Introduction to Statistics

Ungrouped Data

Ungrouped data is data in its original or raw form. The observations are not classified into
groups.

For example, the ages of everyone present in a classroom of kindergarten kids with the teacher
is as follows:

3, 3, 4, 3, 5, 4, 3, 3, 4, 3, 3, 3, 3, 4, 3, 27.

This data shows that there is one adult present in this class and that is the teacher.  Ungrouped
data is easy to work when the data set is small.

Grouped Data

In grouped data, observations are organized in groups.

For example, a class of students got different marks in a school exam. The data is tabulated as
follows:

Mark Interval No. of students

0 − 20 13

21 − 40 9

41 − 60 36

61 − 80 32

81 − 100 10

This shows how many students got the particular mark range. Grouped data is easier to work
with when large amount of data is present.

Frequency

Frequency is the number of times a particular observation occurs in a data.

Class Interval

Data can be grouped into class intervals such that all observations in that range belong to that
class.
Class width = upper class limit - lower class limit

Mean
Finding mean for Grouped Data when class Intervals are not given

For grouped data without class intervals,

Mean, x̄ = where f is the frequency of i observation x .
∑ xi fi
th
i i
∑f
i

Finding mean for Grouped Data when class Intervals are given

For grouped data with class intervals,

∑x f
Mean, x̄ =
i i

∑ fi

Where f is the frequency of i class whose class mark is x .

i
th
i

U pper class limit+Lower class limit
Class mark = 2

Direct method of finding mean

Step 1: Classify the data into intervals and find the corresponding frequency of each class.
class

Step 2: Find the class mark by taking the midpoint of the upper and lower class limits.

Step 3: Tabulate the product of class mark and its corresponding frequency for each class.
Calculate their sum (∑ x f ).  i i

Step 4: Divide the above sum by the sum of frequencies (∑ f ) to get the mean.
i

Assumed mean method of finding mean

Step 1: Classify the data into intervals and find the corresponding frequency of each class.

Step 2: Find the class mark by taking the midpoint of the upper and lower class limits.

Step 3: Take one of the x ’s (usually one in the middle) as assumed mean and denote it by
i
′
a
′
.
Step 4: Find the deviation of ′
a
′
from each of the x s ′
i

d i = xi − a

Step 5: Find the mean of the deviations

∑f d
i i
¯ =
d
∑ fi

Step 6:  Calculate the mean as 

∑ fi di
x̄ = a +
∑f
i

Relation between Mean of deviations and mean

d i = xi − a

Summing over all x s, ′

i

∑ d i = ∑ xi − ∑ a

Dividing throughout by ∑ f i = n, Where ′

n
′
is the total number of observations.
¯ = x̄ − a
d

¯ = a
⇒ x̄ − d

Step-Deviation method of finding mean

Step 1: Classify the data into intervals and find the corresponding frequency of each class.

Step 2: Find the class mark by taking the midpoint of the upper and lower class limits.

Step 3: Take one of the x s (usually one in the middle) as assumed mean and denote it by
′
i
′
a
′
.

Step 4: Find the deviation of a from each of the x s ′

d i = xi − a

Step 5: Divide all deviations −d  by the class width (h) to get u s. i
′
i
x −a
i
ui =
h

Step 6: Find the mean of u s  ′

i

∑ fi ui
ū =
∑ fi

Step 7:  Calculate the mean as 

∑f u
i i
x̄ = a + h × = a + hū
fi
Relation between mean of Step- Deviations (u) and mean
x −a
i
ui =
h
x −a
i
∑f
i h

ū =
∑ fi

1 ∑ fi xi −a ∑ fi
ū = ×
h ∑f
i

1
ū = × (x̄ − a)
h

Important relations between methods of finding mean

All three methods of finding mean yield the same result.

Step deviation method is easier to apply if all the deviations have a common factor.
Assumed mean method and step deviation method are simplified versions of the direct
method.

