Short Questions and Answers: Organisation of Data - 1

Q.1. Define classification of data. 
Ans. Classification of data is the process of organising raw data into groups or classes based on common characteristics so that it becomes easier to present, compare and analyse the information statistically.

Q.2. List the methods of classification of data. 
Ans. The methods of classification of data are:
(i) Chronological classification (by time, e.g., year-wise data)
(ii) Geographical classification (by place or region, e.g., state-wise data)
(iii) Qualitative classification (by attributes or qualities, e.g., type of industry)
(iv) Quantitative classification (by numerical values or amounts, e.g., income levels)

Q.3. When is data said to be raw? 
Ans. Data is said to be raw when it is newly collected and not arranged or processed in any systematic order; such data require classification, tabulation or summarisation before meaningful analysis can be done.

Q.4. Give an example of geographical classification. 
Ans. An example of geographical classification is data showing sugar production in various states of India, for example, tonnes produced in Maharashtra, Uttar Pradesh, Karnataka, etc.

Q.5. Give one point of difference between qualitative and quantitative classification. 
Ans. In qualitative classification, data are grouped according to non-numerical attributes or qualities (for example, types of occupation); while in quantitative classification, data are grouped according to numerical measures or quantities (for example, income brackets).

Q.6. State one point of difference between chronological and spatial classification. 
Ans.In chronological classification, data are arranged in order of time (for example, production by year); while in spatial classification, data are arranged according to geographical location (for example, production by state or district).

Q.7. What are the objectives of classification? 
Ans. The following are the main objectives of classification:
(i) It makes data comparable by grouping similar observations together.
(ii) It makes data more attractive and easier to understand.
(iii) It summarises data into brief, simple and logical forms.
(iv) It enhances the utility of data by bringing similar items together and highlighting patterns.
(v) It helps to draw out differences and relationships among the data.

Q.8. State the features of a good classification. 
Ans. The following are the features of a good classification:
(i) Classification should be comprehensive so that all collected data can be suitably grouped.
(ii) It should clearly indicate the group to which each item belongs.
(iii) It must be homogeneous, i.e. items in the same group should be similar.
(iv) A good classification must be stable, meaning the grouping remains consistent during the investigation.
(v) A good classification is made in accordance with the objective of the investigation.

Q.9. What is a variable? 
Ans. A variable is a characteristic or quantity that can assume different values for different observations; it may be qualitative (categories) or quantitative (numbers) and is used to study variation across cases.

Q.10. Define discrete variable. 
Ans. A discrete variable is one that can take only certain separate values; its value changes by finite jumps (usually integers) and does not assume intermediate fractional values between two adjacent values.

Q.11. What is a continuous variable? 
Ans. A continuous variable can take any numerical value within a given range, including fractions and decimals; it is measured rather than counted (for example, height or temperature).

Q.12. Give two examples each of a discrete and continuous variable. 
Ans. Examples of a discrete variable: Number of members in a family, result of rolling a die (1-6), goals scored in a hockey match.
Examples of a continuous variable: Height, weight, temperature.

Q.13. State some features of discrete variable. 
Ans. The following are some features of a discrete variable:
(i) A discrete variable can take only particular, separate values.
(ii) Its value changes by finite "jumps" from one possible value to another.
(iii) It does not take any intermediate fractional values between two adjacent values.

Q.14. When is frequency distribution with unequal classes more appropriate? 
Ans. Frequency distribution with unequal class intervals is more appropriate when observations are concentrated around certain values or when the range of data is wide and unequal class widths help to present the distribution more meaningfully; unequal classes also allow class marks to coincide with values around which observations cluster.

Q.15. Why is there no class mark in a discrete frequency distribution? 
Ans. Frequency array is used for discrete variables, where each distinct integral value is treated separately as a class. Since there are no class intervals (only individual values), there is no midpoint or class mark to compute for such a distribution.