units before looking at the data.
- Check whether axes use a linear or logarithmic scale — logarithmic scales compress large ranges and can mislead you about the size of changes.
- Note whether multiple data series are plotted (different colored lines, different symbols) and identify them using the legend/key.
Worked Example
A table shows temperature (°C) in column 1 and solubility of NaCl (g/100 mL) in column 2. A question asks: "At 40°C, what is the solubility of NaCl?" You simply locate 40°C in column 1 and read across to column 2 — no calculation needed.
Exam Focus
- Why it matters: Roughly one-third of all Interpretation of Data questions are straightforward "look-up" questions — free points if you read carefully.
- Typical question patterns:
- "According to Table 1, what was the value of X when Y equaled Z?"
- "Based on Figure 2, which trial produced the highest result?"
- "What are the units of the dependent variable in Experiment 3?"
- Common mistakes:
- Misreading units (e.g., confusing mL with L, or °C with K).
- Reading the wrong row or column because you rushed.
- Ignoring the legend and confusing two data series on the same graph.
Recognizing Trends and Patterns in Data
Beyond reading individual values, you must identify relationships between variables.
Key Relationship Types
| Relationship | Description | What It Looks Like on a Graph |
|---|---|---|
| Direct (positive) | As X increases, Y increases | Line slopes upward left to right |
| Inverse (negative) | As X increases, Y decreases | Line slopes downward left to right |
| No relationship | Changes in X don't affect Y | Flat/horizontal line or scattered points |
| Non-linear | The rate of change itself changes | Curved line (exponential, logarithmic, etc.) |
Tips for Identifying Trends
- Look at the overall direction — don't let one or two outlier points mislead you.
- Distinguish between a leveling off (plateau) and a true decrease.
- If the question asks about a trend "between Trial 2 and Trial 5," only examine that range — the trend outside that range is irrelevant.
Exam Focus
- Why it matters: Trend questions test conceptual understanding and appear in nearly every ACT Science passage.
- Typical question patterns:
- "As temperature increases, the rate of reaction ___."
- "Which of the following best describes the relationship between X and Y?"
- "Over the range of 20–60°C, does variable Z increase, decrease, or remain constant?"
- Common mistakes:
- Describing the overall trend when the question asks about a specific range.
- Confusing correlation with causation — the ACT usually just asks you to describe, not explain.
- Calling a curve "linear" when it's clearly leveling off or accelerating.
Translating Between Data Representations
Some questions ask you to move information from one format to another — for example, selecting a graph that matches a table, or vice versa.
Strategy
- Anchor on extreme values. Check the highest and lowest data points in the table and verify they appear at the correct positions on the graph.
- Check the shape. If the table shows values increasing at an accelerating rate, the correct graph should show an upward curve — not a straight line.
- Verify intermediate points. Spot-check one or two middle values to confirm.
Example
A table shows that as pressure doubles from 1 atm to 2 atm, volume decreases from 10 L to 5 L. The matching graph should show an inverse relationship — a curve that decreases and flattens (consistent with PV = nRT, or Boyle's Law behavior).
Exam Focus
- Why it matters: These questions test deeper comprehension — not just reading data but understanding its shape and behavior.
- Typical question patterns:
- "Which graph best represents the data in Table 2?"
- "If the data in Figure 1 were presented in a table, which table would be correct?"
- Common mistakes:
- Only checking one data point and selecting an answer that matches at that point but diverges elsewhere.
- Forgetting to check axis scales — two graphs can look similar but have different scales.
- Confusing direct and inverse curves when variables are swapped on the axes.
Interpolation and Extrapolation
Interpolation means estimating a value within the range of given data. Extrapolation means estimating a value beyond the range of given data by extending the observed trend.
How to Do It
- For interpolation: find the two nearest known data points that bracket the value, and estimate proportionally between them.
- For extrapolation: identify the trend (linear, exponential, etc.) and extend it logically.
Worked Example
Given data: at t = 10 s, distance = 20 m; at t = 20 s, distance = 40 m.
- Interpolation (t = 15 s): The midpoint gives \frac{20 + 40}{2} = 30 m.
- Extrapolation (t = 30 s): The rate is \frac{40 - 20}{20 - 10} = 2 m/s, so at t = 30 s, distance \approx 40 + 2(10) = 60 m.
Memory Aid: Inter = "between" (like an interstate runs between states). Extra = "beyond" (like an extraterrestrial is beyond Earth).
Exam Focus
- Why it matters: These questions are extremely common and test whether you can go beyond the explicitly stated data.
- Typical question patterns:
- "Based on Figure 1, at a temperature of 35°C the solubility would most likely be closest to ___."
