• Our Services

Can’t find a perfect paper?

# Exploratory and Confirmatory Factor Analysis

This is a statistical process used in the description of the variability among the correlated and observed variables by the potentially fewer number of unobservable variables known as factors (Bryant, " Yarnold, 1995). It is the method in which the values of identified data are communicated as tasks of some probable causes to discover the most significant ones.

Scholastic Assessment Tests (SAT): The aptitude assessment instrument analyses the factors that influence students’ education (Kline, 2005). Based on the SAT) scores, this determines the ability of the college students to achieve the requirements. Reasoning test is done verbally and based on mathematical reasoning with subtests such as SAT II being generated. The equation section contains the refreshing items. It takes 3 hours for the SAT I to administer every reasoning section, therefore; allowed seven times every year. The aspects of reasoning approach 500 while the standard deviation is around 110 (Kline, 2005). Every reasoning area is given a percentile score, which represents every participant. The aptitude testing is conducted before and after the involved students are taken through the learning sessions and once they majorly commit themselves to some transformation experience. The nature of the school system and the societal degeneration are among the factors influencing students' performance. The test takers are assigned the scored as they change downward in all the tests. The sub-factors comprise of the selection decision, the reasoning of the student and mental ability to take the assessment. There is the assessment of the skills and knowledge (Kline, 2005). Multivariate statistical processes are followed in testing the academic performance of the participating college students. The joint factor analysis is supported by the multi-dimensional subscales based on SAT.

Exploratory factor analysis (EFA) focuses on the identification of the factors as per the data and increases the amount of variance needed, while confirmatory factor analysis (CFA) assesses a priori hypotheses and is widely controlled by theory. EFA searches for patterns while CFA conducts the statistical hypothesis testing on the projected methods. In EFA, the researcher is not needed to have a certain hypothesis concerning the number of factors that would emerge and the variables and items that would be included in those factors (Tabachnick, " Fidell, 2013). On the other hand, CFA analyses need the investigator to hypothesize, an improvement, the number of factors, if or if not the particular factors are correlated, and the items/measures burden onto and replicate the specific factors. CFA is a reducing method which follows a top-down strategy where people generate the conclusions in relation to theory while EFA is a data-based method or an inductive method which identifies that it follows a bottom-up plan. Therefore people draw conclusions as per the particular observations. If one is not sure of the factors to put in the model, one needs to use EFA. After removing various factors and settling on those things to be put in the model, one should do CFA to scrutinize the method formally to realize whether the selected factors are important (Tabachnick, " Fidell, 2013). In CFA, researchers specify the factor structure based on the appropriate theory and apply CFA to identify is there is empirical assistance for the planned theoretical factor structure and assumes oblique revolution and no cross-loadings. In EFA, people apply the data to identify the underlying structure and use an orthogonal rotation and cross-loadings which are allowed if they are small.

## References

Bryant, F. B. " Yarnold, P. R. (1995) Principal-components analysis, and exploratory and confirmatory factor analysis. In L. G. Grimm, " P. R.Yarnold, (Eds.), Reading and understanding multivariate statistics (pp. 99-136). Washington, DC: American Psychological Association.

Kline, T. J. B. (2005). Psychological testing: a practical approach to design and evaluation. Thousand Oaks, CA: Sage Publications Chapter 10

Tabachnick, B. G., " Fidell, L. S. (2013). Using multivariate statistics (6th ed.). Upper Saddle River, NJ. Chapter 13