Statistical Methods and their Applications

 
Statistical Methods and their Applications
Syllabus

Unit-1
Introduction:


Importance and Scope of Statistics. 
In modern times, Statistics is viewed not as a mere device for collecting numerical data but as a means of developing sound techniques for their handling and analysis and drawing valid inferences from them. As such it is not confined to the affairs of the State but is interfering constantly into various diversified spheres of life-social, economic and political. It is now finding wide applications in almost all sciences-social as well as physical-such as biology, psychology, education, economics, business management, etc. It is hardly possible to enumerate even a single department of human activity where statistics does not creep in. It has rather become indispensable in all phases of human endeavour.
1. Statistics and Planning
Statistics is indispensable to planning. In the modem age which is termed as 'the age of planning', almost all over the world governments, particularly of the budding economies, are resorting to planning for the economic development. In order that planning is successful, it must be based soundly on the correct analysis of complex statistical data. In addition to that Modern states being welfare- oriented states, the collection of statistics is done to solve various problems and to frame suitable policies to solve these problems.
2. Statistics and Business
Statistics is an indispensable tool of production control also. Business executives are relying more and more on statistical techniques for studying the needs and the desires of the consumers and for many other purposes. The success of a businessman more or less depends upon the accuracy and precision of his statistical forecasting. Wrong expectations, which may be the result of faulty and inaccurate analysis of various causes affecting a particular phenomenon, might lead to his disaster. Suppose a businessman wants to manufacture readymade garments. Before starting with the production process, he must have an overall idea as to 'how many garments are to be manufactured', 'how much raw material and labor is needed for that', and 'what is the quality, shape, color size, etc., of the garments to be manufactured'. Thus, the formulation of a production plan in advance is a must which cannot be done without having quantitative facts about the details mentioned above.

Following are some of the typical application areas :

A). Quality Management: Statistical quality control (SQC) helps in separating the assignable causes from the chance causes. SQC techniques also help in reducing and stabilizing process variation that may reduce cost significantly.

B). Process Capability:  It enables us to find out whether the enterprise is capable of meeting customer specification. If it is not meeting customer requirements, then management has to take decisions regarding the modernization of plant, change the technology etc. to achieve process capability. Statistics plays crucial role in this field.

C). Finance: Through Financial ratio analysis, key financial ratios are worked out, by using the Balance sheet and income statement data, to know the financial health of an enterprise. Comparative statistical analysis reveals the performance of the company in the past years and the emerging trends in the future.

D). Cash Forecasting:  Financial analyst has to forecast cash requirements to meet short term obligations for example payment of salary, working capital, payments to suppliers etc. statistical analysis helps in working out realistic cash forecasting.

E). Materials Management: Inventory level, by using appropriate statistical methods one can maintain optimum inventory levels of raw materials and finished products and thus savings are made which impact the profit.

F). Quality assessment for incoming and outgoing items: Statistical sampling techniques help in ensuring quality of incoming and outgoing items. Raw materials and other items are checked for quality to meet customer's needs and inspection of finished products is done on sample basis to reduce complaints from customers.

G). Marketing: Marketing research provides solutions to various problems involved in marketing goods, services or ideas. In this context systematic collection, analysis and interpretation of marketing data is done by using statistical methods. Marketing decisions involve the introduction of new product, reaction of consumers to price like, effectiveness of advertisement campaign etc...

H). Demand Projections: Short term and long term demand projections are required for the company products and services. These projections are required both for the domestic and international markets. Statistical surveys and analysis of the past time series data relating to industry provides reliable sales and demand projections.
 
3. Statistics and Mathematics

Statistics and mathematics are very intimately related. Recent advancements in statistical techniques are the outcome of wide applications of advanced mathematics. Main contributors to statistics, namely, Bernouli, Pascal, Laplace, De-Moivre, Gauss, R. 
A. Fisher, to mention only a few. were primarily talented and skilled mathematicians. Statistics may be regarded as that branch of mathematics which provided us with systematic methods of analys- ing a large number of related numerical facts. According to Connor, "Statistics is a branch of Applied Mathematics which specialises in data." Increasing role of mathematics in statistical analysis has resulted in a new branch of Statistics called Mathematical Statistics.

