Part 3 — GED ® Science · Unit 1 — Read the Evidence
Chapter 3 — Tables, Graphs, and Models
GED® test targets: SP.1.c, SP.3.c, SP.6.a, SP.6.b, SP.6.c (Science Practices chapter)
A fictional maintenance log records how much water remains in a storage tank. One person reads the highest number and says the system is fine. Another sees a downward line and says the whole damn place is running dry. Neither has yet read the time interval, the units, or whether water was being added. A display gives evidence only after you identify exactly what its marks represent.
A graph is an argument made with position, size, or shape. Before believing the picture, read what the picture actually measures. A dramatic line can show a tiny change if the vertical axis starts just below the lowest value. The graph may be technically accurate while making the difference look huge as hell.
Start with the labels
Read the title, both axes, units, scale intervals, and legend. The horizontal axis is commonly called x; the vertical axis is y. In an experiment, the independent variable often appears on x and the dependent variable on y, but labels outrank that convention. Some graphs use time, categories, or two observed variables instead.
| Time after cooling began (min) | Container A (°C) | Container B (°C) |
|---|---|---|
| 0 | 60 | 60 |
| 5 | 54 | 49 |
| 10 | 49 | 41 |
| 15 | 45 | 35 |
| 20 | 42 | 31 |

Constructed cooling data. Both samples cool; A remains warmer at each measured time after the start.
The table gives exact recorded values. The line graph makes the cooling pattern easier to see. At ten minutes, A is 49°C and B is 41°C, a difference of 8°C. That is a comparison at one time. A’s total decrease over 20 minutes is 18°C; B’s is 29°C. That is a comparison of changes. Those answers describe different things.
Match the display to the question
A bar graph compares categories, such as the average growth under three light conditions. A line graph emphasizes change across an ordered variable such as time. A scatter plot shows paired values for two numerical variables, such as water temperature and oxygen concentration. A circle graph shows parts of a whole; the represented parts should total the whole.
A histogram groups numerical observations into intervals. It can show whether most measurements cluster together or spread widely. Unlike separate category bars, adjacent histogram intervals represent a continuous numerical scale. A diagram represents structures or relationships, such as a food web. Its arrows need a key: arrows can mean energy transfer, movement, cause, or sequence.
Suppose a table lists rainfall of 10, 20, and 30 mm in April, May, and June. To represent it with bars, use equal-width bars labeled by month and heights at 10, 20, and 30 on an axis labeled rainfall in mm. A chart with heights 1, 2, and 3 but no scale hides the units. A chart reversing May and June contradicts the table.
Describe a trend without inventing a cause
A positive association means larger values of one variable tend to occur with larger values of the other. A negative association means one tends to increase as the other decreases. No clear association means the points do not show an obvious pattern. Association alone does not prove causation.
Ice-cream sales and heat-related illness can both rise in hot weather. That does not show that ice cream causes the illness. Temperature is a plausible common influence. To establish a causal explanation, examine the design, timing, alternative causes, and supporting evidence.
An outlier is a value far from the general pattern. It might reflect an error or a real unusual event. Investigate it; do not erase it merely because it makes your favorite conclusion inconvenient. Honest graphs sometimes look messy because the world did not read your tidy little plan.
Predict carefully
Interpolation estimates between measured points. For container A at 12.5 minutes, halfway between 10 and 15, a straight-line estimate is halfway between 49 and 45: about 47°C. This assumes the change between those points is approximately linear. It is an estimate, not a recorded measurement.
Extrapolation extends beyond observed values. Extending the initial cooling rate forever could predict temperatures below the surrounding room, even though an ordinary cooling container approaches room temperature. The model stops making sense when its assumptions stop fitting the system.
The slope of a line is change in y divided by change in x. Between zero and five minutes, A’s slope is (54 − 60) ÷ 5 = −1.2°C per minute. Between 15 and 20 minutes, it is (42 − 45) ÷ 5 = −0.6°C per minute. Cooling slowed. A falling line does not necessarily fall at a constant rate.
