Estimating from Data, Graphs, and Tables

The three-step method for estimating

Whenever you need a value that sits between two points you can actually see, follow this routine:

• 1. Find the two points it sits between. Locate the known value just below and just above what you want.
• 2. Picture the line connecting them. Assume the data changes smoothly and steadily from one point to the next, then read off the value where your target lands.
• 3. Round to a sensible level of precision. Match the detail of the data. If readings are whole numbers, an estimate to the nearest whole number (or half) is honest; claiming three decimal places is not.

Worked example: reading between points on a line graph

Here is the reasoning, step by step:

• Find the surrounding points: week 5 sits between week 4 (16 cm) and week 6 (22 cm).
• Read along the line: week 5 is exactly halfway between weeks 4 and 6, so the height is about halfway between 16 and 22. Halfway is 19.
• Round sensibly: the recorded heights are whole centimetres, so "about 19 cm" is an honest estimate, not 19.0 or 19.37.

Estimating from a table or spreadsheet

The same logic works with a table of numbers, you just do the "reading along the line" in your head. Suppose a spreadsheet records the temperature of a cooling cup of water:

Question: about what temperature was the water at 12 minutes? There is no row for 12 minutes, so we estimate between the 10-minute row (60°C) and the 15-minute row (52°C).

• Find the rows: 12 minutes is between 10 min (60°C) and 15 min (52°C).
• Judge the position: 12 is two-fifths of the way from 10 to 15. Across that gap the temperature drops 8°C (from 60 to 52). Two-fifths of 8 is about 3, so the water has cooled roughly 3°C past 60.
• Estimate and round: 60 − 3 = 57. So about 57°C.

A trap to avoid: over-claiming precision

When you estimate, your answer should sound like an estimate. Reporting too many decimal places makes it look like you measured something you only guessed at. Compare these two ways of answering the cooling-water question:

Rule of thumb: round your estimate to roughly the same precision as the data you started from. Match the data, and add a word like "about" or "approximately" so your reader knows it is an estimate.

Inside the data is safer than outside it

Estimating between known points (interpolation) is fairly reliable, because you are surrounded by real measurements. Estimating beyond the last point (extrapolation) is much riskier, the pattern might not continue.

In the plant graph, guessing the height at week 5 is safe. But guessing the height at week 20 would be a stretch: the plant may stop growing, so the line might flatten out. If a question asks you to estimate far past the last data point, be cautious and say so. On the CAEC, recognizing the limits of an estimate is itself a scientific-inquiry skill.

Tips that make estimating quick and accurate

• Anchor to the two nearest points first. Your estimate must land between them. If it does not, recheck your reading.
• Use "halfway" and "a third of the way" thinking. You rarely need exact fractions, rough proportions are usually enough.
• Sanity-check the direction. If the data is rising, your estimate should be higher than the lower point; if it is falling, lower. Make sure the trend matches.
• Round to match the data. Whole-number readings deserve a whole-number (or half-unit) estimate, with a word like "about" attached.

Your turn: practice problems

Use the bean-plant graph and the cooling-water table from above. Estimate first, then round sensibly. Try each before you peek.

1. From the plant graph, about how tall was the plant at week 3?
2. From the cooling-water table, about what temperature was the water at 7 minutes?
3. Using the plant graph, would estimating the height at week 12 be more or less reliable than at week 3, and why?