01 · Explore
The Need for Organisation
Raw data, much like a pile of unsorted junk, is difficult to interpret. Classification is the process of arranging data into groups based on specific criteria to make it comprehensible.
When data is first collected, it is often in a 'raw' or unclassified form. For instance, if you have the marks of 100 students listed randomly, finding the highest score or the average performance is a tedious task. As the volume of data increases—from 100 to 1,000 or 5,000 observations—the difficulty of extracting information grows manifold.
Classification brings order to this chaos. By grouping facts with similar characteristics together, we can easily locate items, make comparisons, and draw inferences. A familiar example is a local junk dealer who sorts items into categories like 'metal', 'plastic', and 'glass' to manage his trade efficiently. Similarly, the Census of India (such as the 2001 Census) collects data on millions of people; in 2001, about 20 crore persons were contacted, and this data only becomes useful when classified by gender, education, or occupation.
Pause & Try
Think it through first. Writing and checking your answer is free with an account.
Question
What is the primary purpose of classifying raw data?
Sign in to see the answer
Write your own answer and compare it with ours. It’s free.
Sign inNew here? Sign up freeNCERT reference: chapter PDF pages 1, 2, 3, 4.
02 · Explore
Methods of Classifying Data
Data can be grouped in various ways depending on the purpose of the study, such as time, location, or specific attributes.
Chronological Classification involves grouping data according to time, such as years, months, or weeks. For example, the population of India recorded every ten years from 1951 to 2011 forms a 'Time Series'. Spatial Classification, on the other hand, organises data based on geographical locations like countries, states, or districts.
When data is grouped based on non-measurable qualities like gender, literacy, or religion, it is called Qualitative Classification. For instance, a population can be divided into 'Male' and 'Female', and further subdivided into 'Married' and 'Unmarried'. Quantitative Classification is used for characteristics that can be measured numerically, such as height, weight, income, or marks obtained in an examination.
| Type of Classification | Basis of Grouping | Example |
|---|---|---|
| Chronological | Time | Population of India from 1951-2011 |
| Spatial | Geographical Location | Wheat yield by country in 2013 |
| Qualitative | Attributes/Qualities | Classification by gender or marital status |
| Quantitative | Numerical Magnitude | Marks of students (0-10, 10-20, etc.) |
Pause & Try
Think it through first. Writing and checking your answer is free with an account.
Question
Distinguish between spatial and chronological classification.
Sign in to see the answer
Write your own answer and compare it with ours. It’s free.
Sign inNew here? Sign up freeNCERT reference: chapter PDF pages 4, 5, 6.
03 · Explore
Understanding Variables
A variable is a characteristic that varies in value. In statistics, variables are broadly classified into continuous and discrete types.
A continuous variable can take any numerical value within a range, including fractions and irrational numbers. For example, the height of a student growing from 90 cm to 150 cm can take every conceivable value in between, such as 102.34 cm. Other examples include weight, time, and distance.
A discrete variable changes only by finite 'jumps' and cannot take intermediate fractional values between two specific points. For example, the 'number of students in a class' must be a whole number; you cannot have 25.5 students. While some discrete variables might involve fractions (like 1/8, 1/4), they are still discrete if they cannot take values between those specific points.
Variable Classification Flow
- 1
Variable
A characteristic that can take different numerical values.
- 2
Continuous
Can take any value (integral or fractional) within a range. Example: Height.
- 3
Discrete
Takes only specific values and changes by jumps. Example: Number of children.
Variables are categorised based on whether they can take any value or only specific, jumping values.
Pause & Try
Think it through first. Writing and checking your answer is free with an account.
Question
Is 'temperature' a discrete or continuous variable?
Sign in to see the answer
Write your own answer and compare it with ours. It’s free.
Sign inNew here? Sign up freeNCERT reference: chapter PDF pages 6, 7.
04 · Explore
The Frequency Distribution
A frequency distribution is a table that shows how different values of a quantitative variable are distributed across various classes.
In a frequency distribution, data is divided into 'Classes'. The number of observations falling into a specific class is called the 'Class Frequency'. For example, if 7 students scored between 30 and 40 marks, the frequency of the class 30-40 is 7.
Each class is defined by Class Limits: the Lower Class Limit (the smallest value) and the Upper Class Limit (the largest value). The Class Interval (or width) is the difference between these two limits. For a class 60-70, the lower limit is 60, the upper limit is 70, and the interval is 10.
Calculating Class Mark
Class Mark = (Upper Class Limit + Lower Class Limit) / 2
The Class Mark (or Mid-Point) is the middle value of a class. For the class 20-30, the Class Mark = (30 + 20) / 2 = 25. This value is used to represent the entire class in further statistical calculations.