4. Statistics and Industry

In industry, Statistics is very widely used in 'Quality Control', in production engineering, to find whether the product is conforming specifications or not, statistical tools, viz., inspection plans, control charts, etc., are of extreme importance. In inspection plans we have to resort to some kind of Sampling a very important aspect of Statistics.

5. Statistics and Economics

According to F.C.Mills, "Statistics is an Statistics and Economics. Statistical data and technique of statistical analysis have proved im.nensely useful in solving a variety of economic problems, such as wages, prices, analysis of time series and demand analysis. It has also facilitated the development of economic theory. Wide applications of mathematics and Statistics in the study of economics have led to the development of new disciplines called Economic Statistics and Econometrics. In the field of production, we can study the effect of certain incentives on production. Again, it is with the help of statistical techniques that we can know the effectiveness of economic policies on production.
 
6. Statistics and Biology, Astronomy and Medical Science

The association between statistical methods and biological theories was first studied by Francis Galton in his work in 'Regression'. According to Prof. Karl Pearson, the whole 'theory of heredity' rests on statistical basis. He says, "The whole problem of evolution is a problem of vital statistics, a problem of longevity, of fertility, of health, of disease and it is impossible for the Registrar General to discuss the national mortality without an enumeration of the population, a classification of deaths and knowledge of statistical theory."
In astronomy, the theory of Gaussian 'Normal Law of Errors' for the study of the movement of stars and planets is developed by using the 'Principle of Least Squares'.
In medical science also, the statistical tools for the collection, presentation and analysis of observed facts relating to the causes and incidence of diseases and the results obtained from the use of various drugs and medicines, are of great importance. Moreover, the efficacy of a manufactured drug or injection or medicine is tested by using the 'tests of significance - (t-test).

7. Statistics and Psychology and Education
 
In education and psychology, too, Statistics has found wide applications, e.g., to determine the reliability and validity of a test, 'Factor Analysis', etc., so much so that a new subject called 'Psychometry has come into existence.

8. Statistics and War
 
In war, the theory of 'Decision Functions' can be of great assistance to military and technical personnel to plan 'maximum destruction with minimum effort'. Thus, we see that the science of Statistics is associated with almost all the sciences - social as well as physical. Bowley has rightly said, "A knowledge of Statistics is like a knowledge of foreign language or of algebra; it may prove of use at any time under any circumstance."

Reference: Fundamentals of Mathematical Statistics, S.C. Gupta and V.K. Kapoor

Limitations of Statistics

Although Statistics has a very wide application in everyday life as well as in Behavioral Sciences, Physical and Natural Sciences, it has certain limitations also. These limitations are as follows: 

1. Statistics deals with aggregate of facts. 

2. It cannot deal with single observation. Thus statistical methods do not give any recognition to an object or a person or an event in isolation. This is a serious limitation of Statistics. 

3. Since Statistics is a science dealing with numerical data, it is more applicable to those phenomena which can be measured quantitatively. 

4. However, the techniques of statistical analysis can be applied to qualitative phenomenon indirectly by expressing them numerically with the help of quantitative standards. 

5. Statistical conclusions are true only on the average. Thus, statistical inferences may not be considered as exact as inferences based on Mathematical laws. 


Source: B.L.Aggrawal (2009). Basic Statistics. New Age International Publisher, Delhi. 

Classification of data: 

The important object of the statistics is to simplify the complexities. A large complex data is not capable of suggesting anything unless, it is to be divided into groups with respect to their nature. Then, it is done according to kind of character, separate the similar from the dissimilar and bring out the distinguishing features.   Thus, it enables to comparison and drawn the conclusions without unnecessarily considering hundreds of individual numbers. Hence, the process of dividing the objects or things into different classes based on their character or feature is known as classification of data.

