From a Scatterplot to Approximate Coordinates
In “Adding a Categorical Variable to a Scatterplot,” you learned to read a plot’s axes and symbols before describing its pattern. Now you will use those axes to estimate the coordinates of individual points. This is useful when a question asks about a particular observation or asks what response value appears at a specified explanatory-variable value.
Every point represents one observational unit and its paired values. The horizontal location gives the explanatory-variable value, \(x\), and the vertical location gives the response-variable value, \(y\). A point’s coordinates are written \((x,y)\), with the \(x\)-value first. When you read coordinates from a graph rather than from the original data, your values are estimates: the dot may sit between labeled tick marks, and the graph may not show enough detail to recover the exact recorded values.
A reliable reading depends on the scale, not just on where a point appears on the page. A point halfway between the marks for 10 and 20 has an estimated value of 15, even if the graph is displayed at a different size. The same visual distance can represent different amounts on different axes, so read the labels and intervals on each axis separately.
A Method for Estimating Coordinates
Start by checking what each axis measures, including its units. Then locate the point relative to the labeled tick marks. If there are grid lines or smaller tick marks, use them as guides. If not, estimate the point’s position within the interval between labeled marks. Report a level of precision the graph supports; do not write many decimal places when the plot only allows a rough estimate.
Identify the explanatory variable on the horizontal axis and the response variable on the vertical axis.
Use the horizontal tick marks to judge where the point lies, interpolating between labeled values when needed.
Use the vertical scale independently. Do not assume that equal-looking distances on the two axes represent equal numerical amounts.
Give the ordered pair in the correct order, include units where helpful, and identify it as approximate if it was read from the graph.
Interpolation here means estimating a value between labeled tick marks by using the spacing of those marks. It is a way to read the location of a dot, not a way to claim that the plot contains an observation at every possible \(x\)-value. Keep that distinction in mind when a question gives an \(x\)-value and asks for a corresponding \(y\)-value.
Estimate a Point’s Coordinates
Worked Example: Dissolved Oxygen at a Stream Site
Imagine a fictional scatterplot of water temperature and dissolved oxygen at several stream sites. Temperature, in degrees Celsius, is on the horizontal axis, labeled at 0, 2, 4, 6, 8, and 10. Dissolved oxygen, in milligrams per liter, is on the vertical axis, labeled at 4, 6, 8, 10, and 12. One dot is directly above \(x=6\) and halfway between the vertical marks for 6 and 8.
Estimate the coordinates. The dot’s horizontal coordinate is about 6 degrees Celsius. Halfway between 6 and 8 on the vertical axis is about 7 milligrams per liter. The estimated ordered pair is therefore \((6,7)\), where the first value is temperature and the second is dissolved oxygen.
Interpret in context. The graph suggests that one sampled site had a water temperature of about 6°C and dissolved oxygen of about 7 mg/L. Because these values are read from a plotted dot, “about” is appropriate; the graph does not establish whether the original recorded oxygen value was exactly 7.
Check the reading. Reading the vertical position as 6 would place the dot on the lower tick mark, and reading it as 8 would place it on the upper one. Since the dot is halfway between those marks, 7 is the reasonable estimate. The coordinates must be reported \(x\) first, then \(y\), not in the reverse order.
In that example, the \(x\)-value fell exactly on a labeled tick, but the \(y\)-value did not. A point can be easy to read on one axis and less precise on the other. Estimate each coordinate separately; do not let a clear reading on one axis make the other seem more precise than it is.
Read a Response Value at a Given \(x\)
When asked for the approximate \(y\)-value at a given \(x\)-value, find that location on the horizontal axis and trace vertically to the relevant plotted point. Then trace horizontally from the point toward the vertical axis to estimate \(y\). This two-direction method helps keep the axes straight: vertical movement locates the explanatory-variable value, and horizontal movement helps read the response-variable value.
If the graph is group-coded, use the legend to identify which point or points the question concerns. As in “Adding a Categorical Variable to a Scatterplot,” symbols represent categories; they do not change what the axes mean. If the question asks about one group, do not accidentally read a nearby point from another group.
Worked Example: Reading a Garden Plant’s Height
A fictional scatterplot shows the number of weeks after planting on the horizontal axis and plant height in centimeters on the vertical axis. The horizontal axis is labeled every 5 weeks from 0 to 20. The vertical axis is labeled every 10 centimeters from 0 to 50, with smaller marks every 2 centimeters. A plotted point lies at \(x=12\) weeks and at the second small mark above 30 centimeters.
Locate the given \(x\)-value. Twelve is between the labeled horizontal-axis values 10 and 15. The dot’s horizontal position is a little less than halfway from 10 to 15, consistent with about 12 weeks.
Read the \(y\)-value. Each small vertical-axis interval represents 2 centimeters. The second small mark above 30 is \(30+2+2=34\) centimeters. Thus the plotted point is approximately \((12,34)\).
Answer the question in context. At 12 weeks after planting, the plant represented by this point is about 34 cm tall. The response is an estimate from the graph; it should not be reported as an exact measurement unless the underlying data provide that value.
