A new approach to reliability assessment based on Exploratory factor analysis

Abstract: We need to collect data in any science and reliability is a fundamental problem for measurement in all of science. Reliability means calculation the variance ratio. Reliability was defined as the fraction of an observed score variance that was not error. here are a lot of methods to estimated reliability. All of these indicators of dependability and stability are in contradiction to the long held belief that a problem with test-retest reliability is that it introduces memory effects of learning and practice. As a result, Kuder and Richardson developed a method named KR20 before advances in computational speed made it trivial to find the factor structure of tests, and were based upon test and item variances. These procedures were essentially short cuts for estimating reliability. Exploratory factor analysis is also a Traditional method to calculate the reliability. It focus on only one variable in the liner model, a statistical method that can be used to collect an important type of validity evidence. but in reality, we need to focus on many more variables. So we will introduce a novel method following.