Types of Classification

Broadly, data can be classified under following categories:
(i) Geographical classification
(ii) Chronological classification
(iii) Qualitative classification
(iv) Quantitative classification

(i) Geographical Classification

In geographical classification, data are classified on the basis of location, region, etc. For example, if we present the data regarding production of sugarcane or wheat or rice, in view of the four main regions in India, this would be known as geographical classification.
                        Region Wise Production of Wheat in India

(ii) Chronological Classification

Classification of data observed over a period of time is known as chronological classification. For example, Production of wheat in Indian from 2000 to 2006.
                         Year Wise Production of Wheat in India

(iii) Qualitative classification

In qualitative classification, data are classified on the basis of some attributes or qualitative characteristics such as sex, color of hair, literacy, religion, etc. It can be further classified as
(a). Simple Classification
(b). Manifold Classification

(a). Simple Classification
when only one attribute is studied, two classes are formed – one for possessing the attribute and the other for not possessing it. This type of classification is known as simple classification. For example, the population under study may be divided into two categories based
on the characteristic ‘Gender’ as follows:


(b). Manifold Classification

When the data are classified according to more than one attribute, it is called manifold classification. For example, Students are classified as University and Non-University students, among each of the class further divided into two sub classes as Boys and Girls; Boys and Girls are again classified as Indians or Non-Indians. In this example, the attribute observed are University or Non-University, their gender and their nationality. It is shown as the following chart.

(iv) Quantitative Classification
Quantitative classification refers to the classification of data according to some characteristics that can be measured numerically such as height, weight, income, age, sales, etc. This classification again classified based on the consideration of characteristics of each variable (s). They are

(a). Univariate Classification
 
Univariate Classification refers to the classification of data according to one characteristic that can be measured numerically. 
For example, the employees of a company may be classified according to their designation as follows:
                       Designation Wise Employes's in a Company

Univariate classification may also be represented as individual frequency distribution, discrete frequency distribution and continuous frequency distribution. 

(i). Individual Series

The data can be represented in the form of each value of the variable.
For Example, students in a class.

(ii) Discrete Series

The data can be represented in the form of each value of the variable and respective frequency. For Example, students in a class.

(iii) Continuous Series

The data can be represented in the form of each value of the variable in class intervals and respective frequency. For Example, students and their marks in a class.


(b). Bi-variate Classification

The data can be classified based on the two variables that can be measured numerically. For example, the employees of a company may be classified according to their designation as well as gender as follows:
Designation Wise and Gender Wise Employes's in a Company

(c) Multivariate Classification

The data can be classified based on more than two variables that can be measured numerically. For example, the employees of a company may be classified according to their designation, Age as well as gender as follows:

         Age, Designation and Gender Wise Employes's in a Company


Tabulation of Data

A table is a systematic arrangement of statistical data in columns and rows. Rows are horizontal arrangements, whereas columns are vertical ones. A table can solve the purpose of the presentation and facilitate comparison. The simplification results from the clear-cut and systematic arrangement, which enables the reader to quickly locate the desired information. Comparison is facilitated by brining related items of information close together.

Components of a Table

The various components of a table may vary case to case depending upon the given data. But a good table must contain at least the following components:
1. Table Number
2. Table Heading
3. Caption
4. Stub
5. Body of Table
6. Head Note
7. Foot Note

