~ 10 ~ ISSN Print: 2394-7500 ISSN Online: 2394-5869 Impact Factor: 5.2 IJAR 2017; 3(12): 10-14 www.allresearchjournal.com Received: 03-10-2017 Accepted: 04-11-2017 Eshrat Ara Research Scholar, Department of Psychology, University of Kashmir, Jammu and Kashmir, India Correspondence Eshrat Ara Research Scholar, Department of Psychology, University of Kashmir, Jammu and Kashmir, India Exploring factors: Made easy Eshrat Ara Abstract Factor analysis is used largely when the researcher has substantial numbers of variables seemingly measuring similar things. It has proven particularly useful with questionnaires. It examines the pattern of correlations between the variables and calculates new variables, which account for the correlations. In other words, it reduces data involving a number of variables down to a smaller number of factors which encompass the original variables. This paper presents a very brief overview on factor analysis. Although, the factor analysis is complicated and has many variations. Keywords: Exploring factors, original variables, factor analysis Introduction Factor Analysis is not a new technique – it dates back to shortly after the First World War. It is an invention largely of psychologists, originally to serve a very specific purpose in the field of mental testing. It has proven generally more useful and is used in the development of psychological tests and questionnaires. Personality theories are heavily dependent on factor analysis (e.g., Raymond Cattell). The development of high-speed electronic computers has made the technique relatively routine since no longer does it requires months of hand calculations. In the social sciences we are often trying to measure things that cannot be directly measured, they are so-called latent variables. For example ‘personality’, we can’t measure personality directly: it has many facets. However, we can measure different aspects of personality: we could get some idea of motivation, stress levels, and so on. Having done this, it would be helpful to know whether these differences really do reflect a single variable. Put another way, are these different variables driven by the same underlying variable? So, factor analysis is a technique for identifying groups or clusters of variables. This technique has three main uses: 1. To understand the structure of a set of variables. For example, pioneers of intelligence such as Spearman and Thurstone used factor analysis to try to understand the structure of the latent variable ‘intelligence’. 2. To construct a questionnaire to measure an underlying variable. For example, we might design a questionnaire to measure burnout, personality, etc. 3. To reduce a data set to a more manageable size while retaining as much of the original information as possible. For example, multicollinearity can be a problem in multiple regression, and factor analysis can be used to solve this problem by combining variables that are collinear. The logic and purpose of factor analysis is to process a large set of correlations revealing the dimensionality. In real data there are large numbers of correlations and it is impossible to analyze visually by simplistic eyeballing approach, and further the patterns of correlations are not clear in real data which complicates the process of dimensionality. There are three basic questions of Dimensionality: 1) number of dimensions, 2) correlation among dimensions (if more than one dimension is reflected), 3) psychological meaning of the dimensions. The answers to these questions require back-and-forth process. Concepts in Factor Analysis If we measure several variables, or ask someone several questions, the correlation between each pair of variables (or questions) can be arranged in what’s known as an R-Matrix. An R- matrix is just a correlation matrix: a table of correlation coefficients between variables. International Journal of Applied Research 2017; 3(12): 10-14