Basic Mathematics

Lecture 12
Statistics: Measures of central tendency, and variation (Grouped data)
Lecturer: Kahenya, N.P

Introduction to lecture 12
This lecture introduces you to basic statistics. It discusses the measures of central tendency and
measures of variation/spread/dispersion for grouped data.
Intended learning outcomes
At the end of this lecture you will be able to;
(i) Define measures of central tendency and measures of variation.
(ii) Calculate measures of central tendency, and variation for grouped data.
References
These lecture notes should be complemented with relevant topics in (Antony & Robert, 2006;
Kahenya, 2017; Upton & Cook, 2001).
Measures of central tendency (Grouped data)

In the previous lecture we dealt with ungrouped data set. In this lecture we shall find the
arithmetic mean, mode, and median by calculation and by use of graphs. One can use a histogram
to find the mode, while the median can be estimated from a cumulative frequencies curve (ogive).

Example 1: Consider the data set below and hence find the; Arithmetic mean, Arithmetic mean
using the assumed mean A, Mode, Median, plot a histogram and estimate the mode, and plot a
cumulative frequency curve and estimate the median.

Class 10 – 14 15 – 19 20 – 24 25 – 29 30 – 34 35 – 39 40 – 44 45 – 49
𝑓 2 5 7 9 14 8 6 1

1
Solution:

∑fx
(i) To calculate the arithmetic mean we use the formula; x̅ = ∑f

In our case x will be the midpoints of the given classes. Hence, we need a table with the additional
rows.

class 10 – 14 15 – 19 20 – 24 25 – 29 30 – 34 35 – 39 40 – 44 45 - 49 Total
𝑓 2 5 7 9 14 8 6 1 52
𝑥 12 17 22 27 32 37 42 47
𝑓𝑥 24 85 154 243 448 296 252 47 1549
∑fx 1549
x̅ = = ≈ 29.79
∑f 52

(ii) Arithmetic mean, using an assumed mean A

When one needs a quick mean or doesn’t have time to calculate one, one can take any class
midpoint more so the one at the center (in case the data is normally distributed), and work with it
to interpret the data set. This is called a working mean or an assumed mean A. To calculate the
arithmetic mean using an assumed mean A we use the formula;

∑fd
x̅ = A + , where d = x − A
∑f

We need some more rows for the values of d. We shall use A = 32 as our assumed mean, since it
lies almost at the middle of the data set.

class 10 – 14 15 – 19 20 – 24 25 – 29 30 – 34 35 – 39 40 – 44 45 - 49 Total
f 2 5 7 9 14 8 6 1 52
x 12 17 22 27 32 37 42 47
d=x−A -20 -15 -10 -5 0 5 10 15
𝑓𝑑 -40 -75 -70 -45 0 40 60 15 -115

∑fd −115
x̅ = A + = 32 + ≈ 29.79
∑f 52

2
(iii) Mode
Mode of grouped data is not as straight forward as for ungrouped data. In grouped data the mode
can be any value in the class interval. Hence, we need a formula to calculate the mode.

f0 − f1
M0 = l + ×i
2f0 − f1 − f2

Where;

l – is the adjusted lower-class limit of the modal class (to get the modal class, find
the modal frequency f0 first).

f0 – the modal frequency (the highest frequency in the data set)

f1 – the frequency BEFORE f0

f2 – the frequency AFTER f0

i – the class interval or class size

In our case, the modal frequency f0 is 14. Hence, the modal class is class 30 – 34 i.e., our mode lies
within this class. To improve on accuracy, we need to adjust the class i.e., 29.5 – 34. 5. Note that
30 is the lower-class limit and 34 is the upper-class limit. After adjusting the class, we have 29.5 as
our adjusted lower-class limit (or lower-class boundary) and 34.5 as adjusted upper-class limit (or
upper-class boundary). The class interval or class size 𝑖 is the difference between the adjusted lower-
class limit and adjusted upper class limit. In our case;

𝑖 = 34.5 − 29.5 = 5

We now have our values as;

l = 29.5, f0 = 14, f1 = 9, f2 = 8, i = 5

f0 − f1 14 − 9 5
M0 = l + × i = 29.5 + × 5 = 29.5 + × 5 ≈ 31.77
2f0 − f1 − f2 2 × 14 − 9 − 8 11

3
(iv) Median
We need to use a formula to find the median Md just like the mode,.

