See Dr. Tyner's dissertation defense:
Task Analysis Instruction
Task analyses involves breaking down a complex, multi-step sequence of behaviors into individual manageable steps. This process is ubiquitous to behavior analytic interventions. Behavior analysts develop task analyses to teach multi-step behaviors to our clients (e.g., vocational tasks or grooming tasks like brushing teeth) or to teach staff how to implement intervention procedures.
To create a task analysis, Snell and Brown (2006) suggest one can a) observe others perform the behavior sequence, b) ask experts in the area, or c) perform the task ones self. Beyond this, a few researchers had conducted studies that suggest task analysis instruction should be broken into as many small steps as possible (Crist, Walls, and Haught, 1984) and include pictures and examples (Graff & Karsten, 2012). In our own lab, we have investigated additional variables that influence the effectiveness of task analysis instruction. Our first study investigated the topography of instruction and compared text based presentations of task analyses with video based presentations. More recently, we have examined the necessity of including descriptions of antecedents and consequences of responses in task analysis instruction.
Topography of Instruction: Graphing Video Models
Clinicians often use graphed data to evaluate the effectiveness of an intervention they are delivering to clients. Learning how to make graphs can be difficult. Published task analyses are a common way to learn how to create graphs. These articles list the steps one takes to make a graph. ACE Lab has been evaluating whether we can improve on this paradigm by incorporating video modeling. With advances in technology, services like YouTube have become resources for learning new skills. If you are having trouble changing something on your computer, simply go to YouTube and you can probably find a video of someone walking through how to fix the problem.
We have applied this to creating graphs. Instead of having a learner follow a written list of steps to create graphs, we created video models for each step. A video model is a brief video that demonstrates a target behavior. For graphing, this involves demonstrating how to set up and organize your data table, insert a graph, and format the graph according to APA and journal standards.
Thus far, our research demonstrates promising results. Video modeling appears to be a more effective instructional tool (Tyner & Fienup, 2015). The benefit is that a learner can see exactly where on the screen they should click, including buttons on toolbars and in pop-up menus. Compared to text –based instructions, video modeling produces more accurate graphing that takes less time to make.
At the top of the page is a screen shot of how the video models can be used. We have a PowerPoint presentation on the left side of the screen. The presentation has embedded videos and buttons to move forward and backward through the process. On the right side of the screen, one can open Excel. This way, a learner can make a graph and complete a step while they see it happen in the video.
Describing Antecedent and Consequence Stimuli
Tyner, B. C., & Fienup, D. M. (2016). A comparison of task analysis with and without descriptions of relevant cues and performance criteria for creating reversal design graphs in Microsoft Excel. Journal of Behavioral Education, 25, 379-392. doi: 10.1007/s10864-015-9242-z [link]
Tyner, B. C., & Fienup, D. M. (2015). A comparison of video modeling, text-based, and no instruction for creating multiple baseline graphs in Microsoft Excel. Journal of Applied Behavior Analysis, 48, 701-706. doi: 10.1002/jaba.223 [link]
Tyner, B. C., & Fienup, D. M. (2014). Adapting user research methodology for behavior analytic instructional design. Behavior Analysis and Technology, ABAI Special Interest Group. Retrieved from http://batechsig.com/2014/12/01/adapting-user-research-methodology-for-behavior-analytic-instruction-design/