Let us know the concepts
1. Table Number
A statistical table should be numbered. There are different ways with regard to the place where table number is to be given. The table number may be shown either in the center at the top above the title or in the left-hand side of the table at the top. 
2. Table Heading
A good table should have a suitable heading. The heading is a brief description of the contents of the table. It should be placed above the table. 
3. Caption
Caption refers to the column heading and explains what information column presents. It may consist of one or more column headings, i.e. under a column defined and placed at the middle of the column. If the different columns are expressed in different units, the unit should be specified along with the captions.
4. Stub
The stubs are row headings. They are placed at the extreme left of the table and perform the same function for the horizontal rows in the table as the captions do for the vertical columns.
5. Body
The body of the table is the central part of table that contains the numerical information presented in table. This is the most vital part of the table.
6. Head Note
Head note is a brief explanatory statement applying to all or a major part of the material presented in the table and is placed below the title entered and enclosed in brackets.  For example, the unit of measurement is frequently written as the head note such as “in thousands” or “million tons” or “in crores”, etc.
7. Foot Note
Anything in a table which the reader supposed to find difficult to understand should be explained in footnotes. Footnotes may be placed directly below the body of the table. 

Types of Tables

Tables may broadly be classified into following two categories.
1. Simple and Complex Tables
2. General Purpose and Special Purpose Tables

1. Simple and Complex Tables
The simple and complex tables can be differentiated on the basis of number of characteristics presented and studied. If the data based on one characteristic is presented, the table is known as simple table. The simple table is also known as one way table. 

On the other hand, in a complex table, two or more characteristics are presented. The complex tables are frequently used in practice because they facilitate to incorporate full information and a proper consideration of all related facts. If the data are tabulated on the basis of only two characteristics, then the table is known as two-way table. If three characteristics are arranged in a table, then the table is known as three-way table. When four or more characteristics are simultaneously presented it is known as manifold tabulation.

Example for simple table
                            Distribution of Marks in Statistics

Example for two-way table



2. General Purpose and Special Purpose Tables

General purpose tables, also known as reference tables or repository tables, and provide the information for general use or reference. They usually contain detailed information and are not used for specific discussion. 
Whereas Special purpose tables, also known as summary tables or analytical tables, provide information for particular discussion. These tables are also called derivative tables since they are often derived from general tables. 

Source: IGNOU

Diagrammatic Presentation of data

Diagrams are advance technique to present the data. As a layman, one cannot understand the tabulated data easily but with only a single glance at the diagram, one can gets complete picture of data. The data is presented in the form of a surface figure other than a
curve is known as diagrams. They are different types of diagrammatic presentation and are
follows:
(i). Bar Diagrams
(ii). Pie Diagrams
(iii). Pictograms
(iv). Cartograms

(i) Bar Diagrams

It represents the height of the bar varies according to the data. All the bars are placed on a common base, but apart from each other to make them distinct. These diagrams are suitable for spatial series or where the whole data not required to be shown but for series which give parts of the whole. 

(a) Simple Bar diagrams
In a Simple bar diagram, the data related to one variable is depicted. Such as, profits, investments, exports, sales, production etc. This type of diagram may be drawn either vertically or horizontally. Both positive and negative values can be presented.

Example: Prepare simple Bar diagram for the given data



(b). Multiple Bar Diagram
In this type of diagram, two or more than two bars are constructed side by side horizontally for a period or related phenomenon. This type of diagram is also called Compound Bar or Cluster Bar Diagram This diagram, on the one hand, facilitates comparison of the values of different variables in a set and on the other, it facilitates the comparison of the values of the same variable over a period of time or phenomenon. 
For Example - Sector Wise Foreign Investment 


Multiple Bar Diagram


(c). Sub-divided Bar Diagram

In this diagram one bar is constructed for the total value of the different components of the same variable. Further it is sub-divided in proportion to the values of various components of that variable. This diagram shows the total of the variables as well as the total of its various components in a single bar.

For Example - Sector Wise Foreign Investment 

                           
Sub-divided Bar diagram


(d). Percentage Bar Diagram

These diagrams represent the data in comparable manner. It gives an idea of total whole and its components in percentages. The heights of bars are equal because they represent 100 in each case. 