N
− cf
Md = l + 2 ×i
f

Where;

l – adjusted lower-class limit of the median class

f – frequency of the median class

cf – the cumulative frequency BEFORE the median class

i – the class size or class interval

𝑁
– the median position
2

The first step is to find the median position. We need to cumulate the frequency to identify the
median position and hence the median class. Cumulative frequency (cf) column is like ranking the
frequencies to identify the middlemost position.

Class 10 – 14 15 – 19 20 – 24 25 – 29 30 – 34 35 – 39 40 – 44 45 – 49
F 2 5 7 9 14 8 6 1
Cf 2 7 14 23 37 45 51 52

The median position is 26.5th (since N = 52 which is an even number. Note that if the N is a very
large number, we ignore the 0.5).

The 26.5th position is at cumulative frequency 37 (here we have positions 24 up to 37).

Hence our median class is class 30 – 34. Which means the l = 29.5. The cumulative frequency
BEFORE the median class cf = 23

The median frequency f = 14.

N
−cf 26.5−23 3.5
Therefore; Md = l + 2
× i = 29.5 + × 5 = 29.5 + 14 × 5 = 30.75
f 14

4
Remark: The mean = 29. 79, mode = 31.77, and median = 30.75 lies almost in the same
neighborhood. Since the three are measures of central tendency, and the data set is almost normally
distributed.

(v) Mode using a histogram

Definition 1: A histogram is one way to represent grouped data. A histogram is a bar graph for
continuous data set. A histogram can be used to get the mode of the data set.

Procedure of drawing a histogram:

A histogram is drawn by plotting the frequencies against the lower-class boundaries.

The size of the bar is proportional to the frequency i.e., the area of the bars represents the given
frequency. Sometimes you may find grouped data with no uniform class intervals. Hence if for
instance the class interval is doubled, then the frequency is halved.

Estimating the mode using a histogram

To estimate the mode, follow the steps below.

- Identify the highest bar. The mode is in this class.

- Draw a diagonal from the top-left corner of this bar, to touch the top-left corner of the bar
opposite this corner.
- Draw a diagonal from the top right corner of this highest bar, to touch the top right corner
of the bar opposite this corner.
- Drop a perpendicular from the point of intersection of these diagonals, to the horizontal
axis.
- The mode is the point where the perpendicular touches the horizontal axis.

5
Example 1

Class 10 – 14 15 – 19 20 – 24 25 – 29 30 – 34 35 – 39 40 – 44 45 – 49
f 2 5 7 9 14 8 6 1
Lower class 9.5 14.5 19.5 24.5 29.5 34.5 39.5 44.5
boundaries

From the histogram the mode is approximately 31.5 (compare with calculated value of 31.77).

(vi) Median using a cumulative frequencies curve

Definition: The cumulative frequencies curve is also referred to as an Ogive or an S- curve. If the
data is normally distributed the curve makes a smooth letter S, hence the name S-curve.

To plot the cumulative frequencies curve, we plot the cumulative frequencies against the adjusted
upper-class limits.

If the first class has non-zero frequency, then you need to add another class with zero frequency
such that the ogive starts from zero.

Join the points to get a smooth curve. Note that in this example the median position is 26.5th.

6
Next, along the cumulative frequency axis identify the 26.5 point and then draw a line parallel to
the horizontal axis. At the point of intersection with the ogive, drop a perpendicular to get the
median. In our case about 31 (see Ogive below).

Class 5-9 10 – 14 15 – 19 20 – 24 25 – 29 30 – 34 35 – 39 40 – 44 45 – 49
f 0 2 5 7 9 14 8 6 1
cf 0 2 7 14 23 37 45 51 52
Upper class 9.5 14.5 19.5 24.5 29.5 34.5 39.5 44.5 49.5
boundaries

Measures of variation (grouped data)

Example 1: The performance in a math quiz for a group of students was recorded as below. Calculate
the interquartile range, and the standard deviation.