Pause & Try
Think it through first. Writing and checking your answer is free with an account.
Question
What is the class mark of the interval 50-60?
Sign in to see the answer
Write your own answer and compare it with ours. It’s free.
Sign inNew here? Sign up freeNCERT reference: chapter PDF pages 7, 8.
05 · Explore
Exclusive and Inclusive Methods
There are two main ways to determine which class a value belongs to when it falls exactly on a class limit.
In the Exclusive Method, an observation equal to a class limit is excluded from one class and included in the next. There are two conventions: 'lower limit excluded' or 'upper limit excluded'. For example, if the upper limit is excluded in classes 0-10 and 10-20, the value 10 would be included in the 10-20 class. This method ensures continuity and is preferred for continuous variables.
In the Inclusive Method, both the upper and lower limits are included in the same class. For example, classes might be 0-9, 10-19, and 20-29. Here, the value 9 is included in the first class. This method creates 'gaps' between classes (e.g., between 9 and 10), which often requires adjustment for statistical analysis.
Pause & Try
Think it through first. Writing and checking your answer is free with an account.
Question
In the exclusive method (upper limit excluded), which class would the value 40 belong to: 30-40 or 40-50?
Sign in to see the answer
Write your own answer and compare it with ours. It’s free.
Sign inNew here? Sign up freeNCERT reference: chapter PDF pages 9, 10, 11.
06 · Explore
Preparing a Frequency Distribution
Tally marking is a manual technique used to count the frequency of observations in each class.
To prepare a frequency distribution, we go through the raw data and place a tally mark (/) against the corresponding class for each observation. To make counting easier, every fifth tally is placed diagonally across the previous four, forming a group of five.
Once all observations are recorded, the tallies are summed to find the frequency for each class. For example, if the class 40-50 has four groups of five tallies and one single tally, its frequency is 21. This process transforms cumbersome raw data into a concise table.
- 1
Determine the range of the data (Largest value - Smallest value).
- 2
Decide the number of classes and the class interval size.
- 3
Set the class limits (Exclusive or Inclusive).
- 4
If using the inclusive method for continuous variables, apply adjustments (subtracting/adding 0.5) to ensure continuity between classes.
- 5
Mark a tally (/) for each observation in its respective class.
- 6
Count the tallies to determine the frequency of each class.
Pause & Try
Think it through first. Writing and checking your answer is free with an account.
Question
How is the fifth tally mark represented in a group?
Sign in to see the answer
Write your own answer and compare it with ours. It’s free.
Sign inNew here? Sign up freeNCERT reference: chapter PDF pages 11, 12, 13.
07 · Explore
Loss of Information and Unequal Classes
While classification makes data manageable, it results in the loss of the exact values of individual observations.
Once data is grouped into a frequency distribution, individual values lose their identity. All observations in a class are assumed to be equal to the Class Mark. For example, in a class 20-30 with frequency 6, we no longer know if the actual values were 21 or 29; we treat all six as 25. This is known as 'Loss of Information'.
To minimize this loss, we sometimes use Unequal Class Intervals. If data is heavily concentrated in one part of the range, we can use smaller intervals for that section and larger intervals for sparsely populated sections. This ensures the class marks are more representative of the actual data points.
Pause & Try
Think it through first. Writing and checking your answer is free with an account.
Question
What is 'loss of information' in the context of data classification?
Sign in to see the answer
Write your own answer and compare it with ours. It’s free.
Sign inNew here? Sign up freeNCERT reference: chapter PDF pages 12, 13, 14.
08 · Explore
Univariate and Bivariate Distributions
Frequency distributions can describe one variable or the relationship between two variables.
A Univariate Frequency Distribution describes a single variable, such as the marks of students or the height of individuals. Most of the distributions discussed so far are univariate.
A Bivariate Frequency Distribution is used when we collect two types of information from each element in a sample. For example, if we record both the 'Sales' and 'Advertisement Expenditure' of 20 firms, we can create a bivariate table. In this table, one variable is classed in columns and the other in rows, with each cell showing the frequency of the corresponding pair of values.
| Feature | Univariate Distribution | Bivariate Distribution |
|---|---|---|
| Number of Variables | One | Two |
| Purpose | Shows distribution of one characteristic | Shows relationship between two characteristics |
| Example | Marks of 100 students | Sales and Ad-spend of 20 firms |
Pause & Try
Think it through first. Writing and checking your answer is free with an account.
Question
What is a bivariate frequency distribution?
Sign in to see the answer
Write your own answer and compare it with ours. It’s free.
Sign inNew here? Sign up freeNCERT reference: chapter PDF pages 15, 16.