Application of Sequential Probability Ratio Test for Reliability and Validity Studies

Assume a new blood pressure cuff has been created. Before this system can be approved by the medical community, doctors must ensure that it is as reliable as the old blood pressure cuff and that the measurements are reasonably consistent. In practise, reliability and validity studies are used to equate an old (standard) system to a new device. A device is said to be accurate when it has low variance, and it is said to be true when it has zero or minimal bias in reliability and validity studies. In the literature, there are several methods for comparing two instruments. Bland and Altman analysis is a common technique for calculating and interpreting the difference between two devices’ measurements since it is easy to calculate and interpret under the assumption of normal distribution. It’s worth noting that data is often collected in a logical order by researchers. We propose alternate methodological methods in this chapter to save researchers’ time if at all necessary. First, we show how to use a statistical approach called the sequential likelihood ratio test, which allows for a quick conclusion before a defined sample size is reached. Following that, By converting an observed discrepancy between two devices’ measurements into a Bernoulli random variable, we relax the normality assumption (which is not guaranteed to be satisfied in practise), and it offers a slightly different understanding than the original Bland Altman study.

Author (s) Details

Jeffrey O. Wand
Department of Mathematics and Statistics, California State University, Monterey Bay, USA.

Steven B. Kim
Department of Mathematics and Statistics, California State University, Monterey Bay, USA.

View Book :- https://stm.bookpi.org/TPMCS-V9/article/view/852

Leave a Reply

Your email address will not be published. Required fields are marked *