Collecting Reliability Data and Creating a Reversal Design Assignment

1.To provide students with experience collecting and calculating reliability data. 

2.Students will also create a reversal design as per ABA guidelines. 

1.This part of the assignment should be done alone.

2.Pick one behavior of a show/movie/TV/politician/leader/news person/speaker (e.g., TED talk). It should be some person who is often seen on screen. Behavior should be appropriate for all. If you doubt the behavior’s appropriate, then stay away from that behavior. It is an exercise for data collection, not an opportunity to watch your favorite show/movie. 

a.Identify a behavior that is frequently seen on the show, so that its easier for you to collect data. Make sure your partner has easy (at no additional cost) access to it.

b.It doesn’t have to be a “problem” behavior. Examples: hair twirling, eye rolling, winking, saying catch phrase, eating, saying “umms,” hugging, checking phone, or fixing clothes. 

c.I would recommend picking video clips/segments without any commercials.

d.You may use a long video clip and want to break it down into five sessions. Or, you may find five different episodes (same actor/character, same behavior) and code first 10 minutes from each.

3.Create an operational definition for the behavior.

a.Make sure to define in a way that another person observing this person would know exactly when the behavior occurs and does not occur. Include examples and non-examples.

4.Next, identify a measurement system that best captures that behavior. 

a.Don’t use frequency system by default. Decide the measurement system based on the dimension of behavior.

b.Each observation session should be minimum 7 minutes.

c.Total video time should be minimum 35 minutes (7 minutes X 5 sessions).

5.Take data on the same show, same character, and same behavior (just different episodes/scenes) and submit it as part A of this assignment. Make sure to note the precise beginning and ending time of the video segment for each observation.

1.This part of the assignment should be done independently. This part of the assignment should be done independently. This part should not be done in collaboration with your RDC. 

2.This assignment continues from part A. You will be creating an ABAB graph using the data collected from part A (which will serve as baseline) and hypothetical data for next 3 phases. You must provide enough hypothetical data to illustrate a functional relation between the behavior and intervention. This often means at least 5 data points per condition (more if you do not have stable data). 

Part A

3.Make sure to refer to the graphing guidelines and all other graphing resources provided related to graphs.

a.For baseline, use the actual data collected for the behavior in part A

b.Create mock data for 5 sessions per each subsequent phase (i.e., treatment, reversal to baseline, and reintroduction of treatment)

c.Finally, create a graph using ABA standards depicting data for baseline and treatment phase. The graph must display a reversal/ABAB design and all four phases must be labeled.

d.Range is listed as the lowest and highest value. E.g., 15-35 instances of throwing.

1.This part of the assignment is completed in collaboration with your reliability partner.

2.You and a classmate will pair to collect reliability data on the behavior you have selected.

3.Make sure you both have different behaviors and have access to the show/movie you have selected.

4.Each student will play the role of ‘primary data collector’ (PDC) for the behavior they have identified and then they will take turn of playing the role of ‘reliability data collector’ (RDC) to collect reliability data for the behavior the partner has collected. 

a.RDC will take reliability data for the behavior that primary student has identified and defined. Use the exact definition as provided by the PDC.

b.RDC will take data on 3 of the 5 sessions that PDC has already observed and collected data for their part A. That is why it is important to note exact beginning and end times of the videos so that data collection is accurate across both observers. 

c.RDC will collect these data independently and submit those data to PDC when reliability data has been collected for all three sessions. 

d.PDC will then take RDC’s data and calculate IOA between both data sets. 

e.So, the partner's job is to watch your picked video, your picked character, your picked behavior, your definition, your measurement system. After they watch 3 sessions, they will simply give you their data sheets. You then have to calculate IOA for each of those 3 sessions. 

f.PDC will next organize their own data and that of RDC and present it in the format shown in sample assignment.

g.Also, list the calculation formula you used to calculate the IOA. 

h.Then you present parts A, B, and C information in one document. 

i.If the IOA score is less than 80%, please report so honestly. There will no point loss for low IOA score, but if it were to be found that there was fibbing of IOA data, that will be grounds for discipline as it violates student honesty and integrity policy of the University. A low IOA score is good from learning perspective. This way you will know how the definition plays a big role in accuracy of data collection.  

Half points of part C will be assigned for the role of PDC and half will be awarded as your role of a RDC.