Interpreting Data and Graphs

Standardized test questions rarely ask you to draw a graph; instead, they hand you one and ask what it means. A single line graph, bar chart, or box plot might anchor three or four separate questions, each requiring a different kind of reasoning: reading a value directly, computing a rate, comparing categories, or judging whether a stated conclusion actually follows from the data. This lesson focuses on that applied interpretation skill rather than on constructing visuals, building on the broader survey of data analysis and visual reporting.

Applied data questions tend to fall into a few recurring categories. Some ask you to retrieve a value directly from the display, such as the height of a bar or the position of a data point. Others require a calculation performed on values you first have to locate, such as a difference, a rate, or a percent change. A third category asks you to evaluate a claim, deciding whether the graph actually supports a stated conclusion or whether it is being used to suggest something the data does not establish. Recognizing which category a question belongs to before you start calculating saves time and prevents you from answering a question that was not asked.

A line graph plots a quantity against time or another continuous variable, and the slope between two points on that line represents a rate. If a graph shows a value \(y_1\) at \(x_1\) and a value \(y_2\) at \(x_2\), the average rate of change between those points is \(\dfrac{y_2 - y_1}{x_2 - x_1}\). Applied questions often disguise this as a word problem: "By how much did membership increase per month between month 2 and month 8?" The arithmetic is simple once you have correctly located the two points on the axis.

Reading this graph, membership starts near 150 and rises by roughly 12 per month. A test question might ask for the rate between month 0 and month 12, which you calculate as \(\dfrac{294 - 150}{12 - 0} \approx 12\) members per month, then use that rate to project a value beyond the graph. Projections beyond the plotted range should be treated cautiously, since extrapolation assumes the trend continues unchanged, an assumption the graph itself cannot confirm.

Bar charts compare discrete categories rather than showing continuous change, so questions built around them typically ask for a difference, a ratio, or a percent change between two bars. Suppose a survey of favorite study formats among four groups produces the results below.

A direct-retrieval question might ask which format was preferred by the fewest students, answered from the chart alone (Reading, at 15). A calculation question might ask for the percent by which Flashcards exceeded Video: \(\dfrac{60 - 40}{40} \times 100 = 50\) percent. Notice that both bars must be read correctly before either arithmetic step begins; a misread value propagates through the rest of the solution.

Many applied questions embed a statistical summary, such as a box plot or a stated mean and standard deviation, inside a longer passage. A box plot displays the minimum, the first quartile \(Q_1\), the median, the third quartile \(Q_3\), and the maximum, and the interquartile range \(Q_3 - Q_1\) tells you how spread out the middle half of the data is. A question might describe two box plots representing test scores from different classes and ask which class had less variability, a judgment made by comparing the widths of the boxes rather than by computing anything new. When a mean and standard deviation are given instead, a value more than about two standard deviations from the mean is typically flagged as an outlier, and applied questions often hinge on identifying that a single unusual data point is inflating or depressing a summary statistic.

Three traps recur across standardized test data questions. A truncated y-axis, one that does not begin at zero, can make a small difference between bars look dramatic, so always check the axis before judging magnitude. A scatter plot showing two variables moving together tempts a causal claim that the data cannot support; correlation only establishes that the variables change together, not that one causes the other. Finally, extrapolating a trend line far beyond the plotted data treats an assumption as a certainty, and a well-constructed question will offer a plausible-looking wrong answer built exactly on that overreach. Practicing with statistical summaries in more depth, including how they are calculated and reported, is covered in statistical data analysis and reporting.

Rate of Change: The amount by which a quantity increases or decreases per unit of the variable it is measured against, found from two points on a graph.

Percent Change: The size of a change between two values expressed as a percentage of the original value.

Extrapolation: Estimating a value beyond the range of data actually plotted, based on the assumption that an existing trend continues.

Interquartile Range: The spread of the middle half of a dataset, calculated as the third quartile minus the first quartile.

Outlier: A data point that falls well outside the typical range of values in a dataset and can distort summary statistics.

Truncated Axis: A graph axis that does not begin at zero, which can visually exaggerate the size of differences between values.

Correlation: A pattern in which two variables tend to change together, without necessarily implying that one causes the other.

Applied practice for this skill involves working through short passages paired with a single graph or chart and answering a sequence of questions of increasing difficulty, from direct retrieval to evaluative reasoning. Timing matters as much as accuracy: the goal is to locate the exact values a question requires without re-reading the entire display for each new question.

Related lessons

• Data Analysis and Visual Reporting