Worked evidence question
Question: Which conclusion is supported by the cooling data? A. A cooled more slowly overall than B. B. Neither container transferred energy. C. Both cooled exactly 1°C every minute. D. A will remain above 40°C forever.
Answer: A. Over the same 20 minutes, A decreased 18°C and B decreased 29°C. B conflicts with the temperature changes in this setting. C conflicts with the unequal changes between intervals. D extends far beyond the observations.
Two studies can report different cooling rates without either being dishonest. They may use different container sizes, starting temperatures, or room temperatures. Compare conditions and measurement methods before declaring a contradiction. A model that fits one set of conditions may need adjustment for another.
What carries forward: translate the display into a sentence that includes the variables, direction, amount when available, and limits. “A cooled 18°C in 20 minutes under these conditions” beats “A is better” by a mile.
Read the Relationship, Not the Dramatic-Ass Picture
Each plotted point on a line graph is a pair: a horizontal value and a vertical value. If the horizontal axis is elapsed hours and the vertical axis is liters remaining, a point at (2, 300) means 300 liters remained after two hours. It does not mean the tank lost 300 liters per hour. The height of a point and the slope between points answer different questions.
For a worked comparison, suppose the tank falls from 500 L at hour 0 to 300 L at hour 2. Change in volume = 300 − 500 = −200 L. Elapsed time = 2 − 0 = 2 hours. Average rate = −200 ÷ 2 = −100 L/hour. The minus sign identifies a decrease. Calling the rate “100 liters” drops the time unit and changes the meaning. That little slash in L/hour is doing a hell of a lot of work.
A straight segment between recorded points is a model of the interval, not proof of every instant. The volume might have fallen steadily, or most of the loss might have occurred in one hour. A table with only endpoints cannot settle that. Likewise, two lines crossing means equal measured values at that location; it does not automatically mean equal rates, equal causes, or equal performance over the whole study.
Uncertainty bars, if supplied, show a stated range of variation or uncertainty around an estimate. Read the caption because bars can represent different quantities. A tiny difference between two means may be weak evidence when the measurements vary widely. Do not invent a formal significance test the passage never provides. Explain the visible difference and the limits of the display.
On the GED® test, the same skill may appear in a graph of animal abundance or chemical concentration. State the relationship using both variables: as time increased, concentration decreased over the measured interval. Then separate that pattern from its explanation. A line can show what happened; it cannot manufacture a missing control group.
Your Turn — Guided and Independent Practice — Practice Check S3
Guided: A table shows water use of 12, 15, 18, and 21 liters across four days. Find the total, mean, and change from day 1 to day 4. Then state the trend without inventing a cause.
Guided answer: Total = 66 L. Mean = 66 ÷ 4 = 16.5 L. Change = 21 − 12 = 9 L. The values increased across the four recorded days. The table alone does not explain why.
Inside guided: Two identical cups of hot water start at 80°C. One is wrapped in a towel. Readings after 10, 20, and 30 minutes: wrapped 74, 69, 65°C; bare 70, 62, 56°C. Which cup cools faster? Predict the bare cup at 40 minutes.
Answer: The bare cup cools faster: it dropped 24°C in 30 minutes, while the wrapped cup dropped 15°C. Its drops were 10, 8, and 6°C, so cooling is slowing as the water gets closer to room temperature. A reasonable estimate at 40 minutes is about 51–52°C. Call it an estimate; the trend is not a law. The towel didn’t make heat; it just slowed the leak.
Independent: A graph’s vertical axis begins at 98 instead of 0 and shows values of 99 and 101. What should a careful reader conclude? A. The difference is 102%. B. The visual gap may exaggerate a small numerical difference. C. The graph is automatically false. D. Axis labels do not matter.
Independent answer: B. The difference is only 2 units, although the cropped scale can make it look dramatic. C is too strong; the graph may be accurate but visually misleading. A miscalculates, and D is nonsense because labels and scale tell you what the marks mean. Common mistake: trusting the picture before reading the damn axis.