|
| The Unfortunate State of Affairs of Graphing Practices in Behavior Analysis |
| Saturday, May 23, 2015 |
| 1:00 PM–1:50 PM |
| 006D (CC) |
| Area: PRA/EAB; Domain: Applied Research |
| Chair: Douglas E. Kostewicz (University of Pittsburgh) |
| Discussant: William J. Helsel (Behavior Technology Consultant) |
| CE Instructor: Richard M. Kubina Jr., Ph.D. |
| Abstract: Behavior analysts take great care in observing and recording data. Line graphs form the main device used to interpret the rigorous collected data. However, the care with which the field of behavior analysis uses when publishing data varies greatly. Behavior analysis relies on nonstandard linear (or equal interval) graphs to determine the effects of almost all times series data. Due to the nonstandard nature and the limitations of linear graphs behavior analysis must continue to face inveterate limitations of non-quantification and shifting slopes due to arbitrary decision of graph creators. The science of behavior analysis would improve at least 10x with standardization, quantification, and vastly improved information for time series behavior. The present symposium provides data from a study showing the state affairs for line graphs in behavioral journals. The symposium also presents and alternative to nonstandard linear graphs in the form of a superior visual display, the Standard Celeration Chart. |
| |
| A Critical Review of Line Graphs in Behavior Analytic Journals |
| DOUGLAS E. KOSTEWICZ (University of Pittsburgh) |
| Abstract: Graphs have played an instrumental role in the science of behavior analysis since its inception. The most common graphic used in behavior analysis falls under the category of time series. The line graph represents the most frequently used display for visual analysis and subsequent interpretation and communication of experimental findings. However, no critical examination of graphical quality has occurred to date. The purpose of the current survey serves to examine the essential structure and quality features of line graphs in behavioral journals. Four thousand three hundred and thirteen graphs from 11 journals served as the sample. Results of the survey indicated a high degree of deviation from standards of graph construction and proper labeling. A discussion of the problems associated with graphing errors precedes future directions for research in the field of behavior analysis. |
| |
| A direct comparison of nonstandard linear graphs and Standard Celeration Charts |
| RICHARD M. KUBINA JR. (Penn State), Douglas E. Kostewicz (University of Pittsburgh) |
| Abstract: Behavior analysts analyze and interpret data as well as construct, communicate, and defend claims of knowledge with data graphics. While behavior analysts do use other inscription devices, such as tables, visual analysis of data serves as the linchpin of the science of behavior. From Skinner to contemporary behavior analysts, researchers and practitioners mostly rely on time series data graphics such as the line graph. With so much analytical and communicative power invested in line graphs behavior analysts must consider how different types of line graphs effect their verbal behavior. This paper describes, compares, and contrasts two types of line graphs, or charts, available for time series data: nonstandard linear charts and Standard Celeration Charts. Standard Celeration Charts provide benefits that all behavior analysts should demand: quantification of trend, variability, immediacy of behavior change, and trend changes. Beyond quantification, Standard Celeration Charts also present numerous advantages over nonstandard linear graphs for the graphic display of behavioral data. |
|
| |