Explore topic-wise InterviewSolutions in Current Affairs.

Following are the essential characteristics of a good classification:-

• Clarity : There should be no uncertainty or ambiguity as to which class or group the collected data is to be kept in. 
• Stability : To make the data suitable for comparison and to meaningfully compare the results, it is necessary that the classification has stability. 
• Extensiveness : The various classes should be created in such an extensive manner that no data-item from the collected data is left out and is necessarily included in some or the other class. If required, a miscellaneous class can be created; i.e. while creating classes on basis of marital status, widower, widow, divorcee, etc. cannot be included in married and unmarried classes. Hence, classification should be complete and extensive. 
• Suitability : Classes should be created in accordance with the objectives. As, in order to know the financial status or saving tendency of people, it would be suitable to create classes on basis of income. 
• Flexibility : The classification should be flexible enough to accommodate change, amendment and inclusion in various classes in accordance with new situations. 
• Homogeneity : Units of each class should be homogeneous. All the units (data-items) included in a class or group should be present according to the property on basis of which the classification was done.

1. Ordinary or Binomial Classification 

2. Multiple-attribute Classification

Continuous series are of five types. These are the following:

1. Exclusive Series 

2. Inclusive Series 

3. Open-Ended Series 

4. Cumulative Series 

5. Series of Mid-Value.

Classification of objects is very facilitating, and makes it is easy to find objects. If a student classifies his book according to his requirement, he will not have difficulty in finding out the related book when necessary. Student can classify his book on the basis of subject or by the author or by alphabetical order or by the publication year. If student makes a pile of books without the classification, then he will have to check all books for finding a single one.

Classification is the process of arranging the available facts into homogeneous groups, classes or sub-classes according to resemblances and similarities.

According to Secrist : “Classification is the process of arranging data into sequences and groups according to their common characteristics, or separating them into different but related parts.”

According to Spur and Smith : “The process of presenting data by arranging it on the basis of similar properties, in classes or divisions, is called classification.”

In this way, classification is an activity in which data are divided into homogeneous classes, sub-classes, on basis of a property or quality.

Objectives of Classification:

Following are the objectives of classification:

i. To make statistical data simple and brief : The main objective of classification is to eliminate the complexity of collected dataitems (figures) and given them a simple and concise form so that they can be understood easily. 

For example: the salary figures of 1000 workers in the factory were collected. It is not possible to draw any conclusion from them in the same form, but if they are classified, pre-grouped and tabulated, they can be easily understood.

 iii. Helpful in Comparison : One purpose of classification is to make the facts comparable. For example, if the examination results of the intermediate students (commerce) of 2 schools are to be compared, then this can be done with the help of classification: Inter (Commerce) Exam Results.

v. To present the basis for tabulation : It is impossible to tabulate unorganized and unrefined data-items without classifying them, moreover, without it statistical analysis is impractical. Thus, the process of classification presents a basis for tabulation.

vi. Lending a Scientific basis : Another objective is to lend statistical data, a scientific basis. This makes data more credible.

vii. Utility : Classification increases the utility of statistical data by lending them uniformity.

Following are the three essential elements of an ideal classification:

• Clarity : There should be no uncertainty or ambiguity as to which class or group the compiled data should be kept in. 
• Stability : Stability is necessary to make statistics comparable and to make meaningful comparisons of the results. 
• Extensiveness : The different classes should be composed widely in such a manner that no item of the collected data should be left out.

Indiscrete or continuous series is constructed from continuous variables. Continuous variables have no definite value, instead, they can have any value with a certain limit or class.

Following are the main benefits of classification:

1. Data is simplified and abbreviated through classification. 

2. Classification shows the uniformity of data and increases their usefulness. 

3. Data becomes comparable through classification. 

4. Data becomes attractive and impressive by classification. 

5. The specific difference of data becomes clear by classification. 

6. Classification provides a scientific basis for data.

In this method, the upper limit of a class and the lower limit of the next class are the same. This method is called exclusive because the variable values equal to the upper limit of a class are not included in that class; rather, they are included in the next class.

1. The value of units is given in discrete series, while class-intervals are given in continuous series. 

2. Discreteness exists in discrete series. Unit-value may have a definite difference, while continuity or indiscreteness is found in continuous series.

Meaning of Exclusive Series : The exclusive series is a type of continuous series. In this, the upper limit of the first class is the lower limit of the next class. The value of the upper limit of the first class is not included in that class and is included in the next class, hence it is called an exclusive series.

For example : If the income class is Rs 100-200 and Rs 200-300, then the individual with Rs 200 salary will not be included in the class 100-200, it will be included he the 200- 300 class.