For Example - Sector Wise Foreign Investment 

                         
Percentage Bar Diagram


(iii). Pie Diagram

In this method, the whole has to be shown together with its components. Total angle is 360 and the corresponding component angles made separately. For example, we can show how the budget is allocated under different heads. A pie diagram is a sub-divided circle.  While making comparison, pie diagrams should be used on a percentage basis and not on an absolute basis.

In constructing a pie diagram one can follow the following steps

1. The first step is to convert the various values of components of the variable into percentages
2. The percentages transposed into corresponding degrees. The total percentage of the various components i.e., 100 is taken as 360° (degrees around the center of a circle)
3.The degree of various components is calculated in proportion to the percentage values of different components. 
It is expressed as:
             
For Example, A researcher made an enquiry on the source of information on purchase of iPhone from 550 sample customers. Present the data in the form of pie diagram


Pie Chart


Graphical Presentation of Data

Diagrams are presenting the data in an attractive manner but there is another method which is reliable, more effective and meaningful is graphical presentation.
A frequency distribution can be portrayed by means of Histogram, frequency polygon, Frequency curve and O-give curves.

(i). Histogram

It is one of the most popular and commonly used devices for charting continuous frequency distribution. A set of rectangles (vertical bars) construct with width to the magnitude of a class interval taken on x-axis and proportionate frequencies are taken on y-axis is known as histogram.

(ii). Frequency Polygon

The procedure for the construction of a frequency polygon by histogram, is to first draw the histogram of the given data. Then, put a dot at the mid-point of the top horizontal line of each rectangle bar and join these dots by straight lines resultant one is Frequency polygon.  Or by taking mid-values of class intervals on x-axis and respective frequencies table on y-axis and plot the points & joining the points by freehand.

(iii). Frequency Curve

The procedure for the construction of a frequency curve by histogram, is to first draw the histogram of the given data, then, put a dot at the mid-point of the top horizontal line of each rectangle bar and join these dots by freehand resultant one is Frequency curve.  Or by taking mid-values of class intervals on x-axis and respective frequencies table on y-axis and plot the points and joining with freehand.

(iv). O-give / Cumulative frequency curves

It is a graphic presentation of the cumulative frequency distribution of continuous variable. It consists in plotting the c.f. (along Y-axis) against the class boundaries (along X-axis). Since they are two cumulative frequency curves viz., less than c.f. and more than c.f.

(a). Less than O give
(b). More than Ogive

(a). Less than O give: This consists in plotting the less than cumulative frequencies against the upper-class boundaries of the respective classes. The points so obtained are joined by smooth free hand curve to give less than ogive.

(b). More than Ogive: In more than Ogive, the more than cumulative frequencies are plotted against the lower-class boundaries of the respective classes. The points so obtained are joined by smooth free hand curve to give more than Ogive.

Example:  The following data relates to Maintenance Expenses on 200 Computer systems. Determine Quartiles, Deciles and Percentiles from the O-give curve.


Source: Sinha, S.C. and Dhiman, A.K. 2002. Research Methodology, Vol. 1. Ess Publication, New Delhi.



UNIT-2 
(Measures of Central Tendency)















Source: Fundamentals of Mathematical Statistics, Das Gupta & Goon Gupta

UNIT-3
(Measures of Dispersion)












Combined Standard Deviation

Suppose there are two groups with sizes N1 and N2 and also have SDs, means are respectively then combined standard deviation calculated as follows.


σcombined=(N1−1)σ12+(N2−1)σ22+N1N2(Xˉ1−Xˉ2)2N1+N2N1+N2−1

Where:

  • N1 and N2 are the sample sizes of the two groups.

  • σ1 and σ2 are the standard deviations of the two groups.

  • Xˉ1 and Xˉ2 are the means of the two groups.



Source: Fundamentals of Mathematical Statistics, SC Gupta and VK Kapoor, S Chand Pub


UNIT-IV
(Skewness and Kurtosis)

         Skewness:

The word skewness refers to lack of symmetry. Asymmetrical distribution is called as skewness distribution. Skewness describes how much statistical data distribution is asymmetrical or deviated from the normal distribution, where distribution is equally divided on each side. If a distribution is not symmetrical or normal, it is known as skewed, i.e., the frequency distribution skewed to the left or right. Skewness is a measure of the asymmetry of a distribution.

