Data Representation

Learning Objectives

• Extract specific numerical values from tables and graphs within 10 seconds
• Read and correctly interpret axis labels, units, and scale conventions
• Identify trends in data: increasing, decreasing, constant, linear, nonlinear
• Distinguish between direct relationships (both variables increase together) and inverse relationships (one increases as the other decreases)
• Interpolate values at points between plotted data points
• Extrapolate values beyond the range of plotted data using the established trend
• Compare data sets across conditions, trials, or experimental groups
• Interpret error bars as a measure of data variability or uncertainty
• Calculate percent change from tabular or graphical data
• Identify outliers and assess their effect on trends or conclusions

High-Yield Concepts

Reading Tables and Line Graphs

Tables and line graphs are the two most common data displays on ACT Science and the two most frequently tested.

For tables: identify the column headers (what is being measured) and the unit labels before reading any data values. The first column usually contains the independent variable (what was controlled or varied). Subsequent columns contain dependent variable measurements. To answer "what was the value of Y when X = ___?", trace the row for the given X value to the column for Y.

For line graphs: identify the x-axis variable (independent), y-axis variable (dependent), units for each, and whether multiple lines are present. A line sloping upward indicates a positive/direct relationship; downward indicates a negative/inverse relationship. A flat line indicates no relationship. For multiple-line graphs, identify which line corresponds to which experimental condition using the legend.

Trends, Interpolation, and Extrapolation

Trend identification is tested in approximately 30-40% of Data Representation questions. The ACT tests five trend types: linear increase, linear decrease, nonlinear increase (accelerating), nonlinear decrease (decelerating), and constant (no change). Recognizing these patterns without precise calculation is usually sufficient -- the question asks "as X increases, Y..." not "at what rate does Y change?"

Interpolation reads a value between two explicitly plotted data points. The assumption is that the trend continues smoothly between plotted points. Interpolated values are considered reliable on the ACT.

Extrapolation reads a value beyond the plotted range by extending the trend. Extrapolated values are less reliable (the trend may change outside the measured range), but the ACT tests extrapolation as a data reading skill regardless.

Comparing Data Sets and Error Bars

Comparison questions ask students to identify which condition, group, or trial produced the highest, lowest, or most changed value. Strategy: read the question to identify exactly what is being compared and under what specific conditions, then locate the relevant rows/lines/bars and compare them directly.

Error bars on graphs represent the range of uncertainty or variability in measurements. A larger error bar means more variability; a smaller error bar means more precision. When two data points' error bars overlap significantly, the difference between them may not be statistically meaningful. The ACT tests whether students can read error bar magnitude and interpret overlapping vs. non-overlapping error bars.

Study Strategy

Begin with reading tables efficiently. Practice locating values at specific row/column intersections and identifying headers and units in under 5 seconds. This foundation enables every other data representation skill.

Next, practice reading line graphs: identifying trend direction, describing the relationship between variables, and reading specific values at given x-coordinates. Practice both interpolation (reading between points) and extrapolation (extending the line beyond plotted points).

Then study multiple data series and comparison questions. Practice reading legends and comparing behavior across different conditions or experimental groups -- which line rises fastest? which condition produced the highest value at a specific x-value?

Study percent change calculations, error bars, and outlier identification as a final cluster. These appear in medium-to-high difficulty questions and require slightly more analytical reasoning beyond simple data reading.

The strategy topics (act-data-representation-strategy, act-science-elimination-strategy, science-passage-mapping) are best practiced as meta-skills during full-passage practice sets rather than in isolation.

Common Mistakes

• Reading the y-value for the wrong x-value by tracing the wrong row or line
• Confusing the independent variable (x-axis, what was controlled) with the dependent variable (y-axis, what was measured)
• Ignoring units, which can cause percent change and comparison errors
• Extrapolating by assuming the trend reverses rather than continuing as shown
• Confusing multiple lines on a graph by not consulting the legend carefully
• Misreading inverted y-axis scales (sometimes the y-axis is plotted with largest values at the bottom)
• Treating error bars as the range of the data values rather than as measurement uncertainty
• Reading percent change as (new - old) / new instead of (new - old) / old
• Answering with the wrong data series on multi-series graphs
• Spending too long on complex tables when simpler scanning would answer the question

Exam Tips

• Read axis labels and units before looking at any data values
• For "as X increases, Y..." questions, trace the general direction of the line rather than calculating exact values
• For comparison questions, identify the specific condition and specific measurement the question asks about before scanning the data
• For interpolation, trust the trend shown -- do not second-guess the data display
• For percent change: (new - old) / old x 100; the original value is always the denominator
• Error bar questions are typically asking whether two conditions are clearly different or possibly the same -- overlapping error bars mean the difference may not be real
• Outlier questions usually ask what the trend would be "without the outlier" -- remove it mentally and re-examine the pattern
• On multi-series graphs, re-read the legend before each question that involves distinguishing between series
• Budget about 5-6 minutes per Data Representation passage (plus reading time) -- these should be among your fastest passages since they are primarily data look-up