Abstract
This article introduces the notion of the quartile–quartile line as an alternative to the regression line and the median–median line to produce a linear model based on a set of data. It is based on using the first and third quartiles of a set of (x, y) data. Dynamic spreadsheets are used as exploratory tools to compare the different approaches and to investigate the effects of sample size on the lines that are produced to fit random samples.
Additional information
Notes on contributors
Sheldon P. Gordon
Sheldon P. Gordon is SUNY Distinguished Teaching Professor of Mathematics at Farmingdale State College. He has served on a number of national committees involved in undergraduate mathematics education with special emphasis on efforts to rethink calculus and courses below calculus. He is a co-author of Functions, Data and Models: An Applied Approach to College Algebra, Functioning in the Real World and Contemporary Statistics: A Computer Approach and a co-editor of the MAA Notes volumes, Statistics for the Twenty First Century and A Fresh Start for Collegiate Mathematics: Rethinking the Courses Below Calculus. He was also an original co-author of the texts developed under the Calculus Consortium based at Harvard.