Symmetric Distribution 

                                    Mean=Median=Mode


 

          Types of Skewness

A)   Positive Skewness (Right Skewness Distribution):

If the given distribution is shifted to the left and its tail on the right side, it is called positively skewed distribution. It is also called the right-skewed distribution. A positively skewed distribution assumes a skewness value of more than zero. Since the skewness of the given distribution is on the right, the mean value is greater than the median and it moves towards the right, and the mode value occurs at the highest point on the left side.

                           

                                                              Positive Skewed Distribution 

             Mean > Median > Mode

 B)   Negative Skewness (Left Skewness Distribution):

If the given distribution is shifted to the right and its tail on the left side, it is called negatively skewed distribution. It is also called the left-skewed distribution. A negatively skewed distribution assumes a skewness value of less than zero. Since the skewness of the given distribution is on the left, the mean value is less than the median and it moves towards the left, and the mode value occurs at the highest point on the right side.

                      

                                                    Negative Skewed Distribution 

       Mean < Median < Mode

         Moments: 

         


           Central Moments (Moments about mean)

            The rth moment about mean is defined as 

                                            


               Non-Central Moments (Moments about zero)        

                The rth moment about zero is defined as 


            Non-Central Moments (Moments about arbitrary constant)        

                The rth moment about an arbitrary constant A is defined as 


 Absolute Measures of Skewness:

            The following are known as absolute measures of skewness:

            1. Skewness (Sk) = Mean - Median

            2. Skewness (Sk) = Mean - Mode

            3. Skewness (Sk) = (Q3 - Q1) - (Q2 - Q1)

            Relative Measures of Skewness:

            In order compare the two or more series, one can use relative              measures

            A. Karl Pearson's Coefficient of Skewness:

            It is proposed by Karl Pearson and is most frequently used                for measuring skewness. The formula for measuring skewness:

                 
          

          Note: 1. It lies between +1 and -1

                    2. If mean = mode, the distribution is symmetric.

                    3. If mean > mode, the distribution is positive skewed.

                    4. If mean < mode, the distribution is negative skewed.

          

            Note: 2.  If mode id ill-defined then one can use inter                          relationship between mean, median and mode (i.e., Mode= 3               Median - 2 Mean) for the formula for coefficient of skewness is 

            In the above formula the limits of skewness are +3 and -3


            B. Bowley's Coefficient of Skewness:
            
         This method is based on quartiles and the formula for                          calculating coefficient of      skewness is 
                             
            Note: 1. It lies between +1 and -1

                    2. If Skb =0, the distribution is symmetric.

                    3. If Skb  > 0, the distribution is positive skewed.

                    4. If Skb  <0, the distribution is negative skewed.


          C. Kelly's Coefficient of Skewness:
      
        The coefficient of skewness proposed by Kelly based on                      percentiles and deciles. The formula for calculating Skewness            is (Based on Percentiles)

                     

         The formula for calculating Skewness is (Based on Deciles)


        Beta Coefficients of Skewness:

        
        Note: 

                    1. If beta1 =0, the distribution is symmetric.

                    3. If beta1 > 0, the distribution is positive skewed.

                    4. If beta1< 0, the distribution is negative skewed.


       Gamma Coefficients of Skewness:

 



        Kurtosis:

       Another measure for getting complete idea about the shape of             distribution is known as Kurtosis. It is also proposed by Karl               Pearson and he called it is "Convexity of a Curve".

    The curve with greater peakedness than normal curve is called           Lepto kurtic. The curve with more flat than normal curve is known    as  Palty Kurtic. The Normal Curve is known as Mesokurtic. The        following figure mentions different curves. 

