FOLLOW US:

Classification of Data

A process of arranging raw data in a very systematic manner considering certain characteristic (under consideration)

Objective of classification

(i) It eliminates unnecessary details to facilitate desired statistical analysis.

(ii) Proper statistical analysis can be done only for well classified data.

(iii) Classification of data focus on a certain area as per requirement, which facilitate to apply statistical tools.

---

There are four type of classification :

(i)Time series data :

It is also known as temporal data or chronological data. It is arranged with respect to time.
e.g. The number of student appeared for CS foundation for last nine years.

(ii) Geographical classification :

It is also known as spatical data. It is collected with respect to region or location.
e.g. Student appeared for CS foundation in the year 2016, of various cities in U.P.

(iii) Qualitative or ordinal data :

It is arranged or classified on the basis of quality (attributes).
e.g. Data on gender, qualification, drinking habit and etc.

(iv) Quantitative or cardinal data :

It is arranged or classified on the basis of numeric characteristics (which can be measured in number, countable).
e.g. Data with respect to height, weight, profit, sold item and etc.

The smallest entity in the field inquiry is called data.There are two types of data:
(1) Primary Data
(2) Secondary Data

Primary Data:A data collected by the investigator or organisation for his own use is called primary data. The different methods for collecting primary data are as under:

---

Personal observation: It is the most reliable method,as the investigator collects the data by personal observation without asking any question.

---

Questionnaires method: When the field of inquiry is large then this is the best method, as it saves time,money and labour. It is also called questionnaires method. The main disadvantage of this method is that it has highest rate of non-response from the individuals lying under the area of study.

---

Personal interview method:THis method is very expensive and prone to personal bias,as the investigator directly asks questions to individuals and collects of data. It is effectively used in natural calamities.

---

Data through enumerator: This method is very expensive but is a reliable method, as the investigator appoints the person who collects the data from the individual.

Secondary Data

When the data as being already collected by any Marketing Research Co. or any other agency then this type of data is known as secondary data.

Sources of Secondary Data

(i) Journals, Newspapers and periodicals.

(ii) Government publications.

(iii) International organisations like WHO, Word Bank, IMF and etc.

(iv) NSSO,CSO and other similar organisation.

(v) Private and quasi- government sources like NCERT, UGC, ICAR etc.

(vi) Published or unpublished data from several marketing research institutions, agencies private research and etc.

(vii) Publications or records of various regional or local bodies (Association).

(viii) Research papers or thesis are also the source of secondary data.

Graph of Frequency Distribution

We can put graphical representation of frequancy distribution in any of the following ways.

(1) Histogram or Area Diagram or Column Diagram.
(2) Frequency Polygon.
(3) Frequency Curve.
(4) Cumulative Frequency Curve or Ogive.

---

(1) Histogram : It represents mode graphically.To plot this data should be continuous.In this class intervals are plotted on x axis and their respective frequency on y axis.

Note : When class intervals are not equal first calculate frequency density (FD).

F.D =

frequency of the class x width of the smallest class interval / width of the class interval

or
F.D =

frequency of the class / Adjustment factor

---

AF =

width of the class interval / width of the smallest class interval

---

(2) Frequency Polygon :- Frequency polygon is a graphical representation of data in which the mid-point of each classes is taken on x-axis and corresponding frequencies are placed on y axis , joined all the mid-points by line segment. It can also be drawn by constructing a histogram.

---

(3) Frequancy Curve :- It is similar to frequancy polygon, only difference is that the points are joined by drawing a smooth curve or free hand curve. The total area is taken to be unity. It is a limiting form of a histogram or frequancy polygon.

There are four types of frequancy curves

(a) Bell-Shaped Curve :- Commonly used distributions provides bell-shaped curve. Distributio related height, weight, mark, profit, et belong to this category.
(b) U-Shaped Curve
(c) J-Shaped Curve
(d) Mixed Curve : This may have a combination of above three curved.

---

(4) Cumulative Frequency Curve or Ogive Curve :- There are two type of ogive curves less than type ogive curve and more than type ogive curve . The intersection of both ogive curves gives Median of the given data.

(a) Less than Ogive Curve : In this less than frequancy is used that is plotted on y-axis and upper limit of the class interval plotted on x-axis. It is a increasing curve, having S-shape.

(b) More than Ogive Curve: In this more than frequancy is used that is plotted on y-axis and lower limit of the class interval plotted on x-axis. It is a decreasing curve.

Diagrammatic Representation of Data This is an attractive representation of data that is very easy to understand. There are three type of diagrams.