# Measures of Dispersion, Mean Deviation of Grouped Data from Median, Step to Find Mean Deviation from Median of a Continuous Frequency Distribution

Get unlimited access to the best preparation resource for UGC Paper-1-Hindi: Get complete video lectures from top expert with unlimited validity: cover entire syllabus, expected topics, in full detail- anytime and anywhere & ask your doubts to top experts.

## Mean Deviation of Grouped Data from Median

Recall that data presented in the following form are called grouped data

Discrete frequency distribution:

 Observation: Frequencies:

Continuous frequency distribution:

 Observation: Frequencies:

Example:

Find the mean deviation about median for the following data:

 Heights (in cm) Number of Girls

Solution:

Let us first find median:

 Height (in cm) Number of Girls (f) Cumulative frequency (c.f)

lies in c.f. .

Median class is corresponding to the c.f. i.e.,

Now, Median

Where lower limit of the median class

sum of frequencies

c.f. of the class just preceding the median class

frequency of the median class

And width or class-size of the median class

Here,

To find mean deviation let us form the following table

 Height (in cm) Number of Girls (f) Mid value of the heights Absolute Deviation

Mean Deviation (Median)

## Step to Find Mean Deviation from Median of a Continuous Frequency Distribution:

Step 1:

Arrange the intervals in ascending order

Step 2:

Write cumulative frequencies

Step 3:

Identify the median class, as the class having c.f. just greater than , where is the total number of observations (i.e. sum of all frequencies)

Step 4:

Find the corresponding values for the median class and put in the formula:

Median

Where lower limit of the median class

Sum of frequencies

c.f. of the class just preceding the median class

Frequency of the median class

And width of the median class

Step 5:

Now form the table for following columns:

 Given intervals Frequencies Mid-value Absolute Deviation from

Step 6:

Now calculate M.D. (M)

Developed by: