One central aspect of becoming an informed consumer of science or scientist-practitioner is developing analytical skills. These analytical skills are problem-solving skills related to understanding data; practitioners must be able to collect and present their data in a way that allows them to make treatment decisions based on current progress related to their overall habilitation goals. In undergraduate and graduate courses, we often present information to learners in graphs. As a result, learners have more experience interpreting graphs than creating graphs. From personal experience, even upper-level undergraduate students have struggled to create bar or line plots with more than one data series for a within-subject or between-subjects design by hand. Learners can benefit from direct instruction on how to create single-case research graphs, and instructors can use the following resources to help their learners graph data in Microsoft Excel, Google Sheets, SigmaPlot, and Prism.
Video Tutorials
There are many resources available to help instructors teach learners how to create graphs in Microsoft ExcelⓇ for both PC and Mac users following Carr and Burkholder’s (1998) original task analysis and recommendation. Vanselow and Bourret (2012) created a series of interactive tutorials to help learners produce graphs for single-case research designs; refer also to Deochand et al. (2015), Fuller and Dubuque (2019), and Watts and Stenhoff (2021) for more recent updates to the task analysis. The tutorial series is available on the Association for Behavior Analysis International website at https://www.abainternational.org/journals/bap/supplemental-materials.aspx under BAP Supplemental Materials and Volume 5, Summer 2021.
Readers can play the regular video (.mp4) tutorial or interactive tutorial for Introduction to Office 2007, Clinical Graphing Parts 1-3, and Research Graphing Parts 1-4. The goal of the Introduction to Office 2007 video is for the learner to understand terminology and basic functions of Microsoft Office 2007, and the intended audience is anyone who is relatively unfamiliar with the terminology and functions of Microsoft Office. The goals of the Clinical Graphing videos are for the learner to review scatter plots and bar graphs, understand when to use each graph, enter and organize data for graphing, insert a graph, add data series and axis labels, add phase change lines, and export graphs to another file. The intended audience for the Clinical Graphing videos is anyone who needs to make a graph for class or to display client data at work. The goals of the Research Graphing videos are for the learner to review scatter plots, enter and organize data for graphing, insert a scatter plot, format the axes, float the origin, add phase change lines and condition labels, and add labels to data paths. The intended audience for the Research Graphing videos is anyone who needs to make a graph for publication. Although there have been a few updates to Excel since these videos were created, the menu options and capabilities will be similar in the current version of Excel.
Additional Software Packages
Fewer resources exist for making graphs for single-case research designs using other software, but several authors created and/or tested tutorials for SigmaPlot, Prism, and Sheets. Cihon et al. (2021) developed a 20-min video tutorial and successfully used it to teach adult participants with experience in ABA how to create a multiple baseline across participants graph in Systat SigmaPlot. SigmaPlot is more expensive than other available graphing options, but it will create professional graphs for publication. As an aside, we used SigmaPlot to create all the figures to display pigeon data in the Urcuioli laboratory; the software is relatively user friendly and will output into various file formats. Mitteer et al. (2018) and Lehardy et al. (2021) also used video modeling to teach behavior technicians how to create publication-quality multiple baseline designs in GraphPad Prism, and their video tutorials are available at https://youtu.be/cuQjSeMrPTw and https://youtu.be/XcWSlH0jAuk, respectively. All four participants in Mitteer et al. (2018) successfully completed at least 86% of the 35 items on the graphing checklist (i.e., appropriately scaled axes, disconnected data paths across experimental phases, descriptive labels for phases and axes, etc.).
Many undergraduate students used Google products in their K-12 classrooms, and they know Google Sheets better than Microsoft Excel. I generally provide tutorials for graphing in Excel and Sheets so that students can choose either program and obtain the proper graph. Blair and Mahoney (2022) compared the functionality of Sheets to Excel, Prism, SigmaPlot, and Numbers and found that all products were comparable across several preferred formatting characteristics for single-subject research design graphs, including: tick marks and labels, data series and markers, phase change lines, and floating zero. They provide a task analysis with detailed instructions and screenshots of the graphing progress and menus for a withdrawal graph, multiple baseline graph, and multielement/alternating treatments graph. I’ve only noticed one difference between Excel and Sheets in their graphing conventions for bar charts: It’s possible to specify different standard error bar values for each data point/bar that are part of the same data series in Excel but impossible to do so in Sheets. Only one value can be specified for standard error (or standard deviation) for all data in the series in Sheets. This is more of an issue when reporting means and standard error, which is not traditionally what we’re displaying in single-subject research design graphs.