Example of Exclusive Series.

(a) To give a concise form to the large group of data-items

Generally, the inclusive method is used for discrete variables (number of workers, obtained marks, etc.). But the exclusive method is used for continuous variables (income, age, weight, etc.) In such a situation, in terms of easy accessibility, we should change the inclusive series to an exclusive category. For this, the difference between the upper limit of one class and the lower limit of the next class is halved and the result is subtracted from the lower limits (l₁) of the class and added to the upper limits (l₂).

Following are the main objective of classification : 

1. The main objective of classification is to eliminate the complexity of collected data-items (figures) and give them a simple and brief form. 

2. With the help of classification, the similarty-dissmilarity of facts is clarified. 

3. The purpose of classification is to make facts comparable. 

4. The purpose of classification is to make a logical arrangement of the data. 

5. The purpose of classification is to provide scientific basis to the data. 

6. The purpose of classification is to increase the usefulness of data. 

7. The purpose of classification is also to form the basis of tabulation.

Following are the differences between discrete series and continuous series:

A number should be given to each and every table, in order to distinguish and also for easy reference.

1. The size of the table should be according to the size of the paper 

2. The stubs and captions should be arranged logically according to the alphabetical, Chronological, Geographical order and items to be arranged according to the size. 

3. If desired to locate the particular quantity in the body of the table that should be showed by thick colored inks or shaded off.

4. The table should not be overloaded with number of characteristics, rather can be prepare another table. 

5. The table should be complete in all respects by captions and stubs, titles heading and no cell is left blank if left do not put ‘o’ but give (—) dash marks or write N.A 

6. Miscellaneous column can be provided for the data which do not fit to the data, such as Ratio, percentages. 

7. Ditto ( “ ) marks should not be used, as they may confuse with the no. 

8. Footnote may contain about errors, omissions, remarks about the data. 

9. Sources if provided regarding publications, organizations, pages, Journals etc.

Tabulation is a process of systematic arrangement of the classified data in rows and columns, in the form of a table.

50 is lower limit and 60 is upper limit in class interval 50-60.

The difference between the upper limit and lower limit in any class is called class interval.

(a) Average of upper limit and lower limit

These workers are not included in the class (400-500), they will be included in the class (500-600).

Questionnaire is a list of questions used by an enumerator who mails it to the respondents to get the required information.

Individual Series: When the observations are expressed individually, the value of each and every item can be observed and measured separately, it is known as individual series.

Discrete Series : The series in which each unit can be factually measured, is called a discrete series.

In a continuous series, data-items are kept in certain definite classes. When kept in classes the data-items lose their individual identity, and individual items merge in one or the other class group. The classes thus created, have continuity, since where one class ends, the next class begins from there. Due to this class continuity, such a series is called a continuous series.

Continuous series is used the most when number of data-items is more and their magnitude is also more.

Example of Continuous Series :

Meaning of Cumulative Frequency Series : Cumulative frequency series refers to a series in which the frequency classes of different classes are not given separately, but the frequency is written in a cumulative form. In such a category, both the boundaries are not provided, only the upper or lower limit is given. When cumulative frequency is given on the basis of the upper limit, the words ‘less than’ is written before each post value and if it is according to the lower limit, the words ‘more than’ is written while writing the cumulative frequency. While solving the question of cumulative frequency series, it is first converted from cumulative to the simple frequency. 

Example: of Cumulative Frequency Series.

Meaning of Discrete Series : When the values are largely repeated then we can prepare a series showing different values of the variable along with the number of repetitions of each value. Such a series is called a discrete series. No value is written repeatedly in the discrete series. Each value is written only once. If a price or some value comes repeatedly, then the number of times that value repeated is called the frequency of that value.

Statistical data are of two types: 

• Quantitative/Attributive Data : Qualitative data cannot be measured directly. They are measured only on the basis of presence and absence of data. Example: literacy, marital status, employment, etc. are qualitative data. 
• Numeric Data : Numerical data are those that can be measured directly, such as income, age, height, weight, etc.

The purpose of classifying raw data is to organize them sequentially, in which they can easily be made suitable for further statistical analysis. Classification of substances and goods saves precious labour and time, it cannot be done arbitrarily. Thus classification means arranging objects in identical groups or classes.

When facts are classified according to their description or properties (attributes), the process is called Attributive /Qualitative classification. 

Qualitative classification are of two types:

• Binomial Classification : When on the basis of presence and absence of an attribute, the data are classified into two classes- one possessing that attribute, and the other not possessing that attribute, it is called two-fold or Binomial classification. 
• Multifold Classification : The classification, where two or more attributes are considered while classifying facts and is called a multifold classification. For example, data obtained from census is first divided into male and female classes, and then both are further divided into literate and illiterate, and then further into employed and unemployed.