                            


        
        Measures of Kurtosis:

        (A). Karl Pearson's Measure of Kurtosis:

        Karl Pearson has used to second and fourth central moments for          calculating kurtosis. The formula is 
                             


           Note: 

         (B). Kelly's Measure of Kurtosis:


         Kelly has given a measure of kurtosis based on percentiles.                 The formula is given by 

                

                Note: 


   Example1:

  For a distribution Karl Pearson’s coefficient of skewness is                  0.64, standard deviation is 13 and mean is 59.2 Find mode                  and median.

  Example 2:

  Karl Pearson’s coefficient of skewness is 1.28, its mean is 164 and      mode 100, find the standard deviation.

 Example 3:

 For a frequency distribution the Bowley’s coefficient of skewness      is 1.2. If the sum of the 1st and 3rd quarterlies    is 200 and median is 76, find the value of third quartile.

Example 4:

For the following data relates to marks of 150 students. Find the Karl Pearson and Bowley's coefficient of skewness. 

  

Marks

0-10

10-20

20-30

30-40

40-50

50-60

60-70

70-80

No. of Students

10

40

20

0

10

40

26

14


Example 5: 
First four moments about mean of a distribution are 0, 2.5, 0.7 and 18.75. Find coefficient of skewness and kurtosis.

 Example 6: 
Compute the Karl Pearson's and Bowley's coefficient of Skewness from the following data.

Height (In inches)

No of Persons

58

10

59

18

60

30

61

42

62

35

63

28

64

16

65

8


Example 7:
               
 Compute the first four moments about mean from the following data then find skewness and kurtosis.

              
Source: IGNOU Material            

UNIT-V

(Correlation and Regression)

Correlation Analysis: 

Meaning: 

Correlation is a statistical technique to ascertain the            association or relationship between two or more variables.

Definitions:

“When the relationship is of a quantitative nature, the appropriate statistical tool for discovering and measuring the relationship and expressing it in self-formula is known as correlation.”

                                                     ----------Croxton &amp; Cowden

2. “When two or more quantities vary in sympathy so that movements in the one tends to be accompanied by corresponding movements in the others, then they are said to be correlated.”

                                                                             ---- L.R.Connor

3. “Correlation analysis attempts to determine the degree of relationship between variables.” 

                                                                      --------- Ya Lun Chow

4. "Correlation is an analysis of the covariation between two or more variables"

                                                                                    —AM Tuttle

Types of Correlation:

Correlation classified in several different ways. The most important correlation are as follows:

1. Positive or Negative or Zero.

2. Simple, Partial and Multiple.

3. Linear and Non-linear.

Positive and Negative Correlation:

    If both the variables vary in the same direction, correlation is said to be positive. It means if one variable is increasing, the other on an average is also increasing or if one variable is decreasing, the other on an average is also deceasing, then the correlation is said to be positive correlation. Example: 

If both the variables vary in the opposite direction, correlation is said to be negative. It means if one variable is increasing, the other on an average is decreasing or if one variable is decreasing, the other on an average is also increasing, then the correlation is said to be negative correlation. Example: 

When one can’t find any relationship between the variables then, it is said to be zero correlation. It means a change in value of one variable doesn’t influence or change the value of another variable. Example: 

Simple, Partial and Multiple Correlation

When one can studied only two variables, it is a case of simple

correlation. Example: 

If one can studied three or more variables but considers only two variables to be influencing each other and the effect of other influencing variables being held constant is known as Partial correlation. Example: 

When three or more variables are studied, it is a case of multiple correlation. Example: 

Linear and Non-linear Correlation

If the amount of change in one variable bears a constant ratio to the amount of change in the other variable, then correlation is said to be linear. 

If the amount of change in one variable does not bear a constant ratio to the amount of change to the other variable, then correlation is said to be non-linear.