Marks 10 -19 20 - 29 30 – 39 40 – 49 50 – 59 60 – 69 70 – 79 80 – 89 90 - 99

No. of students 1 2 7 10 14 6 5 3 2

7
 Solution:

(i) The interquartile range

Marks 10 -19 20 - 29 30 – 39 40 – 49 50 – 59 60 – 69 70 – 79 80 – 89 90 - 99

No. of students 1 2 7 10 14 6 5 3 2

Cumulative freq. 1 3 10 20 34 40 45 48 50

Recall that IQR = Q3 − Q1 . To find the third quartile we use the formula;

3
N − cf
Q3 = l + 4 ×i
f

This formula is similar with the formula for finding the median (median is the same as Q2 )

3
The third quartile is at position 4 × 50 = 37.5th position (at cumulative frequency 40). Hence, the

3rd quartile class is class 60 – 69, with a frequency f = 6.

Hence the adjusted lower-class limit l = 59.5

The cumulative frequency BEFORE the 3rd quartile class is 34. Class size is 10

3
N−cf 37.5−34
Therefore, Q3 = l + 4 × i = 59.5 + × 10 = 59.5 + 5.833 ≈ 65.33
f 6

Next, to find the first quartile we use the formula;

1
N − cf
Q1 = l + 4 ×i
f
1
The first quartile is at position 4 × 50 = 12.5th position (at cumulative frequency 20). Hence, the

1st quartile class is class 40 – 49, with a frequency f = 10.

Hence the adjusted lower-class limit l = 39.5

The cumulative frequency BEFORE the 1st quartile class is 10. Class size is 10.
1
N−cf 12.5−10
Therefore, Q1 = l + 4
× i = 39.5 + × 10 = 39.5 + 2.5 = 42
f 10

Therefore, IQR = Q3 − Q1 = 65.33 − 42 = 23.33

8
(ii) Standard deviation
To calculate the standard deviation, we use the formula;

∑fx 2
σ=√ − x̅ 2
∑f

We therefore need to find the mean x̅, x 2 , and fx 2 (a table will facilitate this working).

Marks F x Fx x2 fx 2
10 – 19 1 14.5 14.5 210.25 210.25
20 – 29 2 24.5 49 600.25 1200.5
30 – 39 7 34.5 241.5 1190.25 8331.75
40 – 49 10 44.5 445 1980.25 19802.5
50 – 59 14 54.5 763 2970.25 41583.5
60 – 69 6 64.5 387 4160.25 24961.5
70 – 79 5 74.5 372.5 5550.25 27751.25
80 – 89 3 84.5 253.5 7140.25 21420.75
90 – 99 2 94.5 189 8930.25 17860.5
∑f = 50 ∑fx = 2715 ∑fx 2 = 163122.5

∑fx 2 163122.5 2715 2

σ=√ 2
− x̅ = √ −( ) = √3262.45 − 2948.49 = √313.96 ≈ 17.72
∑f 50 50

Exercise

1. The marks obtained by a group of students in a quiz were recorded as below;

Marks 1- 5 6 – 10 11 – 15 16 – 20 21 – 25 26 – 30 31 - 35

No. of students 2 12 20 15 11 6 4

Calculate the interquartile range, quartile deviation, variance, and the standard deviation.

2. Students’ performance in a certain math quiz was recorded as below;

Marks 10 -19 20 - 29 30 – 39 40 – 49 50 – 59 60 – 69 70 – 79 80 – 89 90 - 99

No. of students 10 13 16 19 27 21 15 11 8

Use the above data to find; Arithmetic mean, Arithmetic mean using an assumed mean,
mode, median, mode using a histogram, and median using a cumulative frequencies curve

9
Bibliography
Antony, C., & Robert, D. (2006). Foundation Maths. Prentice Hall.

Kahenya, P. (2017). Foundation Maths. LAP Lambert Academic Publishers.

Upton, G., & Cook, I. (2001). Introduction to Statistics (2nd ed.). Oxford University Press.

10