Why the axis labels matter. The point is not at 12 centimeters simply because its \(x\)-coordinate is 12, and the small marks on the vertical axis are not 1 centimeter apart. Reading the labeled vertical scale first shows that each small interval is 2 centimeters.
When the Graph Does Not Show One Unique Value
A scatterplot displays only the observations that were recorded. A given \(x\)-value may have no plotted point, one point, or several points. If several observations share the same explanatory-variable value, they can have different response values. Therefore, a scatterplot does not always provide one unique \(y\)-value for every \(x\).
If there is no point at the requested \(x\), do not describe a value from a nearby dot as though it were an observed response at that \(x\). If a question explicitly asks for a visual estimate of the pattern at an \(x\)-value between observations, label the result as an estimate of the pattern, not a recorded data value. This reading task alone does not supply a rule for calculating a missing response.
Worked Example: Two Scores at the Same Practice Time
A fictional group-coded scatterplot shows weekly practice time in hours on the horizontal axis and a performance score in points on the vertical axis. The horizontal axis is marked at 0, 2, 4, 6, and 8 hours. The vertical axis is marked at 40, 50, 60, 70, and 80 points. At \(x=4\) hours, the graph shows two dots: one near 60 points and another near 70 points. The legend identifies the dots as two different students.
Read both points. The first student’s approximate coordinates are \((4,60)\); the second student’s are \((4,70)\). Both points have the same horizontal coordinate, but their vertical coordinates differ.
Answer the question precisely. If asked for the score at 4 hours, say that the plot shows two students at about 4 hours, with scores near 60 and 70 points. Saying “the score is 65” would not describe either plotted observation; it would create a single value that the graph does not show.
Use the group coding. If the question asks for the student represented by a particular symbol, check the legend and read that point’s height. The two dots are not interchangeable simply because they share an \(x\)-value.
Conclusion in context. In this invented plot, the same amount of weekly practice is associated with different scores for the two students. The graph lets us read both approximate pairs, but it does not explain why the scores differ.
The same caution applies when a requested \(x\)-value falls between dots. The display may help you describe the general pattern, but a point’s coordinate is an observed pair, not an entire vertical line of possible responses. This is consistent with the earlier distinction in “Spotting Outliers in Bivariate Data” between a point and the overall pattern formed by many points.
Precision, Rounding, and Clear Communication
A graph’s scale determines how much precision is reasonable. If the vertical axis is labeled in tens and has no smaller divisions, a response such as “about 47.3” usually claims more than the graph can support. A value such as “about 50” may be more defensible. If smaller tick marks are present, use them, but still recognize that a dot has thickness and may not align perfectly with a mark.
For a point between two labeled values, use the spacing as a guide. If it appears roughly one-quarter of the way from 20 to 40, an estimate near 25 is sensible. If it appears about three-quarters of the way across, an estimate near 35 may be sensible. The aim is not to manufacture precision; it is to give a reasonable reading at the level the display allows.
Common Mistakes and AP Exam Tips
- Reversing the ordered pair: Coordinates are written \((x,y)\), so give the horizontal explanatory-variable value first and the vertical response-variable value second.
- Ignoring the axis scale: A point’s visual position does not tell you its value without the tick labels and intervals. Read the scale on each axis rather than guessing from the page layout.
- Using the wrong units or variable: Name what each coordinate measures. For example, distinguish 12 weeks from 12 centimeters.
- Reporting unjustified precision: Do not give extra decimal places when the tick marks support only a rough estimate. Use “about” or “approximately” when reading a plotted location.
- Assuming one \(y\)-value for every \(x\): If multiple points share the same \(x\), report the separate approximate \(y\)-values. If no point is plotted at that \(x\), do not claim a nearby observation occurred there.
- Overlooking group symbols: Check the legend if the plot uses different colors or symbols. A point from another category may have a different \(y\)-value at a similar \(x\).
- Confusing a visual estimate with an exact datum: A graph reading is approximate unless the plot or accompanying data give exact values. Make that limitation clear in the answer.
A full-credit response to a coordinate-reading question gives the approximate values in the correct order, identifies the variables and units, and addresses ambiguity when the graph shows more than one point. For instance: “At about 12 weeks, the plotted plant is approximately 34 cm tall.” If two points appear at the requested \(x\), name both values rather than forcing them into one answer.
Check Your Understanding
Use the axis scales and the plotted points described in each question.
- A point lies at \(x=8\) on an axis labeled every 2 units and halfway between \(y=20\) and \(y=30\). What are its approximate coordinates?
- A plot’s vertical axis is marked at 10, 20, 30, and 40, with small marks every 2 units. A point lies three small intervals above 20. Estimate its \(y\)-value.
- At \(x=5\) hours, a scatterplot shows three points with scores near 42, 48, and 55. What should you report if asked for the scores at 5 hours?
- A question asks for the observed response at \(x=7\), but no point is plotted there. What should you avoid claiming?
- Why is “about 32.847” usually not an appropriate coordinate estimate when the graph has axis marks only every 10 units?