What's your question? Figuring out what data you need to answer that question will help you to really focus on the right data and just enough data. Once you have your question, think about these things:
· What information needs to be gathered? Remember you want information that answers the question about the interfering behavior.
· Who will “take” the data? Taking the data means writing down information or tallying information.
· Who will summarize that data? Summarizing the data means looking at the data so that you can describe quickly what is happening. For example, the episodes are getting shorter on average; they used to be 15 minutes long, now they average 7 minutes long. The frequency of hitting is reduced from 10 a day to 2 a day.
But it's not the only thing that teachers do in their day! So data collection (which is the systematic gathering of information) needs to be do-able and use-able. You should collect data to answer a specific question. Two questions you might be collecting information to answer are:
What is the message of the behavior?
Is the behavior intervention plan we have implemented working?
The first question is usually the question being asked during the FBA and the second question is usually the question being asked after the plan is implemented. These two questions, and the information that teams may want to explore, are in a section below.
Data collection tools are just tools that make the collection of information more efficient. To view some data collection tools, click the link.
If your question is related to discovering the message of a target behavior, you'll want to collect data that helps you see patterns. You might want to know:
when the interfering behavior is happening,
who it is happening with, and
what school/work conditions are triggering the behaviors.
Data collection can include teacher notes, interviews, observations, and data by classroom schedule (described on the Data Collection Examples Page).
Once the behavior intervention plan is written, the question changes to whether the target behavior is changing (in the right way). This can be answered with several types of data:
You can look at frequency - how many times an interfering behavior happens.
You can look at how long a behavior lasts – duration. You can judge how intense a behavior episode is - intensity.
Data could also include how long it takes for the student to do something – this is latency.
You might be able to look at permanent products – like the work the student produces.
If an interfering behavior happens frequently, counting it, timing it, or judging intensity might become too difficult for teachers to manage. But there is a solution!
First, consider carefully what information you need to answer the question. You need just enough, not too much and not too little. That may shrink what data needs to be collected.
You can take data for the entire day or just a part of the day. If you use a part of the day, be sure to use the same part of the day each time so you can compare it over time. For example, if you take data during language arts, you can’t compare that effectively with time during math. The situations and settings are too different to allow accurate comparison.
Teams need enough data to be able to be confident that they’ve answered the question. One or two days of data is not enough. Most teams collect information to determine the message of the interfering behavior for about two consecutive weeks.
The data used to determine whether the plan is working will be ongoing. The team needs to look at the trend of the behavior – how is the behavior changing over time, to know that their plan is working (or not).
Remember, collecting the information is not enough. You have to summarize it. In some cases this means averaging over time like we showed on the calendars. You don’t want one rough day or one great day to drive decisions about your behavior intervention plan.
Teams meet regularly to review the behavior intervention plan. Considering the data should be part of each of those reviews. That data should drive when to reduce environmental supports, when to add more supports, and even when to ask a different question and start a new way to collect data.