Association for Behavior Analysis International

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 to construct and use different types of line graphs that effect their verbal behavior. This symposium examines different aspects of presenting time series data and suggests behavior analysis pay greater attention to for the most critical practices behavior analysts engage in.

One may argue that the genesis and continued practice of all forms of behavior analysis (e.g., applied, basic, philosophical) rest upon the foundation of the careful examination of data graphics, mainly time series graphics. Time series graphics derive their name from their construction; a time series scale on the horizontal axis expresses a unit of time and some type of quantitative scale on the vertical axis shows time progressing from left to right. A quick survey of important books on behavioral research show that prominent experimental designs use time series graphics, mainly the simple line chart (e.g., Cooper, Heron, & Heward, 2007; Hersen & Barlow, 1976; Kazdin, 2011; Kennedy, 2005; Kratochwill, 1978; Kratochwill & Levin, 1992; Poling & Fuqua, 1986; Sidman, 1960). However, the guidelines for constructing line charts appear in very few publications. Furthermore, the field of behavior analysis does not have agreed-upon standards for line chart construction. As a result of number of systematic errors occur in published line charts. This presentation shares the results of survey examining 11 prominent behavioral journals and their graphical practices in regards to line chart usage.

Line graphs or charts have two forms, arithmetically and semilogarithmically scaled. Behavior analysts make scant use of semilogarithmic charts and use, almost exclusively, arithmetically scaled charts. As a scientific discipline Behavior Analysis would benefit from an informed explication of each data graphic. Other sources do exist which describe different types of graphics, including line charts, and explain how behavior analysts can use them (e.g., Cooper, Heron & Heward, 2007; Parsonson & Baer, 1978). Comparing and contrasting the structure, hallmark characteristics and limitations of arithmetically and semilogarithmically scaled line charts, however, has yet to occur. Therefore, the present paper asks the following questions. What design attributes guide the construction of arithmetically and semilogarithmically scaled line chart? What purpose does the arithmetically and semilogarithmically scaled line chart serve? How do the technical features and purposes of line charts serve a science of behavior and the behavior analysts who use them? The research questions will not only provide a detailed comparison of the purposes of arithmetic and semilogarithmic line charts but will also offer advantages or disadvantages for behavior analysts conducting visual analysis with time series data.