Instructional Method
Typical classroom instruction involves creating a slide deck with detailed information on a topic, but this may not be the best approach to teaching learners about graphing. I used to spend 3 days (yes, three) in an undergraduate course on ABA teaching students about graphing conventions prior to having them graph their behavior change project data. Students were bored and inattentive, so I begrudgingly condensed those lectures into a single session designed around creating the graphs live in class. Tutorials (live or pre-recorded) are much better for ensuring that learners use the proper graphing conventions for their single-subject data.
Tyner and Fienup (2015) compared the effectiveness of video modeling, no instruction, and text-based instruction for 66 undergraduate students creating graphs for multiple baseline designs. Participants in the video modeling group played short, narrated videos for each of the steps in the task analysis that consisted of watching the graphing steps completed by the demonstrator. Participants in the text-based instruction group read through 41 slides with sentences and/or screenshots of the software menu and current graph properties. Participants in the no instruction control group received the same slides with information about multiple baseline graphs but without directions for implementing those steps in Excel. The task analysis for creating the graphs included organizing the data table, inserting the chart (in Excel), formatting the data series, formatting the chart area, changing the axis values, etc. For example, refer to Dixon et al. (2009) for a complete task analysis of creating a graph for a reversal design. They measured graphing accuracy with a 50-question checklist based on the task analysis. Participants in the video modeling group completed most of the steps correctly when creating their graph. On average, the participants in the text-based instructional group completed fewer steps correctly than the video modeling group when creating their graph, but they performed better than the no instruction control group. Some participants in the text-based instructional group did reasonably well without a proper tutorial, but there was a lot of variability in performance in that group (ranging from ~10-49 steps completed correctly). The video modeling group also completed their graphs for the multiple baseline design in less time (57 minutes on average) compared to the text-based and no instruction groups.
In a replication of Tyner and Fienup (2015), Zonneveld et al. (2024) compared the effectiveness of instructor-led, video model, and no instruction control tutorials for creating an ABAB reversal design graph with 81 first-year master’s students. The video model and no instruction control groups were similar to the original study, and the instructor-led condition involved the instructor projecting their Excel screen and the task analysis to students to demonstrate each step from the task analysis in order in Excel. Students followed along on their own computers and could ask to repeat a step, stop, speed up, or slow down. While participants had used Excel daily for collecting data, they had not graphed single-subject data. Participants in the instructor-led and video model groups performed similarly when creating their graph for the reversal design according to the 63-item checklist, and both groups performed better than the no control group. All participants created better graphs for the reversal (taught) and multielement (generalized) designs at posttest than at pretest.
Having a synchronous (in person or online) demonstration of how to create graphs in Excel is helpful for learners, but video tutorials are especially helpful in remote instruction given that learners can pause, replay, and seek forward to review the task analysis steps at their own pace. Most tutorials were designed to teach learners how to graph multiple baseline design data (e.g., Cihon et al., 2021; Mitteer et al., 2018; Tyner & Fienup, 2015), but there are also tutorials for teaching students how to graph AB design data (Mondati et al., 2024) and functional analysis data (Chok, 2019).
With these resources and programs such as Zoom, Screencastify, or Camtasia (refer also to Blair et al., 2020, August 31 for a tutorial of using Screencastify to record videos), instructors can create video tutorials or instructor-led demonstrations with the appropriate task analysis for graphing single-subject research design data in Microsoft Excel, Google Sheets, GraphPad Prism, and Systat SigmaPlot. Learners with varying degrees of familiarity with the software or general graphing conventions can follow along with clear directions that focus on the commands and navigation menus that will appear on their screen and produce decent graphs. Video modeling and instructor-led demonstrations don’t produce errorless learning, so learners will need individualized feedback for fixing formatting errors. Instructors may also need to emphasize to learners that taking a screenshot of a completed graph from the video tutorial is not the same as using the software to create an original graph. The skills learned for creating one type of graph also don’t completely generalize to errorless generation of other types of graphs (e.g., from AB designs to reversal designs), so differences between the graphs will need to be highlighted during instruction.