The parts of a table are: Table No., Title of the table, Headnote/Sub-title, Captions, Stubs, Body of the table, Footnote & Source.

The boundaries of the class are called class-limits. The lowest value of the class is called lower limit of the class and the highest value of the class is called upper limit.

The blank table given below represents the number of employees with different occupation in a college

Frequency distribution implies the classification of data-items on the basis of some measurable variable. Frequency distribution is a table in which data-items are included in the form of values or classes. The number of units occurring in each value or class are called their frequencies. Thus, the ordered configuration of values or classes and their’ frequencies is palled frequency distribution.

Normal Frequency Distribution:

Cumulative Frequency Distribution:

Statistical series are divided into three types based on their formation 

1. Individual Series 

2. Discrete Series 

3. Continuous Series

1. Individual Series 

2. Discrete Series 

3. Continuous or indiscrete Series

1. Qualitative/Attributive Data 

2. Numerical Data

When the observation is expressed individually, and the value of each object can be seen separately, it is known as the individual series. These cannot be divided into classes, nor can they be tabulated frequency-wise. The value that occurs repeatedly, is written as many times separately. These are arranged in ascending or descending order. Generally, the serial numbers, roll numbers, year, names of places or persons are given in such series. Its special identification is that data values are given in it, not their frequencies. 

For example – Marks obtained by ten students 17,32,35,33,15,26,41,32,11,8.

Mid point =

\(\frac{Upper Limit + lower Limit}{2}\)

Statistical series is the data or data-attributes which are arranged in logical sequence.

(c) Trinomial Table

(d) On the mid-value of class

Following are the essential elements in an ideal classification :

• Clarity : There should be no uncertainty or ambiguity as to in which class or group the compiled data should be kept. 
• Stability : Stability is necessary to make data comparable and to make out meaningful comparison of the results. 
• Extensiveness : The different sections should be composed widely in such a manner that no item of the collected data should be left out. 
• Suitability : The composition of the class should be according to the purpose. 
• For example: to find out the economic condition of the persons or the saving trend, it would be appropriate to create classes on the basis of income. 
• Flexibility : The classification should be flexible enough to accommodate change, amendment and inclusion in various classes in accordance with new situations. 
• Homogeneity : Units of each class should be homogeneous. All the units (data-items) included in a class or group should be present according to the property on basis of which the classification was done.

Objectives of Classification :

• Simplification and conciseness : The main objective of classfication is to eliminate the complexity of the collected data and to lend them a concise from, so that the classified data can easily be understood. 
• Clarification of Similarity and Dissimilarity : By keeping data-items in homogeneous groups having similar properties, separately, the similarity and dissimilarity between them can be easily understood. 
• Helpful in comparsion : he comparative study of data-items becomes easy through classification. If the populations of cities or village are categorised in class of literate and illiterate, married and unmarried, employed and unemployed, then both cities/villages can easily be compared on the basis of these attributes. 
• Logical arrangement : Classification is a logical process. The data-items included in it are presented in a regular and well-ordered manner.
• For example – classifying population data-items in classes of age, gender, caste-creed, state, etc. is a logical activity. 
• Presentation of the basis of tabulation : Tabulation is impossible without classifying unarranged and raw data, and statistical analysis is also impractical without it. Hence, the process of classification presents a basis for tabulation.

The following four essential elements should be present in an ideal classification : 

• Clarity : There should be no uncertainty or ambiguity as to the class or group in which the collected data items are to be kept. 
• Stability : Stability is necessary to make data comparable and to make meaningful comparisions of the results. 
• Extensiveness : The different sections should be composed widely in such a manner that no item of the collected data should be left out. 
• Suitability : The composition of the class should be according to the purpose. 
• For example: to find out the economic condition of the persons or the saving trend, it would be appropriate to divide the sections on the basis of income.

Classification is the process of arranging the compiled facts into homogeneous groups or classes or sub-classes according to their various qualities and properties.

According to Secrist : “Classification is the process of arranging data into sequences and groups according to their common characteristics, or separating them into different but related parts”.

1. Making it simple and concise 

2. Clarifying similarity and dissimilarity.

The number of times a value occurs in a statistical group, is called the frequency of that value.

Raw data is highly disorganized, due to which it is almost impossible to analyze and to arrive at certain conclusions. Statistical methods cannot be used easily on them. In contrast, classified data become well-ordered, easy to understand and suitable for analysis. Conculsions can be easily drawn from them. Hence, classified data is considered better than raw data.