is chapter illustrates nonlinear methods of data analyses for timeseries of reading times, obtained from a self-paced reading task. We present reading data obtained from adult readers, highlighting di erences in reading uency between literate adults (Wallot & Van Orden, 2011b, c), as well as reading data obtained from 2nd, 4th, and 6th graders, highlighting di erences in reading performance between children of di erent ages. ese two di erent data sets illustrate choices that are encountered when nonlinear analysis techniques of timeseries are employed, and we will discuss solutions to maximize the validity and reliability of the analyses.Study 1 concerns the adult data which present a three-factor design with two levels of each factor (i.e., 2 x 2 x 2): the size of the text unit -word units versus sentence units; the uency of readers -Ph.D. candidates in English literature versus undergraduates from an introduction to psychology course; and text repetition -rst reading versus a subsequent reading. e nonlinear methods illustrated are timeseries analyses, which give information about the hypothetical system dynamics that could have produced the time-ordered variability observed across reading times. e payo from these analyses is new information about performance over-and-above and di erent from the results of conventional analyses.Study 2 concerns the children's reading data containing only a single factor, grade level (2nd, 4th, and 6th). Nonlinear methods here illustrate the di erences between the reading performance of the older and younger children, to describe the acquisition of reading skill.In both cases, conventional analyses reveal little about the systems under study. e nonlinear analyses, on the other hand, reveal di erences due to the text unit conditions and the reading uency of the adult's performance, as well as di erences in reading performance across the groups of children. e results give