Image credits
[1] Cover image provided courtesy of fauxels under Pexels license
[2] Image provided courtesy of Pavel Danilyuk under Pexels license
[3] Image provided courtesy of RF._.studio under Pexels license
References
Blair, B. J., & Mahoney, P. J. (2022). Creating single-subject research design graphs with Google applications. Behavior Analysis in Practice, 15, 295-311. https://doi.org/10.1007/s40617-021-00604-5
Carr, J. E., & Burkholder, E. O. (1998). Creating single-subject design graphs with Microsoft ExcelTM. Journal of Applied Behavior Analysis, 31, 245-251. https://doi.org/10.1901/jaba.1998.31-245
Chok, J. T. (2019). Creating functional analysis graphs using Microsoft ExcelⓇ 2016 for PCs. Behavior Analysis in Practice, 12, 265-292. https://doi.org/10.1007/s40617-018-0258-4
Cihon, J. H., Ferguson, J. L., Milne, C. M., & Leaf, J. B. (2021). Teaching behavior analysts to create multiple baseline graphs using SigmaPlot. Behavioral Interventions, 36, 910-926. http://dx.doi.org/10.1002/bin.1833
Deochand, N., Costello, M. S., & Fuqua, R. W. (2015). Phase-change lines, scale breaks, and trend lines using Excel 2013. Journal of Applied Behavior Analysis, 48, 478-493. http://dx.doi.org/10.1002/jaba.198
Dixon, M. R., Jackson, J. W., Small, S. L., Horner-King, M. J., Lik, N. M. K., Garcia, Y., & Rosales, R. (2009). Creating single-subject design graphs in Microsoft ExcelTM 2007. Journal of Applied Behavior Analysis, 42, 277-293. http://dx.doi.org/10.1901/jaba.2009.42-277
Fuller, T. C., & Dubuque, E. M. (2019). Integrating phase change lines and labels into graphs in Microsoft ExcelⓇ. Behavior Analysis in Practice, 12, 293-299. https://doi.org/10.1007/s40617-018-0248-6
Lehardy, R. K., Luczynski, K. C., Hood, S. A., McKeown, C. A. (2021). Remote teaching of publication-quality, single-case graphs in Microsoft Excel. Journal of Applied Behavior Analysis, 54, 1265-1280. http://dx.doi.org/10.1002/jaba.805
Mitteer, D. R., Greer, B. D., Fisher, W. W., & Cohrs, V. L. (2018). Teaching behavior technicians to create publication-quality, single-case design graphs in GraphPad Prism 7. Journal of Applied Behavior Analysis, 51, 998-1010. http://dx.doi.org/10.1002/jaba.483
Mondati, A. D., Reeve, S. A., Vladescu, J. C., DeBar, R. M., & Thomas, R. R. (2024). Remote teaching of AB graphs in Microsoft Excel. Behavior Analysis in Practice. https://doi.org/10.1007/s40617-024-00943-z
Tyner, B. C., & Fienup, D. M. (2015). A comparison of video modeling, text-based instruction, and no instruction for creating multiple baseline graphs in Microsoft Excel. Journal of Applied Behavior Analysis, 48, 701-706. http://dx.doi.org/10.1002/jaba.223
Vanselow, N. R., & Bourret, J. C. (2012). Online interactive tutorials for creating graphs with Excel 2007 or 2010. Behavior Analysis in Practice, 5(1), 40-46. https://doi.org/10.1007/BF03391816
Watts, Z. B., & Stenhoff, D. M. (2021). Creating multiple-baseline graphs with phase change lines in Microsoft Excel for Windows and macOS. Behavior Analysis in Practice, 14, 996-1009. https://doi.org/10.1007/s40617-021-00552-0
Zonneveld, K. L. M., Cox, A. D., Asaro, M. M., Hranchuk, K. S., Alami, A., Kelly, L. D., & Frijters, J. C. (2024). Comparing instructor-led, video-model, and no-instruction control tutorials for creating single-subject graphs in Microsoft Excel: A systematic replication and extension. Journal of Applied Behavior Analysis, 57, 502-514. http://dx.doi.org/10.1002/jaba.1053
Blog post contributed by Melissa Swisher