Novice Interpretations of Progress Monitoring Graphs: Extreme Values and Graphical Aids

Abstract

Curriculum-Based Measurement of Reading (CBM-R) is frequently used to monitor instructional effects and evaluate response to instruction. Educators often view the data graphically on a time-series graph that might include a variety of statistical and visual aids, which are intended to facilitate the interpretation. This study evaluated the effects of brief training, graphical aids, and presence of extreme/outlier values on accuracy. Novice visual analysts ( N = 173) were randomly assigned to a training condition: control, trend line calculation, or extreme value identification. All participants interpreted graphical displays of progress monitoring data with and without trend lines and extreme values. Novice visual analysts’ performance on a knowledge test was uniformly high. Results indicate that the overall probability of an accurate response was .80, which improved to a maximum of .94 within the condition with trend lines and no extreme values. Specifically, results indicate that even in the presence of extreme values, trend lines facilitate novice visual analysts’ accuracy.