Median
Finding median of Grouped Data when class Intervals are not given

Step 1: Tabulate the observations and the corresponding frequency in ascending or descending
order.

Step 2: Add the cumulative frequency column to the table by finding the cumulative frequency
up to each observation. 

Step 3: If the number of observations is odd, the median is the observation whose cumulative
frequency is just greater than or equal to  (
n+1
).
2

 If the number of observations is even, the median is the average of observations whose
cumulative frequency is just greater than or equal to ( ) and + 1.
n n

2 2

Cumulative Frequency

Cumulative frequency is obtained by adding all the frequencies up to a certain point.

Finding median for Grouped Data when class Intervals are given
Step 1: find the cumulative frequency for all class intervals.

Step 2: the median class is the class whose cumulative frequency is greater than or nearest to
, where n is the number of observations.
n

n
−cf

Step 3: M edian = l +
2
× h
f

Where,
l =lower limit of median class,
n = number of observations,

cf = cumulative frequency of class preceding the median class,

f = frequency of median class,

h = class size (assuming class size to be equal).

Cumulative Frequency distribution of less than type

Cumulative frequency of the less than type indicates the number of observations which are
less than or equal to a particular observation. 

Cumulative Frequency distribution of more than type

Cumulative frequency of more than type indicates the number of observations which are
greater than or equal to a particular observation.

Visualising formula for median graphically

 Median from Cumulative Frequency Curve

Step 1: Identify the median class. 

Step 2: Mark cumulative frequencies on the y-axis and observations on the x-axis
corresponding to the median class.

Step 3: Draw a straight line graph joining the extremes of class and cumulative frequencies.

Step 4: Identify the point on the graph corresponding to cf

n
= .
2

Step 5: Drop a perpendicular from this point on to the x-axis.

Ogive of less than type

The graph of a cumulative frequency distribution of the less than type is called an ‘ogive
ogive of the
less than type’
type .

Ogive of more than type

The graph of a cumulative frequency distribution of the more than type is called an ‘ogive
ogive of
the more than type’
type .

Relation between the less than and more than type curves

The point of intersection of the ogives of more than and less than types gives the median of the
grouped frequency distribution.

Mode
Finding mode for Grouped Data wen class intervals are not given
In grouped data without class intervals, the observation having the largest frequency is the
mode.

Finding mode for Ungrouped Data

For ungrouped data, the mode can be found out by counting the observations and using tally
marks to construct a frequency table.
The observation having the largest frequency is the mode
mode.

Finding mode for Grouped Data when class intervals are given

For, grouped data, the class having the highest frequency is called the modal class. Mode can
be calculated using the following formula. Formula valid for equal class intervals and when the
modal class is unique.
f1 −f0
M ode = l + ( ) × h
2f −f −f
1 0 2

l =lower limit of modal class

h = class width

f = frequency of the modal class

1

f = frequency of the class preceding the modal class

0

f2 = frequency of the class succeeding the modal class

Visualising formula for mode graphically

Graphical Method for finding mode

Step 1: Express the class intervals and frequencies as a histogram.

Step 2: Join the top corners of the modal class to the diagonally opposite corners of the
adjacent classes
Step 3: Drop a perpendicular from the point of intersection of the above on the horizontal x-
axis.

Measures of Central Tendency for Grouped Data

i) Mean is the average of a set of observations.

ii) Median is the middle value of a set of observations.
iii) Mode is the most common observation.

Best suited measure of central tendency in different cases and the

Empirical relationship between them

i) The mean takes into account all the observations and lies between the extremes. It enables
us to compare distributions.

ii) In problems where individual observations are not important, and we wish to find out a
‘typical’  observation where half the observations are below and half the observations are
above, the median is more appropriate. Median disregards the extreme values.

iii) In situations which require establishing the most frequent value or most popular item, the
mode is the best choice.

Mean, mode and median are connected by the empirical relationship

3 Median = Mode + 2 Mean