Speaker 1: Hi students, this video is about the continuation of analysis method, where today we are going to see about the factor analysis. So, this factor analysis, it is one of the statistical data reduction technique, where it tries to explain correlations among the multiple outcomes. So, as a result of this, one or more underlying explanations are factors, it is mean. So, this factor analysis in short is used to reduce a large number of variables into fewer number of factors. So, this technique extracts maximum common variance from all variables and then puts them into a common score. So, this is about the factor analysis. This technique, the technique involves the data reduction, so it attempts to represent a set of variables by a smaller number, so to making it into smaller number to represent the set of variables which is being obtained from the results. So this, the interfered independent variables are called as factors, so we denote it as a factor there. Next we are going to see about the purpose. So, the main application of this factor analytic technique is to first one is to reduce the number of variables. So, in order to reduce the number of variables, this technique can be adopted. Then number two is to deduct the structure in the relationship between variables and that is to classify variables. In order to classify the variables, we use this factor analysis. So, this purpose of factor analysis is to reduce many individual items into a fewer number of dimensions. So, that is the overall major objective of this factor analysis. So, therefore, this factor analysis is applied as a data reduction or structure deduction method, they call it as a data reduction or structure deduction method. So, this is especially popular in the survey research, where the responses to each question represents an outcome. So, for each, the responses of each question represents an outcome. So, because of the multiple questions often are related, so these underlying factors may influence the subject response of that. So, for this reason, the factor analysis is being done. So, the concept, the typical factor analysis suggests answer to the four major questions like this. First one, how many different factors are needed to explain the pattern of relationship among the variables? In order to find out the relationship among the variables, this can be done. So, the first question has to be answered. Then number two is, what is the nature of those factors? What is their characteristics of these factors, which has been chosen? How well do the hypothesized factors explain the observed data? So, observed data, how well it is being explained? So, how well do the hypothesis factors be explain the observed data? How much purely random or unique variance does each observed variable involve? So, for each and every variable, they are going to find out what how much purely random and how much are the unique variance it is being done. So, this is the factor analysis example where these are the variables, it is being grouped item 1, item 2 till item 8 and they are going to find out how it is being related. So, in SPSS, they are going to find out the anxiety, what are the factors which influence this anxiety? Next, we are going to see the types of factor analysis, where the first one is exploratory factor analysis, in short, they say EFA. It is used to identify the complex interrelationships among the items and group items. So, interrelationship among the items and group items that are part of unified concepts. And the second type of factor analysis is confirmatory factor analysis, CFA. So, it is a more complex approach that test the hypothesis that the items are associated with specific factors. So, it is most complex approach which is being widely used. Now, we will see the diagrammatic representation of this EFA and CFA. So, this is all about this EFA, exploratory and confirmatory. So, this is exploratory factor analysis and this is confirmatory factor analysis. So, next we are going to see about the discriminant analysis. So, this discriminant function analysis is used to determine which variables discriminate between two or more naturally occurring groups. So, it is one of the statistical procedure that classifies the unknown individuals and the probability of their classification into certain groups. So, that is the explanation of this discriminant analysis. For example, here I have given an educational researcher want to investigate which variables discriminate between high school graduates who decide first one to go to college, then to attend a trade or a professional school and number 3 is to seek no further training or education. So, in order to find out what is the choice this particular school graduate is going to take. So, for that purpose the researcher could collect data on the numeric numerous variables prior to the students graduation after and after graduation. So, most students will naturally fall either into one of these 3 categories, either they go to college or attend any professional school or trade or else they seek no further education henceforth. So, this discriminant analysis could then be used to determine which variables are the best predictors of students subsequent education choice. So, from that from these 3 which is the most best predictors of the subsequent education choice of the student. Next we are going to see the steps involved in conducting the discriminant analysis. The first one is the problem is formulated before conducting, then number 2 is the discriminant function coefficients are estimated. So, coefficient is nothing but indicates the relative importance of the independent variables in the predicting dependent and then the next step is the determination of significance of these discriminant functions. So, significance is nothing but the importance of these discriminant functions and one must interpret the results obtained. So, they have to has definitely to be interpreted, then the last and the most important step is to assess the validity of that. So, here is the application of this discriminant analysis. So, this is one of the most useful tool for the first one for detecting the variables that allow the researcher to discriminate between different naturally occurring groups. So, in order to discriminate differentiate the natural occurring groups and then for classifying cases into different groups with a better than chance accuracy. So, for classifying the cases into different groups accordingly and with a better chance of accuracy. So, here is a example of that. So, here a biologist could record different characteristic of similar types of or similar types you say or else is groups. So, here it is taken as groups of flowers and then perform a discriminant function analysis to find out or to determine the set of characteristics that allow for the best discrimination between the type. What is the best discrimination of these types of flowers? So, this is an example of discriminant analysis. So, this is the observations of the places. So, here it is taken as the F1 and F2. So, this is how discriminant analysis will look. The next analysis is cluster analysis. So, this is a class of techniques that are used to classify objects or cases into relative groups called clusters. So, it is called classification analysis or numerical taxonomy. So, these are the two another names of this cluster analysis where clustering is a task of grouping a set of objects in such a way that objects in the same group or same group or they may call as the cluster are more similar to each other than those of in the other groups. So, what is the similarity of those which is available objects in the same group which is being compared with the another or each other groups which is being availed. So, this is cluster analysis has been used in marketing for the various purposes. So, best one is the segmentation of consumers. So, in order to segment the consumers this cluster analysis is used on the basis of benefits sought from the purchase of the product. So, what is the benefit they derive out of purchasing the similar I mean a particular product from comparing with the other products. So, it can be used to identify the homogeneous groups of the buyers here. These are the following steps in the cluster analysis. So, the cluster analysis involves formulating a problem first formulating a problem then selecting a distance measure and then selecting a clustering procedure and then deciding the number of clusters available to study and interpreting the profile clusters and finally, assessing the validity of the clustering here. So, now we are going to see the application where and all this cluster analysis being used in the area of medicine. So, what are the diagnostic clusters to so, to answer this question researcher would devise a diagnostic questionnaire that includes most possible symptoms for example, in psychology anxiety and depression. So, the cluster analysis can identify the groups of patients that have the similar symptoms of anxiety and depression. So, same like that marketing where they find out the conduct the survey covering needs attitudes demographics and behavior of the customers. So, in order to identify the homogeneous groups of customers they have this similar needs for having the same needs and attitudes. So, for that in marketing area they use and in education area to measure the psychological attitude and achievement characteristic of the students. So, this is an best example for high achievers in all subjects or students that excel in a certain subjects, but fail in others. So, in order to find out that this clustering is used in educational institutions and at last biology where taxonomy is set in before itself. So, researchers collect the data set of different plants and note different attributes of their phenotypes and the cluster analysis can group their observation into a series of cluster and with that they it is helpful for them to build the taxonomy of groups and subgroups of the same similar plant available. So, this is an example of the cluster analysis where the x axis denotes the income and the y axis this is the y axis denotes the age. So, the dots which is available are known as the objects and it is grouped as clusters. So, this is the cluster it is being grouped as clusters. So, with this our video comes to an end. Thank you for listening.