Methods of Measuring Correlation

Various methods are used to measure correlation. The commonly used methods are 

a) Scatter diagram method

b) Karl Pearson's method

c) Spearmen's method

d) Concurrent deviation method.

a) Scatter Diagram Method

This method is also known as Dotogram or Dot diagram. Scatter diagram is one of the simplest methods of diagrammatic representation of a bivariate distribution. Under this method, both the variables are plotted on the graph paper by putting dots. The diagram so obtained is called "Scatter Diagram". 

Merits:

1. This method helps us to form a rough idea about the nature of the relationship between the two variables only inspection of the grap.

2. The method is not affected by extreme observations whereas mathematical formulae of determining correlation between two variables are affected by extreme items.

3. This method helps us to get an approximate estimating line or the line of best fit by free hand method. This method consists of the line drawn through the plotted points to locate the best possible line.

Demerits:

1. This method is not suitable if the number of observations are sufficiently large.

2. This method does not provide an exact measure of the extent of the relationship between the two variables.

Product Moment Method or Karl Pearson's Correlation coefficient:

Kart Pearson, a reputed statistician, in 1890, has constructed a well set formula based on mathematical treatment for determining the coefficient of correlation. The formula is named after his name as Karl Pearson's formula and is popularly known as 'Karl Pearson's coefficient of correlation. I is also named as Product moment coefficient.


Spearmen's Rank Correlation Coefficient:

When quantification of variables becomes difficult such beauty of female, leadership ability, knowledge of person etc, then this method of rank correlation is useful which was developed by British psychologist Charles Edward Spearman in 1904. In this method ranks are allotted to each element either in ascending or descending order. The correlation coefficient between these allotted two series of ranks is popularly called as “Spearman’s Rank Correlation". 


Merits:

·  1. It is easy to calculate and understand as compared to Pearson's r. 

· 2. When the ranks of different values of the variables are given, it is then the only method left to calculate the degree of correlation.

· 3. When actual values are given and we are interested in using this formula then, we have to give ranks to calculate correlation.

· 4. This method is employed usefully when the data is given in a qualitative nature like beauty, honesty, intelligence etc.

Demerits:

· 1. This method cannot be employed in a grouped frequency distribution.

·  2. If the items exceed 30, it is then difficult to find out ranks and their differences.

·  3. This method lacks precision as compared to Pearson's co- efficient of correlation, as all the information concerning the variables is not used. It is just possible that the difference between r, and r may be very insignificant.

Concurrent Deviation Method- 

Regression Analysis:

Definition: The average relationship between two variables is known as Regression.  The literal meaning of 'regression' is stepping towards back. 

In general, suppose that there is a single dependent variable or response variable y and that is related to k independent or regressor variables say x1,……..,xk. The response variable y is a random variable and the regressor variables x1,……,xk are measured with negligible error. The relationship between y and x1,……..xk is characterized by a mathematical model and it is known as the regression model. It is also known as the regression of y on x1,……..,xk. This regression model is fitted to a set of data. In many situations the experimenter knows the exact form of the functional relationship between y and x1, ..., xk.

If k=1, the model is known as Simple linear regression
If k =2 or more than 2 are called multiple regression.

Regression Line of Y on X:
                                        Y = a + b X
                                        Where, Y - Dependent variable
                                                    X - Independent variable
                                                    a - Constant or intercept
                                                    b - Regression Coefficient


Regression Line of X on Y:
                                        X = a + b Y
                                        Where, X - Dependent variable
                                                    Y - Independent variable
                                                    a - Constant or intercept
                                                    b - Regression Coefficient

SMA Multiple Choice Questions

                                            (Open the link)






Comments

  1. The experiences we create define the impact we leave on learners and educators alike. Whether it’s through clear guidance, structured support, or meaningful feedback, every interaction contributes to academic growth. High-quality solutions such as academic writing services help students navigate challenges with confidence, ensuring their learning journey feels supported rather than overwhelming. Ultimately, the goal is to create experiences that empower students, build skills, and inspire long-term academic success.

    ReplyDelete

Post a Comment