- "If the experiment were extended to 100 seconds, the expected value would be approximately ___."
- Common mistakes:
- Assuming a linear trend when the data is clearly curved — always look at the shape before extrapolating.
- Extrapolating far beyond the data range with false confidence; the ACT typically only asks for modest extrapolation.
- Choosing an answer that matches an exact data point instead of estimating between points.
Mathematical Reasoning with Data
Some Interpretation of Data questions require simple calculations — you won't need a calculator (it's not allowed on ACT Science), but you need to handle basic arithmetic, ratios, percentages, and averages.
Common Calculations
- Averages: \bar{x} = \frac{\sum x_i}{n}
- Percent change: \% \text{ change} = \frac{\text{new} - \text{old}}{\text{old}} \times 100
- Ratios and proportions: comparing two values or determining how one quantity scales relative to another.
- Differences: finding the change between two data points.
Example
If Trial 1 produces 50 mL of gas and Trial 2 produces 75 mL, the percent increase is:
\frac{75 - 50}{50} \times 100 = 50\%
Exam Focus
- Why it matters: About 5–8 questions per test require some computation; getting these right separates mid-range scores from high scores.
- Typical question patterns:
- "What is the average value of X across all trials?"
- "The value in Experiment 2 is approximately what percent of the value in Experiment 1?"
- "By how much did Y change between Trial 3 and Trial 5?"
- Common mistakes:
- Using the wrong denominator in percent change (always divide by the original value).
- Arithmetic errors under time pressure — estimate first, then verify.
- Forgetting to account for units when comparing values from different parts of the passage.
Comparing and Contrasting Data Sets
Many ACT Science passages present results from multiple experiments, groups, or conditions. You must identify similarities, differences, and relationships across data sets.
Strategy
- Identify what changed between the data sets (different concentrations, different organisms, different temperatures, etc.).
- Hold other variables constant mentally — compare data points where only one variable differs.
- Look for consistent patterns — does the same trend appear in both data sets, or do they diverge?
Example
Two graphs show population growth for Species A and Species B under identical conditions. Species A levels off at 500 individuals; Species B levels off at 800. You can conclude Species B has a higher carrying capacity under those conditions — but the ACT might simply ask which species reached a higher maximum.
Exam Focus
- Why it matters: Cross-data-set questions test higher-order reasoning and frequently appear as the harder questions at the end of a passage.
- Typical question patterns:
- "Compared to Experiment 1, the results of Experiment 2 show ___."
- "Which student's model is most consistent with the data in both Table 1 and Figure 2?"
- "At what value of X do the results of Group A and Group B most closely agree?"
- Common mistakes:
- Comparing data points measured under different conditions without realizing they aren't directly comparable.
- Mixing up which data set corresponds to which experiment — always re-check labels.
- Overlooking that two data sets may agree at some points but disagree at others; avoid over-generalizing.
Quick Review Checklist
- ☐ Can you identify the independent and dependent variables in a table or graph?
- ☐ Can you read axis labels, units, and legends accurately and quickly?
- ☐ Do you know how to describe the relationship between two variables (direct, inverse, no relationship, non-linear)?
- ☐ Can you translate data from a table into the correct graph shape, and vice versa?
- ☐ Can you interpolate a value between two known data points?
- ☐ Can you extrapolate a value by extending an observed trend?
- ☐ Do you know how to calculate a percent change, average, or simple ratio from data?
- ☐ Can you compare two data sets and identify where they agree or differ?
- ☐ Can you distinguish between what the data actually shows and what it merely suggests?
- ☐ Can you complete all of this within roughly 52 seconds per question?
Final Exam Pitfalls
Rushing past labels and units. Many wrong answers are designed to trap students who misread an axis label or confuse units (e.g., seconds vs. minutes, grams vs. kilograms). Always spend 5 seconds reading the headers before touching the data.
Assuming linearity. When interpolating or extrapolating, students often assume data follows a straight line. If the graph is curved, your estimate must account for the curve — pick the answer that reflects the bend, not the midpoint.
Using the wrong denominator for percent change. The formula is \frac{\text{new} - \text{old}}{\text{old}} \times 100. Students frequently divide by the new value or the larger value. Always divide by the starting value.
Mixing up data series. When a graph has multiple lines or a passage has multiple tables, it's easy to accidentally pull a number from the wrong series. Re-check which line or table corresponds to the question.
Over-interpreting the data. The ACT rewards answers that stay close to what the data literally shows. If a question asks what the data supports, pick the most conservative, directly supported conclusion — not a broad generalization or causal claim.
Ignoring the specified range. When a question says "between 20°C and 60°C," only describe the trend in that interval. The data may behave differently outside that range, and choosing an answer based on the full data set is a common trap.