Measures of Dispersion for Grouped Data

Interpolation

Reading a value between plotted points, for example finding the median from an ogive.

EnglishInterpolation
Bahasa MelayuInterpolasi
中文内插

How it is used

For n = 30, the median is at position 15, which falls in the class 21–30 (lower boundary 20.5, cumulative frequency 10 before it, class frequency 12, width 10). By interpolation the median = 20.5 + [(15 − 10) ÷ 12] × 10 = 24.67.

Where it shows up in SPM

In grouped-data statistics, Form 5. Interpolation gives an exact-looking value for the median or a quartile from the cumulative frequency table without drawing a graph, using the formula L + [(n/2 − F) ÷ f] × c.

It is the algebraic alternative to reading these off an ogive in Paper 2.

Don't confuse it with

Reading from an ogiveInterpolation calculates the value with a formula, while reading from an ogive estimates it by eye off a drawn curve.
ExtrapolationInterpolation finds a value between known data points; extrapolation would estimate beyond the range of the data.

Open the chapter: Measures of Dispersion for Grouped Data →

Frequently asked questions

What does each letter in the interpolation formula mean?

In L + [(n/2 − F) ÷ f] × c, L is the lower boundary of the median class, n is the total frequency, F is the cumulative frequency before that class, f is the class's frequency and c is the class width. For a quartile, replace n/2 with n/4 or 3n/4.

How do I find which class the median is in?

Work out n/2, then look down the cumulative frequency column for the first class whose running total reaches or passes n/2. That is the median class, whose lower boundary you use as L in the formula.

Related terms

One-hour paid trial · Same-day replyfrom RM50/hr
Book a Trial Class